ELEMENTS OF CHEMISTRY: Jf0r tin disc of Colleges, JUademtes, and Schools. BY M. Y. REGNAULT. ILLUSTRATED BY NEARLY 700 WOOD-CUTS. TRANSLATED FROM THE FRENCH * By THOMAS R. BETTON, M.D. AND EDITED WITH NOTES, By JAM E S C. BOOTH, AND WILLIAM L. FABER, , METALLURGIST AND MINING ENGINEER. f/ IN TWO VOLUMES.—V0L. I. PHILADELPHIA: PARRISH, DUNNING, AND HEARS. 1852. Entered according to Act of Congress, in the year 1852, by AMBROSE W. THOMPSON, in the Clerk’s Office of the District Court for the Eastern District of Pennsylvania. STEREOTYPED BY L. JOHNSON & CO. PHILADELPHIA. TABLE OF CONTENTS OF YOL. I. PAGE Introduction 9 Difference between physical and che- mical phenomena 9 Definition of chemistry 10 Simple and compound bodies 10 Divisibility of matter 11 States of bodies 11 Force of cohesion or aggregation 12 Chemical affinity 12 Law of multiple proportions 12 External characters used to distin- guish bodies 13 Crystallography 14 Cleavage 15 Fundamental forms 15 Natural joints of a crystal 16 Angles of a crystal 16 Simple and compound forms 16 Dominant and secondary forms 17 Truncation ,.. 17 Bevelment.. 18 Acuinination 18 Centre of the crystal 18 • Axes of the crystal ,.... 19 Position of the crystal 19 Definitions of systems of crystalliza- tion 20 The six systems of crystalline forms... 20 1. Regular system, monometric 21 2. Quadratic system, dimetric 25 3. Hexagonal system 30 4. Rhombic system, trimetric 36 5. Monclinic system 40 6. Triclinic system 42 Molecular decrements 44 Goniometer 51 Imperfections of crystals 55 Hemitropes or twin-crystals 59 Dimorphism and polymorphism 60 Isomorphism 61 Relations of isomorphism to compo- sition 62 Chemical Nomenclature 63 Table of elements and their symbols... 64 Chemical notation and formulas 73 PAGE Division of elements into metalloids and metals 76 METALLOIDS 79 Oxygen, its preparation from red oxide of mercury 79 Preparation from binoxide of manga- nese 81 Gasometer 82 Preparation from binoxide of manga- nese and sulphuric acid 83 Preparation from chlorate of potassa. 84 Physical and chemical properties 85 Blowpipe with atmospheric air 87 “ “ oxygen 88 Hydrogen, preparation from water 89 “ by ignited iron 90 “ by certain metals and acids 90 Chemical and physical properties 91 Hydroxygen blowpipe 92 Mercury-bath to collect gases 94 Compounds of Hydrogen and Oxygen... 96 Protoxide of hydrogen or water 96 Properties of water 96 Crystallized water 97 Deliquescence and efflorescence 98 Distillation 98 Evaporation 100 Solubility of gases in water 101 Synthetic composition of water 103 Graduation of measuring-glasses 103 Eudiometer 104 Relative volumes of hydrogen and oxygen in water 105 Analysis of water by galvanism 110 Equivalents of hydrogen, oxygen, and water Ill Binoxide of hydrogen 112 Presence or catalysis 115 Nitrogen 117 Preparation from the air 117 “ chlorine and am- monia 118 Chemical and physical properties 119 ! Atmospheric Air 119 1 Analysis of air 120 4 TABLE OF CONTENTS. PAGE Determination of water and carbonic acid; aspirator 121 Determination of the proportions of nitrogen and oxygen 125 Analysis of air by the eudiome- ter! 129 The air is only a mixture 131 Compounds of Nitroqen and Oxyqen.... 132 Nitric acid 133 Chemical properties of nitric acid.... 133 Formation “ .... 135 Preparation “ .... 135 Manufacture “ .... 137 Analysis “ .... 138 Protoxide of nitrogen or nitrous oxide 142 Its properties 142 Its analysis by potassium 144 “ the eudiometer 145 Deutoxide of nitrogen or nitric oxide 145 Its properties 145 Woolf’s bottles, theory of their ac- tion 147 , Safety-tubes 148 Analysis of nitric oxide 150 Nitrous acid; its preparation 153 Its properties 153 Its analysis 154 Hyponitric acid ; its preparation 155 Its properties 155 Its analysis 156 Laws of the combination of gases 158 Equivalents of nitrogen and its com- pounds with oxygen 159 Compound of Nitrogen and Hydro- gen 162 Preparation of ammonia 163 . Properties “ 164 Analysis “ 165 Anhydrous nitric acid 168 Sulphur 169 Its diffusion and properties 169 Its purification 172 Compounds of Sulphur and Oxygen 173 Sulphurous acid; its preparation 174 Properties of sulphurous acid 176 Synthesis “ “ 176 Oxidation “ “ 177 Bleaching by “ “ 177 Compound of sulphuric acid and chlo- rine 178 Sulphuric acid, monohydrated 178 Distillation of oil of vitriol 179 Analysis of sulphuric acid 180 Fuming “ “ 185 Manufacture of oil of vitriol 187 Hyposulphuric acid; its preparation.... 194 Its analysis 195 Hyposulpliurous acid 196 Hyposulphuric acid, monosulphuretted . 197 “ “ bisulphuretted 197 “ “ trisulphuretted 197 Equivalent of sulphur 198 Compounds of Sulphur and Hydrogen.. 201 Sulfhydric acid or sulphuretted hydro- gen 201 PAGE Its properties 202 Its analysis 204 Bisulphide of hydrogen.; 206 Compound of Sulphur and Nittogen.... 207 Selenium, its compounds with oxy- gen 208 Selenious acid 209 Selenic acid 209 Equivalent of selenium 211 Selenhydric acid 212 Tellurium, its compounds with oxy- gen 213 Chlorine, its preparation 215 Hydrate of chlorine 217 Bleaching by chlorine 218 Compounds of Chlorine and Oxygen 219 Chloric acid, its preparation and pro- perties 219 Analysis of chloric acid 220 Perchloric acid 222 Hypochlorous acid 223 Its analysis 224 Chlorous acid 225 Its analysis 226 Hypochloric acid 227 Equivalent of chlorine 228 Compound of Chlorine and Hydrogen... 230 Synthesis of chlorohydric acid 230 Preparation “ “ 231 Analysis “ “ 233 Compound of chlorohydric acid and ammonia 234 Compounds of Chlorine and Sulphur.... 234 Compounds of Chlorine and Nitrogen... 237 Aqua regia 237 Bromine, its properties 240 Bromic acid 241 Bromohydric acid 242 Iodine, its properties 244 Its preparation 245 Compounds of iodine and oxygen 245 “ “ hydrogen.... 247 “ “ nitrogen 248 “ “ sulphur 249 “ “ chlorine 249 Fluorine 250 Preparation of fluohydric acid 250 Engraving on glass by fluohydric acid 251 Analysis of fluohydric acid 251 Phosphorus, its properties 254 Distillation of phosphorus 256 Its preparation in the arts 257 Phosphoric matches 259 Compounds of Phosphorus and Oxy- gen 260 Preparation of anhydrous phosphoric acid 260 Preparation of hydrated phosphoric acid 262 Three series of hydrated phosphoric acid and their salts 263 Analysis of phosphoric acid 263 Phosphorous acid, its preparation 264 Hydrated phosphorous acid 264 Composition of phosphoric acid 266 Hypophosphorous acid 266 TABLE OF CONTENTS. 5 PAGE • Oxide of phosphorus 267 Equivalent of phosphorus 268 Compounds of Phosphorus and Hydro- gen 270 Phosphuretted hydrogen, gaseous.... 271 “ " liquid 273 “ “ solid ' 273 “ “ analyzed... 274 Compound of phosphorus and nitrogen 275 “ “ sulphur. 276 “ “ chlorine. 276 “ “ iodine.... 278 Arsenic, its properties 279 Its preparation 280 Compounds of Arsenic and Oxygen 280 Arsenious acid, properties of 281 “ “ analysis of 281 Arsenic acid 282 Compounds of arsenic and hydrogen.... 282 “ “ chlorine 283 “ “ sulphur 284 Researches on Poisoning by Arsenic.... 285 Antidotes 285 Distinctive characters of arsenious acid 285 Marsh’s apparatus 286 Detection of arsenious acid in animal matter 289 Boron, Boracic acid 292 Preparation of boracic acid 293 Properties of “ “ 294 Analysis of boracic acid, and equiva- lent of boron 295 Chloride of boron 295 Eluoride of boron 296 Silicium.—Silicic acid or silica 297 Preparation of silica 298 Equivalent of silicium 298 Chloride of silicium 299 Fluoride “ 301 Silicofluohydric acid 302 Carbon, different forms of 304 Power of charcoal to condense gases. 307 Decolorizing power of bone-black 307 Compounds of Carbon and Oxygen 309 Carbonic acid, its preparation 309 Its properties 310 Carbonated waters 311 Liquid and solid carbonic acid 313 Analysis of the gaseous acid 316 Carbonic oxide, its preparation 319 Its eudiometrie analysis 321 Chloroxycarbonic gas 321 Oxalic acid 322 Method of analyzing bodies composed of carbon, hydrogen, and oxygen... 323 Equivalent of carbon 327 Compounds of Carbon and Hydrogen... 330 Protocarburetted hydrogen, marsh gas 330 Its preparation 331 Its eudiometrie analysis 331 Bicarburetted hydrogen, olefiant gas... 331 Its eudiometrie analysis 332 Compound of Carbon and Sulphur 333 Analysis of sulphocarbonic acid 335 Compounds of Carbon and Nitrogen 337 PAGE Cyanogen, its origin and preparation... 338 Analysis of cyanogen 339 Cyanhydric. or prussic acid 342 Its analysis 344 On the Equivalents of the Metal- loids 346 METALS 349 State of metals in nature 351 Geology , 351 Stratified and non-stratified rocks 352 Position of stratified rocks 352 Causes of alteration in rocks 355 Primary and secondary formations... 358 Principal kinds of rocks 358 Geological division of the formations 361 Metallic veins 363 Physical Properties of the Metals 365 Opacity, lustre, and colour of the metals 365 Crystallization of metals 367 Malleability and ductility 368 Wire-drawing 369 Tenacity of metals 370 Conductibility and capacity for heat. 371 Chemical Properties of the Metals 371 Division of metals into six classes, ac- cording to their affinity for oxygen 372 Action of dry oxygen on the metals.. 374 Action of moist oxygen “ ... 375 Action of sulphur “ ... 376 Action of chlorine “ ... 376 Action of bromine and iodine “ ... 377 Action of phosphorus and arsenic on the metals 377 Action of carbon, boron, and silicium on the metals 377 Alloys 377 Properties of Metallic Oxides 380 Five classes of metallic oxides 380 Preparation of oxides 381 Behaviour of metalloids to metallic oxides 383 Behaviour of oxygen and hydrogen to metallic oxides 383 Behaviour of carbon and sulphur to metallic oxides 384 Behaviour of chlorine to metallic oxides 385 Behaviour of metals to metallic oxides 386 Properties of Metallic Chlorides 386 Preparation of metallic chlorides 386 Behaviour of metalloids to metallic chlorides 387 Behaviour of oxygen, hydrogen, and carbon to metallic chlorides 387 Properties of Metallic Bromides,Iodides and Sulphides 387 Behaviour of metalloids to sulphides. 388 Properties of Metallic Phosphurets and Arseniurets 389 On Salts 389 Neutral, acid, and basic salts 390 Coloured tests of neutrality 390 Laws of the constitution of salts 391 Monobasic and polybasic salts 395 Haloid salts 397 6 TABLE OF CONTENTS. PAGE Water of crystallization 398 Its determination 400 Solubility of Salts 402 Determination of solubility 404 Curves of solubility 406 Boiling points of saline solutions 410 Frigorific mixtures 412 Decomposition of Salts by Acids 413 Laws of these decompositions 414 Decomposition of Salts by Bases 416 Reciprocal action of Salts on Salts 417 Mutual action of Salts in the dry way 418 Mutual action of Salts in the wet way .. 418 Determination of the Electronegative body of a Salt or Binary Com- pound 421 The negative element of binary com- pounds 421 Determinations of oxides.., 422 Determination of sulphides and sele- nides 422 Determination of phosphurets and arseniurets 423 Determination of chlorides and bro- mides 423 Determination of iodides, fluorides, and cyanides 424 The acid of an oxysalt 424 Determination of nitrates and nitrites 424 Determination of chlorates, perchlo- rates, and hypochlorates 425 Determination of bromates, iodates, an d periodates 426 Determination of sulphates and sul- phites 426 Determination of hyposulphates and hyposulphites 427 Determination of phosphates, phos- phites, and hypophosphites 428 Determination of arseniates and ar- senites 429 Determination of carbonates, borates, and silicates 430 Determination of sulphosalts 431 I. ALKALINE METALS 434 Potassium 434 Preparation of potassium 435 Compounds of Potassium and Oxygen... 410 Anhydrous oxides of potassium 441 Salts of protoxide of potassium or po- tassa 442 Hydrates of potassa 442 Analysis of potassa 444 Equivalent of potassium...,. 445 Carhonates of potassa 446 Alkalimetry 448 Nitrate of potassa 453 Natural nitre-beds 454 Artificial nitre-beds 454 Lixiviation of nitrous matters 456 Extraction of crude saltpeter 457 Refining saltpeter 459 Testing saltpeter 461 Sulphates of potassa 464 Chlorate “ 464 Hypochlorite “ 467 Oxalates “ 467 PAGE Compounds of Potassium and Sul- phur 468 Sulphosalts of monosulphide of po- tassium 469 Chloride of potassium 470 Iodide “ 470 Cyanide “ 471 Distinctive Characters of the Salts of Potassa 471 Sodium „ 473 Compounds of Sodium and Oxygen 474 Anhydrous oxides of sodium 474 Salts of protoxide of sodium or soda 474 Hydrate of soda ; 474 Sulphates “ 475 Carbonates “ 477 Manufacture of soda-ash 477 Test of commercial sodas 480 Nitrate of soda 480 Phosphates of soda 480 Tribasic or common phosphates 481 Bihasic or pyrophosphates 483 Monobasic or metaphosphates 483 Constitution of the phosphates of soda 484 Chlorate of soda 486 Borates of soda 486 On the nature of flame 487 Manufacture of borax 491 Test of borax 493 Hyposulphite of soda 494 Compound of Sodium and Chlorine 495 Extraction of rock-salt 496 Evaporating salines 500 Extraction of sea-salt 503 Working up residues of salt-works... 506 Sulphides of sodium 510 Distinctive Characters of the Salts of Soda 510 Lithium, characters of its salts 512 Ammonia, its compounds 514 Ammoniacal salts and amides 515 Chlorohydrate of ammonia 516 Sulf hydrate “ 518 Sulphate “ 519 Nitrate “ 519 Phosphates " 519 Carbonates u 520 Behaviour of potassium and sodium to ammonia 520 Action of the battery on an ammoniacal solution 521 Distinctive Characters of Ammoniacal Salts 521 Determination of the alkalies and am- monia 522 Determination of potassa and soda in mixtures containing a single acid 523 II. ALKALINO-EAltTHY METALS. 528 Barium 528 Anhydrous oxides 529 Salts of Protoxide of Barium or Ba- ryta 532 Sulphate of baryta 532 Nitrate “ 532 Carbonate “ 532 TABLE OF CONTENTS. 7 PAGE Compounds of Barium with Sulphur and Chlorine 533 Sulphide of barium 533 Chloride “ 533 Distinctive characters of the salts of baryta 534 Strontium, anhydrous oxides 535 Salts of Protoxide of Strontium or strontia 535 Nitrate, carbonate, and sulphate of strontia 535 Salts of Strontium with Sulphur and Chlorine 536 Sulphide and chloride of strontium... 536 Distinctive characters of the salts of strontia 536 Calcium, oxides 537 Protoxide of calcium or lime 537 Determination of the equivalent of calcium 538 Lime-burning 539 Salts of Protoxide of Calcium or Lime 542 Sulphate of lime 542 The hydrous and anhydrous sul- phates 542 Boiling plaster, plaster-casting.... 545 Stucco 546 Native carbonates of lime 547 Calcareous incrustations 548 Nitrate of lime 549 Phosphates of lime 549 Chlorate of lime 550 Hypochlorite of lime 550 Manufacture of bleaching-salt 551 Chlorimetry 552 Sulphide, Chloride, and Fluoride of Calcium 556 Distinctive Characters of the Salts of Calcium 557 Magnesium, magnesia 558 Salts of Magnesia 559 Sulphate of magnesia 559 Nitrate “ 560 Carbonate “ 560 Phosphates “ 561 Silicates “ 562 Magnesium with sulphur and chlorine. 562 Distinctive Characters of the Salts of Magnesia '. 563 Determination of the alkaline earths, and their separation from each other and from the alkalies 564 III. EARTHY METALS 567 Aluminum, alumina 567 Composition of alumina 569 Salts of Alumina 570 Sulphate of alumina...... 570 Manufacture of the alums 571 Silicates of alumina, clays 575 Chloride of aluminum 577 Distinctive Character of the Salts of Alumina 578 Glucinum, salts of glucinum 579 Zirconium, salts of zirconia 582 Thorinum, Yttrium, Erbium, Terbium 583 Cerium, Lanthanum, Didymium......... 583 PAGE Determination of the earths; their sepa- ration from the alkalies and alka- line earths 584 Washing gelatinous precipitates 584 CHEMICAL ARTS DEPENDENT ON THE PRECEDING BODIES 586 Gunpowder, chemical theory of its ef- fects 587 Composition of different gunpowders.... 589 Preparation of materials 591 Saltpeter and sulphur, pulverization. 591 Charcoal, carbonization 592 Manufacture of gunpowder 594 By stamping 594 Granulation 596 By revolving 597 By mills 598 Polishing the grains 601 Round powder 602 Testing the strength of powder 606 Analysis of powder 608 Lime and Mortar.,., 611 Common mortar from fat lime 613 Hydraulic lime and cement 614 Composition of limestones.. 615 Manufacture of hydraulic cement 616 From fat lime 617 Concrete., 618 Analysis of limestones 618 Mastic and adhesives 620 Cementing apparatus 622 Glass, simple and compound silicates.. 623 Kinds of Glass 626 1. Alkaline glass 626 Bohemian 627 French 628 Blowing window-glass 631 Casting glass 635 2. Bottle glass......... 636 3. Lead glass, crystal or flint glass 637 Glass tubes... 638 Optical glass 639 Paste and enamel 640 Imperfections and Alteration of Glass. 641 Rupert’s drops and Bologna vial 641 Devitrification of glass,.... 642 Action of moist air on glass.. 642 Glass-working by the blowpipe.. 642 Coloured glass and painting on glass... 646 Analysis of glass, 648 POTTERY 651 General description, 651 Materials employed in the manufac- ture of... 651 Pottery, the Paste of which becomes compact by burning 652 Porcelain 652 Composition of clay for porcelain 652 Mechanical preparation of 653 Processes of forming articles of vari- ous forms 655 1. Throwing on the potter’s lathe 655 2. Press-work, 655 3. Moulding, or casting 655 Manufacture by the potter’s lathe, described 655 8 TABLE OF CONTENTS. PAGE Combining moulding with press- work 656 Moulding 657 Glazing 659 Porcelain kilns 659 Porcelain burning, or baking 663 French china 663 Composition of paste 663 “ glaze 663 Stoneware 664 Pottery, the Paste of which remains porous after burning 664 PAGE Earthenware, or fayence 664 Composition of paste 665 “ glaze 666 Common earthenware 666 Common building-bricks 666 Tiles 666 Fire-bricks for the construction of furnaces 667 Crucibles 667 Ornaments and Painting 667 Chemical Analysis of Earthenware 671 ELEMENTS OF CHEMISTRY. INTRODUCTION. § 1. When we bring the various bodies of nature into juxtaposi- tion, or into contact with each other, several kinds of phenomena result. Sometimes, these phenomena are displayed by important changes in the constitution of the bodies: at other times, on the contrary, the bodies acquire properties more or less fugitive, but which, in nowise, alter their apparent constitution, and do not sensibly change their respective weights. Thus, when a glass rod is rubbed with a piece of cloth, the rod acquires the property of attracting light bodies, such as the down of a quill, small bits of paper, etc., but the glass rod presents no apparent alteration whilst in possession of this property. When we place a magnet close to, or, better still, in contact with a bar of soft iron, we communicate to the latter the property of attracting objects of iron, but this property vanishes as soon as the magnet is Avithdrawn. If we rub with a magnet, not a bar of soft iron, but a bar of steel, the latter acquires the property of attracting objects of iron, even in the absence of the magnet, and preserves this pro- perty for some time. Under these various circumstances, the glass rod, the iron and steel bar, by acquiring neAV properties, experienced no sensible alteration in their constitution, and preserved their weight un- changed. If we mix together copper filings and pulverized sulphur, we may obtain a very intimate mixture of the two substances. To Avhatever degree of fineness, hoAvever, the particles of each may be reduced, avc can always distinguish with a lens or a microscope the particles of the copper from those of the sulphur, and can, therefore, conceive that their mechanical separation is possible. But, if Ave submit the mixture to the action of heat, a very bril- liant phenomenon soon ensues: a brilliant light is evolved, with a great quantity of heat. After the occurrence of this phenomenon, the microscope discovers a complete change in the constitution of the mass: it is impossible to distinguish the particles of copper 10 introduction. from those of the sulphur; the particles of the two bodies are in timately united—they have combined, and formed a new substance, perfectly distinct from its constituent parts. A piece of iron, exposed to the air, soon becomes covered wit! an ocherous coat, commonly called rust. If the piece of iron be long exposed to a damp atmosphere, it is so completely transformed into this ocherous substance as to lose all the characteristics of iron. If the iron had been carefully weighed before its exposure, its weight, compared with that of the resulting ocherous mass, would show that the latter was considerably heavier. Under these circum- stances, the iron has combined with one of the constituent prin-* ciples of the air, oxygen; and it has also combined with a portion of water, which always exists in the air in the state of vapour; and the result of these combinations is a new substance entirely differ- ent in its properties from those which entered into its composition. Thus, the various bodies of nature present, when in presence of or contact with each other, two very distinct classes of phenomena: phenomena more or less durable, discovered by no material change in their constitution, and phenomena, on the contrary, w'hich pro- duce an important alteration, and a complete change in their nature and in all their properties. The former class of these phenomena belongs to Physics: the latter is the province of Chemistry. Thus, we may define Che- mistry to be that portion of the natural sciences which treats of the phenomena resulting from the contact of bodies, when these pheno- mena effect an entire change in the constitution of these bodies. But, as it is essential that bodies thus made to react on each other should be clearly described, and their characteristic general pro- perties be previously perfectly well known, chemical science ne- cessarily contains a descriptive part, in which we treat, as it were, of the description or appearance of each body, by means of which it can always be subsequently recognised. § 2. Division of Bodies into Simple and Compound.—Che- mists divide bodies into simple and compound bodies. Compound bodies are those from which several substances may be extracted, differing in their properties from each other, and also from the primary body. Thus, common sea salt can be decomposed into two substances, chlorine and sodium ; nitre or saltpetre can also be decomposed into potassa and nitric acid. These last two sub- stances are themselves compound: from potassa, we can extract potassium and oxygen ; and from nitric acid, oxygen and nitrogen. On the contrary, chlorine, sodium, potassium, oxygen, and nitrogen have never yet been resolved into other principles, and hence have been designated by chemists as simple bodies. We, therefore, give the name of simple bodies to those sub- stances which, although subjected to the various manipulations of the laboratory, are never resolved into other substances. We do DIVISIBILITY. 11 not hereby mean to affirm that these bodies are really simple; for it is very possible that the future progress of science will enable us to effect decompositions which resist our present means ; and then many, perhaps all, of the bodies now regarded as simple, will be found to be compound. § 3. Divisibility of Matter.—Daily experience teaches us that bodies may be reduced into very minute particles; but is this divisibility of matter indefinite, or is it arrested at a certain point, at which the particles are no longer separable by mechanical means ? The ancient philosophers discussed this question at length, but without approaching its solution. The researches of modern chemistry have been more successful, and have proved, almost incontestably, that there is a limit to the divisibility of mat- ter. Chemists admit that ultimate analysis shows bodies to be composed of excessively small particles, indivisible by mechanical means; to these they have given the name of molecules or atoms. The molecules of simple bodies are themselves necessarily simple. The molecules of compound bodies are, on the contrary, compound: but all these complex molecules resemble each other, and are formed in the same manner. § 4. Different States of Bodies.—Bodies are presented to us in three different conditions, or states: the solid state, the liquid state, and the gaseous state. Some bodies may be readily obtained in these three different states: thus, water, which is fluid at the ordinary temperature of our latitude, becomes solid, under the form of ice, during the intense cold of winter ; whilst, by subject- ing it to the action of heat, it is easily made to assume the state of an aeriform fluid, or vapour. The solid and liquid states are common to many bodies; such as the majority of the metals, lead, tin, copper, silver, gold, etc. Some of them, such as iron and platinum, require, in order to pass from the solid to the fluid state, the highest degree of temperature of which our furnaces are capa- ble. Latterly, by means of the voltaic pile, a much greater degree of heat has been obtained, sufficient to render gaseous several of the metals, as gold, silver, copper, etc. The majority of substances which are gaseous at the ordinary temperature, become fluid when subjected to great pressure and a very low temperature. Hydrogen, oxygen, and nitrogen gas are the only ones which have hitherto resisted liquefaction: but this result cannot be doubted when we shall have attained a greater degree of compression and a more reduced temperature. The greater part of the gases, which have been liquefied, have been rendered solid by intense cold. It was sufficient to gradu- ally remove the pressure which kept the gas liquefied; the latter then endeavoured to assume the gaseous form: but, to effect this, the absorption of a certain proportion of latent heat, which was abstracted by the gaseous from the fluid parts, was necessary, and 12 introduction. thus the temperature of the latter became sufficiently reduced to congeal the fluid. We may therefore conclude that all natural bodies could assume the three states, were they subjected to favorable conditions of pressure and temperature. We may, however, remark that many solid bodies cannot be liquefied, because they are decomposed when submitted to the action of heat. Thus, carbonate of lime is decom- posed at a red heat, by disengaging one of its constituent prin- ciples, carbonic acid gas; and at this temperature, it has not undergone fusion. The disengagement of the carbonic acid may be prevented, by enclosing the carbonate of lime in a gun-barrel hermetically sealed: it then fuses at a temperature not much greater than that which effects its decomposition when subjected to the pressure of the atmosphere. § 5. Force op Aggregation, or Cohesion.—The force which unites the similar molecules of a simple or of a compound body, is called the force of aggregation, or cohesion. This force is very great in solid bodies, almost insensible in liquids, and entirely null in elastic fluids. In the latter, the particles, on the contrary, repel each other, and only maintain their actual distances by means of the pressure reacted by the sides of the containing •vessel. § 6. Chemical Affinity.—The force which unites the simple molecules constituting a molecule of a compound body, bears the name of chemical affinity. By virtue of this force, the molecules of simple bodies combine to form compound bodies. Chemical affi- nity greatly varies, according to the circumstances in which bodies are placed: it is not readily exerted between solid bodies, because the contact of the molecules cannot be perfect. The free exercise of chemical affinity demands the disaggregation of bodies, and, as this disaggregation can only be imperfectly effected by mechanical trituration, they must be reduced to a liquid or gaseous form. Corpora non agunt, nisi soluta, was an expression of the old che- mists, signifying this fact. In many cases, it is sufficient to liquefy or render gaseous only one of the bodies. The chemical affinity between two bodies varies greatly, accord- ing to temperature. Thus, lime and carbonic acid readily combine at the ordinary temperature to form carbonate of lime, and car- bonate of lime decomposes at a red heat, parting with its carbonic acid. At the ordinary temperature, the chemical affinity between lime and carbonic acid is very strong, whilst, at a red heat, it is null. § 7. Law of Multiple Proportions.*—When two simple bodies, A and B, combine, 1 molecule of A will combine with 1, 2, o, 4,.... molecules of B ; or, again, 2 molecules of A combine * First advanced by Dalton, in 1807. affinity. 13 with 1, 2, 3, 4, 5, 7,.... molecules of B; or, lastly, 3 molecules of A may combine with 5, 7,.... molecules of B, and so on. It therefore follows, that, in the various combinations which a sub- stance B may form with the same weight of a substance A, the ponderal quantities of the substance B will be to each other in rational and commensurable proportions. This fact, which has been clearly demonstrated by experience, is the principal proof we advance to establish the limited divisibility of matter and the existence of indivisible molecules. Experience even shows that the most simple proportions are those which most frequently occur: we generally find between compound bodies the propor- tions of 1 : 2, of 1: 3, of 1: 4, of 1: 5, or of 2 : 3, of 2 : 5, of 2 : 7. This law, which governs the proportions in which two bodies combine, is called the latv of multiple proportions. In our subsequent study of compound bodies, we shall meet with ex- periments establishing incontestably the truth of this law. § 8. Of the different Physical and Organoleptic Cha- racters ry which Bodies are distinguished.—In order to specify and describe bodies, we use various characteristics, founded either upon the appearance or physical properties of the bodies, or upon the impressions they produce on our organs. The former are called physical characters, the latter have received the name of organoleptic characters. The principal physical characters used in chemistry are— 1st. The various states of the body, that is, the various condi- tions of temperature and pressure in which the body presents the solid, liquid, or gaseous form. 2d. Its color in these different states. 3d. The nature of its lustre, when this can be specified by com- parison. Thus, we say, metallic, vitreous, resinous lustre, etc. 4th. Its degree of hardness, if the body be solid; and its greater or less fluidity, if it be liquid. 5th. Its specific gravity or density, that is, the weight of a unit of volume of the body. 6th. The regular or crystalline forms which it assumes. 7th. The appearance of the recent fracture of the body when solid. Thus, we say, vitreous, crystalline, laminated, granular fracture, etc. The organoleptic characters are those impressions produced on the organs of taste, smell, and touch: thus, we mark, by com- parison, the taste and smell of substances, and say that a sub- stance is rough, or has an unctuous or greasy feel. Of the physical characters we have just enumerated, some may be accurately measured or ascertained, and are therefore of great value in defining a substance. Such are, the specific gra- vity of substances, and the temperature at which they change their condition. The exact appreciation of their crystalline form 14 introduction. is also of vast importance : hence, the study of crystalline forms plays an eminent part in the classification of substances and in our modern chemical theories. § 9. Of Crystalline Forms.—A superficial observation of the different bodies in nature would lead us to suppose that their ex- ternal form had no regularity, and was susceptible of infinite variation. More attentive study, however, will soon teach us that the majority of them can assume, under certain circumstances, regular* forms, which are always perfectly similar in the various individuals of*the same substance. Still further, the majority of substances which appear to us under irregular external forms, present, in their recent fracture, evident indices of a regular or crystalline texture; so that the whole body is merely an aggrega- tion of an infinity of small crystals, dovetailed in each other. These rudimentary crystals are often so small that they can only be distinguished by examining the fracture with a lens or micro- scope ; whence it may be inferred that there are others still smaller, which escape our means of observation. The crystalline texture of bodies, far from being an exception, is, on the contrary, most frequently to be met with. The greater part of the substances which we prepare in our laboratories, are capable of being crystallized, that is, of assuming regular geometric forms; and we may remark that, when this operation takes place under identical circumstances, the forms of the various individual crystals perfectly resemble each other; so much so, as to constitute one of their most certain distinctive characters. At first sight, the crystalline forms assumed by the various bodies in nature appear to vary ad infinitum ; but a close study of these various forms has discovered some general laws which they obey, and which considerably lessen their number. The consideration of the crystalline forms of bodies already occupies an important place in our theories of chemistry. This importance will become greater when the science of crystallo- graphy shall be more extended, and when chemists shall define with accuracy the crystalline forms of the bodies which come under their notice. I have thought it necessary to explain here the prin- ciples of this science: they merely require of the .reader a know- ledge of elementary geometry. PRINCIPLES OF CRYSTALLOGRAPHY. § 10. Crystals are terminated by plane faces: and, in general, each plane face corresponds to another, exactly parallel to it, in the crystal; at least, when the crystal is isolated, and regularly * The science of crystallography has been chiefly established by the brilliant labours of Bergmann, of Rom6 de Lisle, and of Ilaliy. CRYSTALLOGRAPHY. 15 terminated throughout. Most frequently, crystals are imbedded in a solid mass, so that only one summit of the crystal is visible, and only one-half of it can be observed : it is therefore often difficult to verify the proposition just advanced as to the parallelism of the opposite faces; but nearly all imbedded crystals have been some- times found isolated and perfect, and thus justify the assertion. From analogy, we may therefore infer that such would be the case with all crystals, were they not imbedded, and we may repre- sent them perfect at both extremities. § 11. Crystals have always salient and never re-entering angles. Yet, Avhen we observe a mass of a great number of crystals, as, for example, the cavity of a rock, the walls of which are covered with crystals, and called by mineralogists a geode of crystals, or a crys- tallization obtained in the laboratory, we see many re-entering angles, which would seem to invalidate our remark ; but these re- entering angles are produced by the junction of two individual crystals, and are never seen in an isolated, individual specimen. § 12. Cleavage.—Crystals are not broken with equal readiness in all directions: the fracture generally follows the plane faces. These fractures in the direction of the plane faces may be indefi- nitely reproduced on the same crystal, parallel to each other, so that the substance may be divided in many laminae with parallel faces. This is called a lamellar fracture. This property of crys- tals has been long known to lapidaries, who have profited by it, to divide precious stones. Thus, the diamond presents a lami- nated fracture in four different directions : lapidaries avail them- selves of this, to remove the defective portions, and thus abridge, considerably, the cutting of the diamond. They call this opera- tion cleaving the diamond. The name of cleavage is applied to the parallel faces thus obtained in a crystal by fracture. The same crystal possesses several directions of cleavage: usually expressed by saying, many cleavages: but these cleavages are not always equally easy. Certain cleavages are readily known by the fracture, even when it is accidental: others are obtained only after much care; and even then, often very imperfectly. Thus, carbonate of lime presents three equally easy cleavages, inclined toward each other at an angle of 105° 5', and, in consequence of Avhich, the substance always breaks into rhombohedrons. Sulphate of lime also presents three cleavages, but one is much more easy than the other two : it follows, therefore, that the crystal tends to separate into laminae, and, by means of a knife, Aye can obtain laminae of exceeding thinness. If these laminae be broken betAveen the fingers, the other cleavages immediately appear, and give rise to parallelo- gramic laminae. §13. Fundamental Forms.—The combination of the planes of cleavage constitutes a geometrical figure, constant in all the 16 INTRODUCTION. individuals of the same crystallized substance: these are called f undamental forms. § 14. Natural Joints of a Crystal.—The lines in the direc- tion of which the faces of a crystal are divided, are called the joints. They are distinguished into acute and obtuse, according as the faces constituting these joints form with each other an acute or an obtuse angle. § 15. Angles of a Crystal.—Three or a greater number of faces, uniting at one point, form a solid angle, which mineralogists term, though improperly, the angle of the crystal. The angles are classed according to the number of their faces: thus we say, an angle with 3 faces (fig. 1), with 4 (fig. 2), with 6 (fig. 3), etc. etc. Fig. 1. Fig. 2. Fig. 3. § 16. Simple and Compound Forms.—Sometimes, crystals are only terminated by faces similar to each other. Such are, the regular octahedron, formed by 8 equilateral triangles (fig. 4); the regular hexahedron, or cube, terminated by G squares (fig. 5); the hexagonal dodecahedron, formed by 12 isosceles triangles (fig. 6). These are called simple forms. We call, on the contrary, com- Fig. 4. Fig. 5. Fig. G. pound, those forms which include faces of different kinds. Fig. 7 represents a compound form : it is composed of 6 square faces and 8 equilateral triangles. Fig. 8 is also a compound form, and constituted by G rectangular faces and 12 isosceles triangles. If, in a compound crystal, we conceive the faces of the same kind to be extended so as to hide completely the faces of the other kind, we will obtain a simple form. The triangular Fig. 7. CRYSTALLOGRAPHY. 17 faces of fig. 7 being extended, give the regular oc- tahedron (fig. 4). If, on the contrary, we extend the square faces so as to conceal the triangular faces, we will obtain the hexahedron (fig. 5). Hence we see that compound forms result from the combination of so many simple forms as there are faces of different kinds in those compound forms: we may, therefore, call the compound form of which we have just spoken (fig. 7), a combination of the octahedron and hexahedron. It often happens that, by extending the simi- lar faces of a compound crystal, we obtain an unlimited form, which cannot of itself terminate a crystal. Thus, for example, if we suppose the 6 rectangular faces of fig. 8 to be produced, we will obtain a regular prism with 6 indefinite faces. If, on the contrary, we produce the 12 triangular faces, we obtain a solid, the hexagonal dodecahedron (fig. 6). It is evident that the faces which form an unlimited solid cannot, of themselves, produce a crystal: they will always appear in combina- tion, either with faces which, being produced, will give a solid, or with faces which, under the same conditions, give open or indefinite forms. § 17. Dominant and Secondary Forms.—Generally speaking, in a compound crystal, one of the simple forms constituting it is more developed than the others, and gives the crystal its general aspect: this is then called the dominant form, whilst the other forms of the combination are termed second- ary; their faces are also called modifying faces. Thus, figs. 9 and 10 represent com- binations of the octahedron with the hexahe- dron; but in fig. 9 the facets o pertaining to the octahedron are more developed than the facets a belonging to the hexahedron, and give the crystal an octahedral form: we therefore say that it is an octahedron modi- fied by the faces of the cube. On the con- trary, in fig. 10, the aspect of the cube pre- dominates, and it will be termed a hexahedron modified by the faces of the octahedron. § 18. Truncation.—When, in a combination of several simple forms, an edge or joint of the dominant form is succeeded by a face parallel to this edge, as in fig. 11, the edge is said to be truncated, and the modifying face is called the truncated face, or facet of the edge.* This truncated face may incline equally toward Fig. 8. Fig. 9. Fig. 10. * The edge is also said to be replaced by a plane, either evenly or obliquely; or, as in 8 19, by two planes.—J. C. B. 18 introduction. both faces of the dominant form which enclose the truncated edge : the truncation is then said to be right, or tangent, as in fig. 11. In the contrary case, it is said to be oblique. Frequently, the angles of the dominant form are truncated, and the faces of the truncation are right or oblique, according as their inclination to the faces of the domi- nant form which make the angle is equal or unequal. Fig. 9 represents a regular octahedron, the angles of which are truncated by the faces of the hexahedron : fig. 10 repre- sents a hexahedron the angles of which are truncated by the faces of the octahedron. In both figures, the truncation is right, or tangent. When an oblique truncated face of an angle inclines equally on the two faces, forming one of the edges of that angle, the trunca- tion is said to rest, symmetrically on this edge : thus, in fig. 11, the truncated face inclines equally to the two faces : it rests sym- metrically on the edge. In the contrary case, it is said to rest obliquely. Again, we say that a truncated face rests symmetrically on a face of the dominant form, when the line of intersection of these two faces forms equal angles with the two adjacent edges of the dominant form ; we say, on the contrary, that it rests obliquely, when these angles are unequal. § 19. Bevelment.—The edges of the dominant form are often replaced by two faces parallel to these edges, and equally inclined toward the adjacent faces: in this case, Ave say that the edge has been beveled. Such is the case in fig. 12, where a bevelment has taken the places of the edges of the hexahedron. § 20. Acumination.—An angle of the dominating form is often repla- ced by another more obtuse an- gle ; the angle is then said to be acuminated. Sometimes, the fa- cets of the acumination are equal in number to the faces forming the primitive angle, as in fig. 13: at others, the latter are double, as in fig. 14. The terminal faces rest symmetrically, either on the faces (fig. 13), or on the edges of the angle (fig. 14). § 21. Centre of the Crystal.—In all crystals, whether Fig. 11. Fig. 12. Fig. 13. Fig. 14. CRYSTALLOGRAPHY. 19 simple or compound, there is a point at which every right line which passes through and is terminated by the faces of the crys- tal must be bisected. This point is the centre of the crystal. § 22. Axes of the Crystal.—In all simple forms, there are certain right lines pass- ing through the centre of the crystal, and around which the faces are symmetrically ar- ranged ; these lines have been called the axes of the crystal. Sometimes the crystal has several planes or systems of axes, as the regular hexahedron. In fact, if we connect by lines the centres of the opposite faces (fig. 15), we shall have three right lines possessing the above property, and which, consequently, are axes: we obtain a second system of axes by joining the opposite angles (fig. 16). This gives a system of 4 axes, forming with each other angles of 70° 32': lastly, if we join, two by two, the centres of the opposite edges (fig. 17), we obtain a system of 6 axes, com- prising angles of 60°. All the axes of the hexahedron, which make a part of the same system, are equal to each other. Fig. 15. Fig. 1G. Fig. 17. Fig. 18. Fig. 19. In the hexagonal dodecahedron (fig. 18), we obtain the axes by joining the opposite angles; and thus have a plane of 8 equal horizontal axes, forming with each other angles of 60°, and a single vertical axis perpendicular to the plane of the three others. In the oblique rhombic octahedron (fig. 19), the axes are still the lines joining the opposite angles ; the three axes of this figure are all unequal and inclined toward each other. § 23. Position of the Crystal.—In order to study more readily crystalline forms, it is useful to give to crystals a deter- minate position; and it has been agreed to place them so that one of their axes shall he vertical. Thus, in the hexadron, we gene- rally adopt, as the plane of axes, that system of three rectangular axes which join the centres of the opposite faces. As these three axes perfectly resemble each other, it is evident that either may 20 introduction. be chosen, and that the figure will present exactly the same aspect, whichever axis may be vertical. When, in the system of axes of a crystal, there is found an axis having no analogue in the system, this axis is always chosen for the vertical position, and is then termed principal axis ; the others are called secondary axis. The hexagonal dodecahedron (fig. 18) is placed so that its single axis may be vertical. The rhombic octahedron (fig. 19) presenting three unequal axes, either may be chosen as the principal axis: but, the selec- tion once made, it should be continued during the whole study of the crystal. Neither should the direction of the secondary axis be arbitrary, when we are about to study the various crystals presented by the same substance. In all our figures, the crystals will be placed so that one of the secondary axes shall have the direction of the plane of the figure. § 24. Definitions of Systems of Crystallization.—The accurate study of the various compound forms furnished to us by the mineral kingdom has showrn that simple forms cannot indiscri- minately combine with each other. We never meet, in the same crystal, simple forms, which have not an identical system of axes. Thus, we frequently find the cube and regular octahedron in combination ; the rhombohedron is found combined with a regular six-sided prism : but we never see the rhombohedron or six-sided prism combined with the regular octahedron or hexahedron. The rhombohedron and regular six-sided prism have a system of axes composed of 3 similar axes forming with each other angles of 60°, and situated in the same plane, and a fourth axis perpendicular to the other three; whilst, in the cube and regular octahedron, there is found no analogous system of axes. We give the name of System of Crystallization to the collection of the different forms which have similar systems of axes. Crystallographers have made six systems of crystallization: 1. The first, or regular system of crystallization, is characterized by 3 axes of equal length, and intersecting at right angles.1 2. The second system is characterized by 3 perpendicular axes, but two only are of equal length.2 3. The third system is characterized by 4 axes, three of which are of equal length, disposed in the same plane, and intersecting at an angle of 60°; the fourth axis is different, and is perpendi- cular to the system of the three others.3 1 This is the Monometric or Tesseral system of Dana: the Isometric of Haus- mann: Tessular of Mohs and Haidinger: Tesseral of Naumann: the Regular of G. Rose: the Cubic of Dufrenoy. 2 This is the Dimetric system of Dana : Pyramidal of Mohs : Tetragonal of Nau- mann : Monodimetric of Hausmann: Quadratic of Kobell: Bino-Singulaxe of Weiss. 3 This is the Hexagonal or Rhombohedral system of Dana: the Rhombohedral of Mohs: Hexagonal of Naumann: Monotrimetric of Hausmann. T. F. B. CRYSTALLOGRAPHY. 21 4. The fourth system is characterized by 3 axes, unequal, but intersecting at right angles.1 5. The fifth system is characterized by 3 unequal axes: two of these axes intersect obliquely, but the third is perpendicular to the plane of the other two.2 6. The sixth system is characterized by 3 unequal axes, which intersect obliquely.3 We will now review successively the principal crystalline forms constituting these different systems. § 25. The forms of this system are characterized by 3 axes of $qual length, intersecting at right angles. As we have seen above, other systems of axes are found in these forms; but as these other systems present no new feature, and the regular system of crys- tallization being completely defined by the three equal and similar axes, we shall consider them alone. The principal simple forms belonging to this system are— 1st. The regular octahedron (fig. 20), formed by 8 equilateral triangles: the edges are equal; the solid angles are equal and have 4 faces. The dihedral angles of the faces are of 70° 32'. The rectangular axes join the opposite angles, and each face divides the three axes in equal lengths. It therefore folloAvs, that if we designate by a the length of these axes, comprised between the centre of the crystal and the point where it meets the faces, we may define the octahedron by saying, that it is the face which divides the three rectangular axes into equal lengths, a. The following formula, which expresses the equality of the three axes, has been agreed upon to represent this face: (a : a : a). 2d. The hexahedron or cube (fig. 21) form- ed by 6 square faces: the three rectangular axes joining the centres of the opposite faces: each face is perpendicular to one of the axes, and parallel to the other two; so that we may represent each of these faces, and conse- quently the entire hexahedron, by the formula (a : oo a : oo a). I.—REGULAR SYSTEM OF CRYSTALLIZATION. Fig. 20. 1 This is the Trimetric system of Dana: the Prismatic of Mohs : the Rhombic and Anisometric of Naumann: Binary of Weiss: Trimetric of Hausmann. a This is the Monoclinic system of Dana: the Ilemi-prismatic of Mohs: the Monoclinoheclral of Naumann : the Clino-rhombic of Kobell: the Augitic of Ilaidinger. 3 This is the Triclinic system of Dana: the Tetarto-prismatic of Mohs: the Trichinohedral of Naumann: the Clino-rhomboidal of Kobell: the Anorthic of Haidinger. T. F. B. Fig. 21. 22 INTRODUCTION. 3d. The dodecahedron (fig. 22), formed by 12 rhombic faces, the angles of which are of 109° 28', and 72° 32'. The angles are of two kinds: 6 angles A to 4 faces corresponding in position to the an- gles of the octahedron, and 8 angles 0 to 3 faces corresponding to the angles of the hexa- hedron. Each face of the dodecahedron is parallel to an octahedric axis, and bisects the other two: these faces may be therefore represented by the formula (a: a: oo a.) 4th. The tetrahedron (figs. 23 and 24) is formed by 4 faces which are equilateral triangles. This solid may be derived from the regular octahe- dron by supposing the al- ternate faces of the octahe- dron to be extended so as to cause the intermediate faces to disappear. As we cause one or other system of alternate faces to disappear, we shall obtain two tetrahedrons (figs. 23 and 24) perfectly equal, but easily distinguished from each other by their position. This mode of generation of the tetrahedron has given to this form the name of hemioctahedron. 5th. The tetrahexahedron or pyramidal hexahedron (fig. 25) is a solid with 24 faces, the general appearance of which is that of a hexahedron on whose faces 4-sided pyra- mids have been superimposed. Most frequently, the height of the 4-sided pyramids implanted on the faces of the hexa- hedron is equal to one-half of the axis of the hexahedron. Sometimes, however, the ratio between the height of the pyramids and the axis of the hexahedron is more complex, but it is always represented by a very simple ra- tional fraction. Thus, we meet the ratios f, 2 ,1 .1 f)J 35 5* If we suppose the alternate faces of the tetrahexahedron to be extended, so as to cause the intermediate faces to disappear, we obtain a figure with 12 pentagonal faces, the penta- gonal dodecahedron (fig. 26), which may be also called hemi-tetrahexahedron, on account of its mode of ge- Fig. 22. Fig. 23. Fig. 24. Fig. 25. Fig. 26. CRYSTALLOGRAPHY. 23 neration. The same tetrahexahedron may generate two hemi- tetrahexahedrons, according as one or other of the alternate sys- tems of faces is extended. The two figures will be perfectly equal, and differ only in the direction of their faces. Oth. The trisoctahedron or pyramidal octahedron (fig. 27) is a solid of 24 faces, having the general appearance of a regular octahedron, on the faces of which 8-sided pyramids have been superimposed. As there are several tetra- hexahedrons, presenting dif- ferent ratios between the elevations of the pyramids and the length of the axes, so there are several trisocta- hedrons : the ratios between the elevations of the pyra- mids and the length of the axes is always represented by very simple fractions, as J, J, f. By extending the alternate faces of the trisoctahedron we obtain a hemihedric figure (fig. 28), the hemi-trisoctahedron, which has hitherto been but rarely observed. 7th. The icositetrahedron (fig. 29) is a solid with 24 faces, 48 edges, and 26 angles. This figure is obtained by supposing the solid angles of the regular octahedron to be replaced by more obtuse 4-sided pyramids, as in fig. 80, and then sup- posing the faces of these pyramids to he extended so as to cause the faces of the octa- hedron to entirely disappear. The ratios be- tween the elevations of these pyramids and the length of the axes of the octahedron may be different; so that several icosotretrahe- drons may exist: but the ratio is always represented by a very simple rational frac- tion : hitherto we have discovered only the relations J and J. Icositetrahedrons are rarely seen in crys- tals, and are only mentioned here to com- plete the list.* § 26. Compound Forms of the regular System of Crystallization.—The combi- Fig. 27. Fig. 28. Fig. 29. Fig. 30. * The origin assigned for the 24-liedron is for that in figs. 29, 80, or " but may be derived from the rhombic 12-hedron by the evenly replacing its edges by a plane. The latter form is common in Garnet, Leucite, Analcime, &c. The general formula for the 24-liedron is a : a : la, in which »i = 2 or 3.—J. C. B. 24 INTliODUCTION. nations of the various simple forms just enumerated give rise to various forms, of which we shall mention the principal. Figs. 31, 32, and 33 represent three combinations of the regu- lar octahedron and the hexahedron, in which the faces of the octahedron are marked o, and those of the hexahedron a. In fig. 31, the faces of the octahedron predominate : the contrary is true in fig. 32; and lastly, in fig. 33 the two forms are equally developed: this last species has received the name of cubo-oetahe- dron.* Figs. 34 and 35 repre- sent combinations of the dodecahedron and octa- hedron. In fig. 35 the faces o of the octahedron predominate, whilst in fig. 34 the predominating are the faces d of the dode- cahedron. In fig. 36 we find a combination of the hexa- hedron with the dodeca- hedron, the hexahedron a predominating. Fig. 37 represents a combination of the hex- ahedron predominating, with the tetrahedron: it will be seen that of the 8 solid angles of the hexahedron, 4 only are truncated by the faces o of the tetrahedron, or hemi-octa- hedron. We find in this combination an exception to the law laid down (§ 10); namely, that, in every regularly terminated crystal, every face has a face parallel to it. In the combination of the hexahedron with the tetrahedron, the facets o of the tetrahedron have no parallel facets on the opposite angles of the hexahedron. Fig. 31. Fig. 32. Fig. 33. Fig. 34. Fig. 35. Fig. 36. Fig. 37. * They are also said to be in equilibrium.—J. C. B. CRYSTALLOGRAPHY. 25 Fig. 38 represents the combination of the two tetrahedrons, which are obtained by ex- tending the two systems of alternate' faces of the same octahedron. The preceding forms result only from the combination of two simple forms of the regu- lar system; but compound forms are some- times seen more complicated, resulting from the combination of three or more simple forms. Thus, fig. 39 repre- sents the combination of the predominating hexahe- dron a, writh the octahedron o and the dodecahedron d. Fig. 40 is a combination of the predominating hexa- hedron a with the dodeca- hedron d and the tetrahe- dron o.* Fig. 38. Fig. 39. Fig. 40. II.—SECOND SYSTEM OF CRYSTALLIZATION. § 27. The forms belonging to this system are characterized by 3 rectangular axes, two of which are equal, and the third un- equal. We choose the unequal axis as the principal one, and give the crystal such a position as will make this axis vertical. The ratio between the principal axis and the secondary axes may vary very much; and, generally speaking, this relation is com- plex and irrational. In the regular system of crystallization, the 3 axes being equal, the faces are symmetrically arranged with reference to the 3 axes. Such is not the case in the second system: the 2 secondary axes being similar, the faces are symmetrically arranged with reference to these two axes, but differently with reference to the principal axis. Therefore we meet in this system faces perpendicular to the principal axis which have no analogues to the 2 secondary axes: and, again, we see faces perpendicular to the 2 secondary * One simple form is entirely omitted in the above enumeration, which has been observed in the Diamond, Garnet, &c., the hexakisoctahedron or solid of 48 triangular planes. It may be viewed as arising from a combination of the 12-hedron, d, and the 24-hedron, or 6 X 8-hedron by replacing the edges of this * l i combination evenly by a plane. Its formula is a : — <« : — and 5 different values for m and n have been observed. This simple form has two hemi-forms, one of which is like the hemioctahedron with a low 6-sided pyramid raised on each plane ; the other is formed by supposing' a pair of adjacent planes to be extended so as to exclude adjoining pairs of planes. The latter forms a 24-planed solid with tra- pezoidal planes, but different in appearance from fig. 29, and is termed the hemi-octakishexahedron.—J. C. B. 26 introduction. axes which have necessarily no analogue to the principal axis. These faces thus form open prisms, which cannot, of themselves, terminate a crystal. No similar occurrence is found in the regu- lar system of crystallization. Simple Forms of the Second System of Crystallization.— Octahedrons with a square Base (fig. 41). The faces of these octahedrons are isosceles triangles: their edges are of two species : 8 terminal edges D, which converge toward the principal axis CC, and 4 lateral edges. A section through these lateral edges, and, conse- quently, through the secondary axes, gives a square (fig. 41 a), and is called the base of the octahedron. Sections through the terminal edges give rhombs (fig. 41 b). In the regular system of crystallization we have but one octa- hedron : in the second svstem. on the contrary, we have an in- finity of octahedrons with square bases, which differ from each other by the inclination of their faces, or the ratio which the length of the principal axis bears to that of the 2 equal secondary axes; and, in order to define the octahedron, this ratio, or the inclination of the faces, must be given. In all octahedrons of the second system of crystallization, the principal axis always joins the apices of the octahedron ; but the secondary axis may present two positions, differing either with reference to the lateral edges of the octahedron, or to the sides of the base. These axes may join the opposite angles of the base, as in fig. 42; or the centre of the opposite sides, as in fig. 43. In fact, we thus obtain two oc- tahedrons, with square bases and perfectly equal systems of axes : the octahedron of which the base is fig. 43 has its faces inclined in the same manner as the edges of the octahedron the base of which is fig. 42. We distinguish them by calling the octahedron, of which the base is fig. 42, an octahedron of the first class, or direct octahedron; and that of which fig. 43 is the base, an octahedron of the second class, or inverse octahedron. When the same substance crystallizes in octahedrons with square Fig. 41. Fig. 41 a. Fig. 41 b. Fig. 42. Fig. 43. CRYSTALLOGRAPHY. 27 bases, it does not always present the same octahedron. On the contrary, we often observe several very different octahedrons, but which bear to each other a very simple relation. This relation is thus expressed: if we reduce these octahedrons to the same base, their elevations or principal axes ivill bear to each other rational and extremely simple relations. One of these octahedrons is chosen as the principal or primitive form; and that is generally preferred which occurs or which pre- dominates most frequently.* The principal or primitive form is expressed by the formula (a : a : c). The formula of all octahedrons in which the base is situated, with reference to the secondary axes, in the same manner as in the principal octahedron, that is, of all direct, or octahedrons of the first class, then becomes, [a : a : me): whilst those of the inverse, or octahedron of the second class, in which the inclination of the faces is equal to that presented by the edges of the corresponding octahedrons of the first class, will be (a : oo a : me). If an arbitrary value could be assigned to m, the number of these octahedrons would be infinite: but observation has shown, that in the various octahedrons with square bases presented by any one substance, the number m always has a rational and extremely simple rule: thus, the formula of the primitive octahedron being (a : a : c), we always find, in the same substance, octahedrons expressed by, (a : a : 2c) (a : a : Jc) (a : a : 3c) or, (a : a : Jc) (a : a : 3c)' {a : a : Jc); that is to say, octahedrons which, with equal secondary axes, have a principal axis, 2, 3, 4 times greater, or 2, 3, 4 times smaller, than the principal axis c of the primitive octahedron. We also find corresponding octahedrons of the second class : * That which is most frequent in a particular mineral species or chemical combination. It is termed the primary or radical 8-hedron.—J. C. B. 28 INTRODUCTION. (a : oo a : c) (a : oo a : c) (a : oo a : 2c) (a : oo a : Jc) (a : oo a : Sc) ’ (a : cc a : %c) {a : oo a : 4e) (a : cc a : There are, however, still two cases which are often met with in substances belonging to the second system of crystallization, and they are derived from the formulae (a : a : me) {a : oo a : me), by giving m its limited values, that is, by making m — o, or, m = go. Right Terminal Plane.—By diminishing the value of m, we obtain octahedrons more and more flattened, and when m = o, the octahedron is reduced to its base: we will call this extreme form of the octahedron, the right terminal plane. This terminal face can never exist isolated; but it is frequently met with in combination in compound forms. The formula of this face might be {a : a : 0 a : ao c). The prism predominates in the combination.* Fig. 49. Fig. 50. * I have taken the liberty of altering the text in this paragraph, as an error has inadvertently crept in. Two simple forms of this system have been omitted, the dioctahedron and its prism. The dioctahedron may be said to arise from the 30 INTRODUCTION. III.—THIRD SYSTEM OF CRYSTALLIZATION. § 29. The forms belonging to the third system of crystallization are characterized by 4 axes, of which 3 are equal, and intersect at angles of 60°: the fourth axis is different, and perpendicular to the other three. This is chosen as the principal axis, and the three axes become secondary. The ratio between the length of the principal axis and that of the secondary axes is indefinite. It is evident that, in this system of crystallization, the faces are symmetrically disposed with reference to the three secondary axes, whilst they are disposed, with reference to the principal axis, in a manner entirely different from their former arrangement. In the third, as well as in the second and following systems, indefinite forms are met with which cannot, of themselves alone, terminate a crystal. The following are the principal forms of this system : Hexagonal Dodecahedrons (fig. 51).* These forms have 12 faces, 18 edges, and 8 angles. The faces are isosceles triangles. The edges are of two kinds : 12 terminal D, and 6 lateral G. The angles also are of two kinds: 2 terminal angles C, which are regular and six-sided, and 6 lateral angles with four sides, which are nearly symmetrical. A section through the lateral edges gives the base, which is a regular hexagon, and contains the three secondary axes. Sections through the two parallel ter- minal edges give rhombs. Two classes of dodecahedrons may be distinguished, according to the manner in which the secondary axes are arranged with reference to the base. In the first class, the axes join the angles of the base, as in fig. 52. In the second, the axes join the centre of the opposite side, as in fig. 53. Fig. 51. Fig. 52. Fig. 53. 8-hedron by bevelling all its terminal edges until the planes of the 8-liedron dis- appear. It gives an 8-sided pyramid, whose alternate terminal edges are equal, and the adjacent ones unequal. Its formula is a : na : me. It is never isolated, and in combination appears as 3 in fig. 50. The other omitted form is an 8-sided prism, with its alternate angles equal, adjacent unequal. None are represented in the figures, but its planes would appear replacing the edges of combination be- tween a and g, fig. 49.—J. C. B. * They are generally termed triangular 12-hedrons, but may be termed hexago- nal with reference to the plane of the three axes.—J. C. B. CRYSTALLOGRAPHY. 31 The faces of the direct, or dodecahedrons of the first class, di- vide the secondary axes, in equal lengths a, and are parallel to the third. The faces of the corresponding inverse, or dodecahedrons of the second class, divide one of the secondary axes, in a length a, and their prolongation divides the other two secondary axes, in the lengths 2a. The dodecahedron of the first class is therefore expressed by the formula (a : a : ao a : c) and the corresponding form of the second class is (2a : a : 2a : c). Besides the two dodecahedrons of which we have just spoken, there may be an infinity of others, belonging to both classes, and of which the formulae will be (a : a : oo a : me) (2a : a : 2a : me). But, in one and the same substance, dodecahedrons always bear to each other very simple relations. If we refer these dodecahe- drons to systems of equal secondary axes, their principal axes will hear to each other very simple rational relations: or, in other words, the value of m will he 2, 3, 4, or, §, We will, therefore, choose one of these dodecahedrons as the principal or primitive form, or type, and will select that which most frequently occurs, or generally predominates in combinations. By giving to the primitive form the formula (a : a : oo a : c), the formulae of dodecahedrons will be— Dodccah. of 1st class. Dodecah. of 2d class. (a : a : oo a : c) (2a : a : 2a : c) (a : a : oo a : 2c) (2a : a : 2a : 2c) (a : a : oo a : 3c) (2a : a : 2a : 3c) (a : a : oo a : 4c) (2a : a : 2a : 4c) (a : a : oo a : Jc) (2a : a : 2a : Jc) (a : a : oo a : Jc) (2a : a : 2a : (a : a : go a : \c) (2a : a : 2a : Jc) We frequently find, in addition, indefinite forms, which may be considered as the extreme forms of the dodecahedrons, and which are obtained by making m = o or m — oo, in the two general for- mulae (a : a : oo a : me) (2a : a : 2a : me). 32 INTRODUCTION. By making m = o, the dodecahedrons are reduced to a single face, parallel and similar to the base. We Avill call this face the right terminal plane. The formula of this face will be, therefore, (a : a : go a : 0c), but it is generally written (oo a : co a : co a : me); that is to say, that this face is considered as the limit of the hex- agonal dodecahedrons having the principal axis me, but of which the secondary axes, Avithout losing their equality, have been pro- duced to infinity. By making m = oo, we obtain tAvo 6-sided prisms, expressed by (2a : a :2a : oo c) (a : a : oo a : co c). Rhomb ohedrons, or, Hemidodecahedrons (fig. 54).—Rliombohedrons have 6 faces, 12 edges, and 8 angles. The faces are rhombs. The edges are of two kinds: 6 terminal X, and 6 lateral. The angles are also of two kinds: 2 regular 3-sided angles C, and 6 lateral 3-sided but irregular angles E. The principal axis joins the two terminal angles C, whilst the secondary axes join the centre of the opposite lateral edges. Sec- tions passing through the two oblique diagonals CE and C'E' are rhombs, the planes of which are perpendicular to the faces of the rhombohedron to which these diagonals belong. There are three of these sections in a rhombohedron, and they are called principal sections. The rhombohedron may be considered as being derived from the hexagonal dodecahedron, by a mode of generation similar to that by which we have derived the tetrahedon from the regular octahe- dron ; that is to say, by supposing that the alternate faces of the dodecahedron were produced so as to eclipse the intermediate faces: we then have remaining only the faces of the dodecahe- dron, pairs of which are parallel. But, as we may choose either system of alternate faces, it is evident that we will obtain two rhombo- hedrons (figs. 54 and 55) which are perfectly equal, and which wrould be lost in each other, if one of them of 60° were made to revolve around its principal axis. We shall call these rhombohedrons direct, or, rhombohedrons of the first class ; and inverse, or, rhombohedrons of the second class. As the faces of rhombohedrons coincide with those of the hexagonal dodecahedron, their formulae will be the same as Fig. 54. Fig. 55. CRYSTALLOGRAPHY. 33 those of the faces of the dodecahedron: but, in order to dis- tinguish them from the latter, the coefficient J is prefixed. Hence, the formula of rhombohedrons of the first class will be : a : oo a : me), and that of the second, : a' : aoc : me). We have accented the first two axes, in the latter formula, in order to express that rhombohedrons of the second class divide, ■when produced, the axes which are directly divided by rhombohe- drons of the first class. The same substance, belonging to the third system of crystal- lization, often presents several rhombohedrons of the first and of the second class. If we suppose the secondary axes of these rhombo- hedrons to be equal, we shall find that the lengths of the principal axes bear to each other rational and simple relations. One of these rhombohedrons is generally selected as the primitive form, and the others are compared with it. We shall consider as belonging to the first class, those rhombo- hedrons of which the faces are arranged in the same manner as the faces of the principal rhombohedron: and to the second class, those of which the faces are arranged in the direction of the edges of the principal rhombohedron. Rhombohedrons present, like octahedrons of the second system of crystallization, series of figures more obtuse and more acute. Each obtuse rhombohedron of this series has its faces inclined toward the principal axis, in the same manner as the edges of the acute form which immediately follows it; so that each form is the first acute rhombohedron of that which precedes it, and the first obtuse rhombohedron of that which follows it. The following for- mula express the rhombohedrons of this series: Principal rhombohedron (a : a : oo a : c) 1st obtuse “ (a' : a' : oo a : %c) 2d “ “ (a : a : co a : \c) 3d “ u (a' : a’ : go a : c) 1st acute “ (a' : a' : cc a : 2c) 2d “ “ (a : a : oo a: 4c) 3d “ “ {a! : a' : oo a : 8c). We sometimes meet, however, rhombohedrons of which the prin- cipal axes are 3 or 5 times the length of the principal axis of the primitive rhombohedron. 34 INTRODUCTION. Didodecahedrons (fig. 56).—This form has 24 faces : it presents the general appearance of an hexagonal dodecahedron the faces of which are replaced by bevelments, having their edges directed in the same direction as the diagonals of the faces of the dodecahe- dron. The faces of the didodecahedron are scalene triangles. The 36 edges are of three kinds : 1st, 12 terminal edges D, which correspond, in position, to the terminal edges of the hexagonal dodecahedron of the first class; 2d, 12 terminal edges F, corre- sponding, in position, to the terminal edges of the dodecahedron of the second class; 3d, 12 lateral edges Gr, corresponding, two by two, with the lateral edges of the dode- cahedron. The angles are of three kinds: 1st, 2 angles with 12 symme- trical faces C, corresponding to the terminal angles of the octahe- dron ; 2d, 6 lateral 4-sided and symmetrical angles A, corre- sponding to the lateral angles of the hexagonal dodecahedron of the first class; 3d, 6 lateral 4-sided and symmetrical angles E, corresponding to the lateral angles of the hexagonal dodecahe- dron of the second class. The principal axis joins the terminal angles C; the secondary axes join the first lateral angles A. The most common formula of didodecahedrons is Fig. 56. (a : na : pa : me); the coefficients n, p, m, representing very simple whole numbers, as 1, 2, 3, 4—or very simple fractions, as J, etc. Didodecahedrons are never seen in crystals, as predominating forms; but they are frequently met with as modifying faces, in combinations, principally in those in which the 6-sided prism pre- dominates. Scalenohedrons, or, Ilemididodecaliedrons (fig. 57).—These are the hemihedral forms of didode- cahedrons, and are obtained by producing, until they meet the pairs of faces adjacent to the second system of alternate terminal edges (fig. 56). The two scalenohedrons, thus derived from each dido- decahedron, have the same ratio of position as the two rhombahedrons derived from the same hex- agonal dodecahedron: if one of them of 60° be made to revolve on its principal axis, it will as- sume the same position as the other. The formulse of the two scalenohedrons derived from the same didodecahedron (a : na : pa : me) are Fig. 57 CRYSTALLOGRAPHY. 35 J(a : na : pa : me) J(a' : na' : pa' : me). § 30. Compound Forms of the Third Sys- tem.—Fig. 58 represents a combination of the primitive dodecahedron r with the first 6-sided prism g. Fig. 59 represents a com- bination of the principal rhombohedron?’,of which the formula is \{a : a : cc a : c), with its first obtuse rhornbo- hedron of which the for- mula is \{a' : a' : oo a : Jc) : this latter rhombohedron t', being the predominating form. Fig. 60 represents a combination of the principal rhombohedron r, £(a : a' : oo a : c), with its first obtuse rhombohedron l(a' : a' : oo a: |c), and with its first acute rhombohedron 2r, J(a' : a' : oo a : 2c), the rhombohedron r predominating. Fig. 61 represents the combination of the primitive rhombohe- dron, : a : oo a : c), Fig. 58. Fig. 59. Fig. 60. with its second acute rhombohedron 4/*., J(a : a : oo a : 4e), the rhombohedron 4r predominating. Fig. 61. Fig. 62. Fig. 63. Fig. 64. 36 INTRODUCTION. Fig. 62 represents a combination of the first 6-sided prism g with a rhombohedron of the second class. Fig. 63 represents a combination of the prin- cipal rhombohedron r with the second 6-sided prism a. Fig. 64 shows a combination of the 6-sided prism g with the terminal face c. Lastly, in fig. 65 is seen a combination of the scalenohedron 3z, predominating, Fig. 65. \{a '.la: la: c\ with the principal rhombohedron r, IV.—FOURTH SYSTEM OF CRYSTALLIZATION. § 31. The forms of the fourth system of crystallization are dis- tinguished by 3 rectangular axes, -which are all unequal and of different kinds : it therefore follows, that the choice’of the princi- pal axis is entirely arbitrary. The relations of length between the 3 axes are indefinite, and in general are irrational. The forms of this system are the following: Simple Forms.—Right Octahedrons with Rhombic Base (fig. 66). The faces of these octahedrons are scalene triangles. The edges are of three kinds: 4 terminal edges D, joining the extremities of the principal with those of the 1st secondary axis; 4 terminal edges F, joining the extremities of the prin- cipal with those of the 2d secondary axes; 4 lateral edges C, joining the extremities of the secondary axes. The angles are of three kinds: 2 terminal angles C, at the extremities of the principal axis; 2 lateral angles A, at the extremities of the 1st secondary axis; 2 lateral angles B, at the extremities of the 2d secondary axis. Sections made through the terminal edges give rhombs (figs. 67 and 68): the same is true of the section passing through the lateral edges and giving the base (fig. 69). When a substance assumes the form of several octahedrons with rhombic base, all these octahedrons maintain simple relations be- tween the length of their axes. Fig. 66. CRYSTALLOGRAPHY. 37 Fig. 67. Fig. 68. Fig. 69. The formula of the primitive form is (a : b : c). The other octahedrons which may result from the same form, will then be expressed by ( a : b : me) ( a : mb : c) (ma : b : c ) (ma : nb : c ), m and n being very simple rational numbers. The first three formulae may be considered as particular cases of the fourth. The number of octahedrons of the fourth system which may be presented in the same substance is therefore still greater than that of the second sytem. But, in reality, this number is very limited, and we rarely meet any octahedrons but those which have for formulas (a : b : c) (a : b : Jc) (a : b : 2c) [a :b : \c) (a : b : 3c), and the extreme forms which are obtained by making m and n = 0, or ad infinitum in our general formulas. By making m or n = 0, we reduce the octahedron to single faces, perpendicular to one of the axes of the crystal. We thus obtain: 1st. A face perpendicular to the principal axis, by making c = 0 : the formula of this face should be, therefore, [ma : nb : 0c): we generally give it the formula (cc a : oo b : c), which supposes that it is derived from the octahedrons (ma : nb : c) having the 38 INTRODUCTION. axis c, but of which the secondary axes have been produced to infinity. 2d. A face perpendicular to the 1st secondary axis, obtained by supposing a = 0: the formula of this face should therefore be (Oa : mb : nc): it has the formula {a : go b : oo c): that is to say, it is supposed to be derived from the octahedrons (a : mb : nc), having the secondary axis a, and the two axes b and c produced indefinitely. 3d. A face perpendicular to the 2d secondary axis, obtained by making b = 0: the formula should be, (ma : Ob : nc): it has the formula (go a : b : oo c), and is supposed to be the extreme face of the octahedrons having the axis b, and of which the axes a and c have been produced ad infinitum. By making m or n equal to infinity in the general formula, we obtain three systems of prisms of which the edges are parallel to each of the three axes: 1st. The first system comprises vertical prisms, of which the faces are parallel to the principal axis: their general formula is (a : mb : go c) : they have the same base as the octahedron from which they are derived. The formula of the vertical prism derived from the primitive octahedron is (a : b : oo c). 2d. The second system comprises horizontal prisms, of which the faces are parallel to the 1st secondary axis, and of which the general formula is (a : oo b : me): the formula of the prism derived from the primitive octahedron is (a : cob : c). 3d. The third system comprises horizontal prisms, of wdiich the faces are parallel to the 2d secondary axis: their general formula will be (oo a : mb : c), and that of the prism derived from the primitive octahedron is (co a : b : c). § 32. Compound Forms.—The principal compound forms of this system are the following : (Fig. 70.) Combination of the principal octahedron o with the more obtuse octahedron |, the terminal face c, and the second horizontal prism / of the principal octahedron. (Fig. 71.) Combination of the principal octahedron o with its vertical prism g, and the vertical prism |. (Fig. 72.) Combination of the principal octahedron o with its first horizontal prism d, and the vertical prism f. CRYSTALLOGRAPHY 39 Fig. 70. Fig. 71. Fig. 72. Fig. 73. (Fig. 73.) Combination of the vertical prism g of the primitive form with the 2d horizontal prism/of the primitive form, and with a second more acute horizontal prism 2f. (Fig. 74.) Combination of the second horizontal prism / of the primitive form, the first horizontal prism f, and the right terminal face c. Fig. 74. Fig. 75. (Fig. 75.) The same combination, with the terminal face pre- dominating. (Fig. 76.) Combination of the 1st vertical prism g of the primi- tive form with the right terminal face c, the terminal face pre- dominating. (Fig. 77.) Combination of the vertical prism g of the primitive form, with the first horizontal prism and the terminal face c. (Fig. 78.) Combination of the principal octahe- dron o with the lateral faces a and b. Fig. 76. Fig. 77. Fig. 78. V.—FIFTH SYSTEM OF CRYSTALLIZATION. § 33. The fifth system of crystallization is characterized by 3 unequal axes, two of which are oblique to each other, and the third 40 INTRODUCTION. is perpendicular to the other two. The relations of length between these three axes are absolutely indefinite, and, in general, irra- tional. Any one of these axes may be chosen as the principal.* Fig. 79 represents an octahedron belong- ing to this system : one of the oblique axes is taken as the principal axis c. The faces are scalene triangles, but they are of two kinds. The edges are of four kinds: 4 terminal edges, joining the axes a and c ; the oppo- site axes alone being equal, on account of the obliquity of the axes; 4 terminal edges which join the axes b and c, and which are equal, because the two axes b and c are perpendicular to each other; lastly, 4 la- teral edges which join the perpendicular axes a and b, and which, consequently, are equal to each other. The section made by the edges D, D' (fig. 80) is a parallelogram comprising the two oblique axes: it is called the principal section. The section made by the lateral edges give the base of the octahedron, which is a rhomb (fig. 81). In order to perfectly define the octahedron, it is no longer enough to give the length of the three axes: we must also as- sign the value of the angle h formed by the oblique axes b and c. The octahedron of the fifth system has not all its faces similar; therefore it is not, strictly speaking, a simple form. It may be considered as a combination of two oblique prisms, of which the first is formed by the faces BAC, CAB', BA'C', and C'A'B', and the second by the faces BCA', CA'B', BAC', and CA'B'. We may distinguish these two prisms by calling the first the anterior oblique prism of the octahedron, and the second, the posterior oblique prism of the octahedron. This distinction is necessary, for it frequently happens that, in the compound forms of this system, the octahedrons do not appear entire, but exhibit only one of their oblique prisms; at other times, one of these prisms predominates greatly over the other. The value of the axes, a, b, c being susceptible of infinite va- riety, as well as the angle 5 of the two oblique axes, it is evident that the fifth system will comprise an infinity of different octahe- Fig. 79. Fig. 80. Fig. 81. * Properly speaking, the choice of a principal axis is limited to one of the two oblique axes, between which the choice is somewhat arbitrary.—J. C. B. CRYSTALLOGRAPHY. 41 drons. But when one and the same substance presents, in its crystalline form, several octahedrons of the fifth system, it will be found that, in all these octahedrons, the angle 8 is the same, and that the lengths of the axes a, b, c of one of these octahedrons, always present commensurable, and, in general, very simple rela- tions with the lengths of the corresponding axes of all the others. So that, if we select one of these octahedrons as a term of com- parison, and assign it the formula (a : b : c), all the octahedrons of the same substance will be comprised in the general formula (ma : nb : pc), the quantities m, n, p, being rational numbers, commensurable, and generally very simple, as 2, 3, 4, or J, J, &c. The forms most frequently met with in this system are the ex- treme forms obtained by successively making m = oo, n — oo, p = oo, or by substituting, successively, m = 0, n = 0, p = 0, in the general formula. By making p— oo, we obtain vertical prisms, parallel to the principal axis c, and of which the general formula is (a : mb : oo c), the formula of the principal prism will be (a : b : oo c). By supposing n = oo, we obtain horizontal prisms, parallel to the 2d secondary axis b, and of which the general formula is (a: cob: me), that of the principal horizontal prism being (a: oo b : c). Lastly, by making m = go, we have oblique prisms parallel to the secondary axis a, and of which the general formula is (go a : b : me), the formula of the principal oblique prism being (oo a : b : c). p = 0 gives a terminal face parallel to the axes a and b, to which we may assign the formula {ma : nb : Oc): but we generally write the formula (oo a : go b : c), Avhich supposes this face to be the ex- treme of the octahedrons (ma : nb : c) having the principal axis c, but of which the secondary axes have been produced to infinity. n = 0 gives a terminal face parallel to the axes a and c, of which the formula would be (ma : Ob : pc): we generally assign to it the formula (oo a : b : oo c): this face is then considered as the extreme of the octahedrons (ma : b : pc) having the secondary b, and of which the axes ma and pc have become infinite. 42 INTRODUCTION. Lastly, m = 0 gives a terminal face parallel to the axes b and c, of which the formula would be (0a : nb : pc); but to which we ordinarily give the formula (a : cob : oo c), because it is supposed to be derived from the octahedrons (a : nb : pc) which have the secondary axis a, and of which the axes nb and pc have been pro- duced ad infinitum. § 34. The following are the most simple compound forms met with in this system: Fig. 82 represents a combination of the perfect primitive octa- hedron o, o' (a : b : c) with the principal vertical prism g (a : b oo c). Fig. 83 is a combination of the perfect principal octahedron o, o' (a : b : c) with the principal vertical prism g (a : b : oo c), and with the terminal faces b (oo a : b : oo c) parallel to the axes a and c. Fig. 82. Fig. 83. Fig. 84. Fig. 85. Fig. 84 presents a combination of the anterior oblique prism o, o' of the principal octahedron (a : b : c) with its vertical prism g (a : b : oo c), and with the terminal face b (go a : b : oo c). Fig. 85 exhibits to us a combination into which enter the pos- terior oblique prism o' of the principal octahedron (a : b : c), its principal vertical prism g (a : b : oo c), and the three systems of terminal faces parallel to the axes, namely, the terminal face b parallel to the axes a and c, of which the formula is (oo a : b : oo c); the terminal face parallel to the axes b and c, and having for a formula (« : oo b : oo c); and, lastly, one oblique face d. § 35. The forms of the sixth system of crystallization have 3 unequal axes, oblique, and bearing to each other indefinite rela- tions : the choice of the principal axis is of no moment. It fol- lows, from the inequality and obliquity of the axes, that the forms of this system have not symmetrical faces, and that the parallel faces alone are similar. Fig. 86 represents an octahedron belonging to this system: the parallel faces alone are equal to each other, so that the faces are of four kinds. The edges are of six kinds: the anterior terminal D is different VI.—SIXTH SYSTEM OF CRYSTALLIZATION. CRYSTALLOGRAPHY. 43 from the posterior edge W: the right ter- minal F is different from the left terminal edge F': the right lateral G is different from the left lateral edge G'. The angles are of three kinds, and formed by unequal edges. Sections made through the terminal or lateral edges are parallelo- grams. When the same substance presents several octahedrons, one is chosen as the primitive form: this takes then the formula (a : b : c); but, in order to define it completely, we must assign the value of the angles a, e, y, which the oblique axes form with each other. If we then determine the axes of the other octahedrons of the same substance, we shall see that these axes always form with each other the same angles a, e, y, and that their absolute lengths present very simple numerical ratios with those of the correspond- ing axes of the primitive octahedron; so that all these octahe- drons may be represented by the formula Fig. 86. (ma : nb : pc), m, n, and p being rational, and, in general, very simple numbers. The octahedrons of this system present four different pairs of parallel faces: they may enter into combinations either by a single pair, or several at a time. It is therefore useful to dis- tinguish each of these pairs by a particular formula. This will be easy, if we preserve the letters a, b, e for the semi-axes on which the positive co-ordinates in analytical geometry are calculated, and, on the contrary, the letters a', b', o' for the portions of the axes directed in the sense of the negative co-ordinates. We can thus represent, The face A B'C and its parallel by (a : b : e) ABC a “ “ (af: b : c) A'B C “ “ “ (af : b' : c) A'B'C “ “ “ (a : V : e). The extreme forms of this system will be obtained by making, successively, p = go , n — oo , m — co , or, p = 0, n — 0, m — 0. We will thus obtain three systems of prisms: Vertical prisms of which the faces are parallel to the principal axis c; Inclined or oblique prisms having their faces parallel to the axis b; 44 INTRODUCTION. Inclined prisms having their faces parallel to the axis 3 alcohol lamp, and open the stopcocks. When we have no gasometer at our disposal, we may make the experiment with a bladder filled with oxygen gas. For this purpose we soak a bladder in water to render it flexible, and fasten to it a metallic stopcock r. In order to fill it with oxygen, we compress it, to expel the air, and then screw the piece r (fig. 134) to a cop per mounting having a stopcock s. This mount- ing is fastened to the upper part of a bell-glasH C, placed in the pneumatic trough and pre viously filled with oxygen. The cocks are open ■ ed and the glass is plunged into the water ol the trough. The oxygen contained in the glasd is necessarily driven out by the water, into thi» bladder. If the latter is not full enough, thi» cocks are closed, the glass is refilled with oxy- gen, which is again passed into the bladder. The piece r is then unscrew- ed, and a tubulure t affixed (fig. 135), which is introduced into the flame, and the jet of oxygen is regulated by the pressure of the arm. § 66. Oxygen is also the essential element in the respiration of animals. An animal perishes in a few moments, if immersed in air previously deprived of its oxygen. Fig. 133. Fig. 134. Fig. 135. 89 Equivalent = 12.50 (or 1). HYDROGEN. § 67. Hydrogen* (from i'Swp, water, and ytw»», I generate) is a gas which, as its name imports, enters into the composition of water. * Water is a compound of oxygen and hydrogen. In the labora- tory, hydrogen gas is always extracted from water. We have obtained oxygen by decomposing, by heat alone, either the oxide of mercury, the peroxide of manganese, or the chlorate of potassa. An analogous process will not succeed with hydrogen. Water cannot he decomposed by heat alone; but the hydrogen may be separated from the water by heating this fluid with substances which absorb its oxygen. Several metals effect this decomposi- tion. Some, such as potassium and sodium, do it when cold: others, as iron and zinc, require an elevated temperature. If we introduce into a bell-glass, a fragment of potassium or sodium, it will be seen to rise toward the top of the glass, by vir- tue of its feeble specific gravity, and an infinity of small bubbles is disengaged from its surface. These bubbles are formed by the hydrogen gas which collects in the upper part of the glass. The metal rapidly disappears by combining with the oxygen of the water: it forms an oxyde which dissolves, and which can be re- covered by evaporating the water contained in the bell-glass. In order to make this experiment accurately, a bell-glass is filled with mercury over the mercurial trough, a small quantity of water is Fig. 136. introduced into the upper part of the glass, and the fragment of potassium, wrapped in tissue-paper to prevent its combination with * Hydrogen gas was obtained toward the close of the 17th century; but it was only in 1766, that Cavendish, a celebrated English philosopher, made known its principal properties. 90 ELEMENTS OF CHEMISTRY. the mercury, is passed in: the potassium rises rapidly through the mercury, until it reaches the water in the bell-glass. § 68. In order to decompose water by means of iron, we ar- range a porcelain tube in a long furnace, called a reverberatory furnace (fig. 136). Into this tube are introduced several bundles of fine iron wire. To one of the ends a of the tube, we affix, by means of a cork and a curved tube, a small balloon filled with water; and to the other end, a discharging tube cd, which con- ducts the gas beneath a bell-glass in the pneumatic cistern. The porcelain tube is slowly heated, in order to prevent its fracture from too sudden an elevation of temperature, until it reaches a red- heat. The water in the balloon is then made to boil. The steam passes over the incandescent iron, which deprives it of its oxygen, and the hydrogen is set free and collected in the bell-glass. § 69. Iron alone, when cold, will not decompose water: it must be heated to a red-heat. This is not the case when a powerful acid, as the sulphuric, is added to the water. The cause of this decomposition is analogous to that which effects the decompo ■ sition of the peroxide of manganese by concentrated sulphuric acid when cold (§ 61). Experience has shown that, when several bodies are in contact, and that, by the interchange of their elements, new compounds may be formed having great affinity with each other, or possessing, under the circumstances in which they arc produced, great fixedness, either isolated or in combination, these new compounds are nearly always formed. We shall, subsequently adduce several examples in confirmation of this proposition. The present experiment is apposite to the subject. The first combina- tion of the iron with oxygen, the protoxide of iron, is a powerful base, having a great affinity for sulphuric acid. Iron alone, whe’i cold, cannot decompose water ; but, in contact with sulphuric acid, its affinity for oxygen is exalted, on account of the affinity of thf acid for the protoxide: the water is then decomposed, and thf resulting oxide of iron combines with the sulphuric acid to fortf, a salt, the sulphate of the protoxide of iron.* The formula of sulphuric acid is S03: that of water is HO, M we shall presently see. The reaction may be then expressed by the following equation: Fe+S03+H0=Fe0,S03+H. The process followed in the laboratory is founded on this reac- tion, but the iron is generally replaced by zinc. Zinc is used, either in the state of the laminated metal found in commerce, and which is cut into small pieces, or in that of granulated zinc. To obtain in the latter form, we melt the metal in an earthen crucible, and * May it not be, that the dilute acid renders the iron more electropositive, and hence decomposes the water with greater facility ?—J C. B. HYDROGEN. 91 pour the liquid into a vessel full of water, which divides it into an infinity of small irregular feathered masses, presenting a large sur- face. The zinc is introduced into a two-mouthed bottle (fig. 13T). Through one we pass a discharging tube, which conducts the gas under a bell-glass full of water, and through the other a tube surmounted by a funnel, which descends nearly to the bottom of the bottle. The bottle is first to be about half filled with water through this tube, and then, through the same way, we introduce small quantities of sulphuric acid. Reaction commences as soon as the acid comes in contact with the zinc : the temperature rises, and hydrogen gas is copiously given off. When the disen- gagement of the gas begins to slacken, Ave add more sulphuric acid. The sulphate of the protoxide of zinc remains in solution in the fiuid, and may be obtained by evaporation. When an apparatus has been used to generate large quantities of hydrogen, it fre- quently happens that the fluid, on cooling, deposits a considerable quantity of this sulphate in a crystallized form. §70. Hydrogen gas is colourless, and, when perfectly pure, is also inodorous. That prepared in the way just described has always a disagreeable, nauseous smell; but this quality arises from the admixture of a very small quantity of foreign substances, which are to be separated, as will be hereafter explained. Hydrogen gas has never yet been liquefied by any degree of pressure, assisted by the lowest temperature hitherto produced. It is the lightest gas known ; its density is 0.0692, that of the air being 1.0000. A litre of this gas weighs, under the normal condi- tions of temperature and pressure, 0sm.0896. Hydrogen gas is, therefore, 14J times lighter than air: its utility in aeronautics is founded on this property.* A balloon made of goldbeater’s-skin, and 2 or 3 decimetres (8 or 12 inches) in diameter, inflated with hydrogen, will rise in the air. A volume of 60 cubic metres of hydrogen gas weighs 5kil.58: an equal volume of atmospheric air weighs, under the same cir- cumstances, 77kil.59.* If, therefore, a balloon of the capacity of 60 cubic metres weighs, with its car and the contents, less than 72kil.21, it will ascend in the air.f Soap-bubbles inflated with hydrogen rise in the air, and take fire if they approach very closely a lighted candle. In order to obtain these bubbles, we fill a bladder, provided with a stopcock, with hydrogen gas, and adapt a small tube to the mounting of the bladder; the extremity of this tube is dipped into soapsuds, and Fig. 137. * 100 cubic inches at 60° F. and 29.92 Bar. weigh 2.162 grains.—J. C. B. -f- 2119 cubic feet (60 cubic meters) of hydrogen weigh 12J lbs. avoird., and the same bulk of air 171£ lbs., so that if a balloon of that capacity, with its car, &c. weighed less than 158| lbs., it would ascend.—J. C. B. 92 ELEMENTS OF CHEMISTRY. removed with the liquid drop which adheres to it; then, by open- ing the stopcock, we obtain soap-bubbles, Avhich separate sponta- neously when they are sufficiently large. §71. Hydrogen gas is eminently combustible: it burns in the air with a feeble flame. If we place above this flame a cold body, water, which is the product of combustion, is deposited. This experiment is made either by approximating a lighted taper to the opening of a bell-glass filled with hydrogen, or by adjusting a delicate curved tube to the mouth of the vessel containing the hydrogen (fig. 138). The gas is allowed to pass over for some time, in order to be sure that no atmospheric air remains in the bottle, and then a lighted taper is brought near to the curved tube: the hydrogen gas inflames and burns with a feeble flame. This apparatus is called the philoso- pher s lamp. A mixture of hydrogen and atmospheric air is ex- plosive. The most powerful possible explosion is a mixture of 2 volumes of hydrogen and 5 volumes of air. This disposition to explode must not be forgotten in making the experiment of the philosopher’s lamp. If the air be not com- pletely expelled from the bottle at the moment of lighting the gas, the flame extends to the explosive mixture contained within, the bottle bursts into a thousand pieces, and the operator runs a risk of being seriously injured. The explosion of a mixture of 2 volumes of hydrogen and 1 volume of oxygen is incomparably more intense than that of a mixture of hydrogen and atmospheric air. The flame of hydrogen gas is not very brilliant, but it produces a great degree of heat. The heat becomes excessively intense when the combustion is assisted by oxygen gas. The experiment is easily made by the gasometer (fig. 139): it is sufficient to place the tube c in the flame of the hydrogen ; this flame then becomes much smaller, because the combustion of the gas takes place in a more confined space. The current of oxygen is increased or diminished by opening or closing the stopcock. The proportion of oxygen is most cor- rect when the flame is reduced to its smallest pos- sible size. The flame of hydrogen, fed by oxygen, produces the highest degree of heat hitherto known : it effects the fusion of substances such as lime, which undergo no change in the most elevated temperature which we can produce in our furnaces. Various kinds of apparatus have been invented to effect the com- bustion of hydrogen by oxygen. Newmann’s blowpipe consists of a reservoir B of thick sheet iron (fig. 140), hooped with iron, upon Fig. 138. Fig. 139. HYDROGEN. 93 which is mounted a forcing pump P, by means of which the explo- sive mixture (2 volumes of hydrogen and 1 volume of oxygen) is introduced, under great pressure, into the reservoir. This pump receives, through the tube t, the gaseous mixture contained in a bell-glass in the pneumatic cistern, or in a gasometer resembling fig. 139. To do this, we begin by making a vacuum in the reser- voir, which can be done with the same pump, merely by changing the action of the valves. When the va- cuum is effected, the pump is made to act as a forcing-pump, and the gaseous mix- ture is forced into the reservoir. This receptacle has a pipe s, terminating in a fine point, and furnished with a cock, at the end of which the mixture is inflamed. In order to prevent the flame from ex- tending into the inside of the reservoir, which would occasion a terrible explosion, the pipe is preceded by a brass tube T, of larger diameter, in which are introduced several layers of metallic washers, which cool the gas and prevent the combustion from extending to the reservoir. Not- withstanding this precaution, the appa- ratus just described has sometimes burst »md caused dreadful injuries. We now prefer keeping the gases separate, and mixing them only at a short distance from the orifice of the pipe: thus all danger of explosion is avoided. For this purpose, two gasometers nre used, one filled with hydrogen, and the other with oxygen. Two tubes, r and s, adapted to the tubes c of these gasometers, con- vey the two gases into a single brass tube L (fig. 141), containing a great many layers of metallic washers, and to which is screwed a pipe terminating in a platinum point. The stopcocks a of the two gasometers are opened, in such a manner as to admit into the gasometer of hydrogen twice as much water as into the gasometer of oxygen. When the burning jet is directed upon a small cylinder of chalk, the lime becomes incandescent, and produces a very brilliant light, which has been called Drummond1s light.* § 72. Hydrogen, being itself combustible, cannot support the combustion of other combustible substances. In order to prove Fig. 140. Fig. 141. * Dr. R. Hare, of Philadelphia, first burned a mixture of the two gases for pro- ducing heat and light. Refer to Encyclopedia of Chemistry, art. Blowpipe. Bags of India-rubber cloth are now used to contain the gases for class-experiments, and the jet, fig. 141, adapted to them.—J. C. B. 94 ELEMENTS OF CHEMISTRY. this, we close, by means of a small plate of glass, the smoothly-ground aperture of a bell-glass, filled with hydrogen, and placed in the pneumatic trough; the glass thus closed is removed without being inverted: and, on the other hand, we fasten a small wax candle to an iron wire, as represented in fig. 142. The glass is partially opened by withdrawing the plate, and the lighted can- dle, being introduced into the bell-glass, is immediately extinguished. § 73. The zinc of commerce is never absolutely pure; it always contains a small quantity of carbon in combination, and sometimes traces of sulphur and arsenic. When this zinc is dissolved in dilute sulphuric acid, a very small portion of the hydrogen com- bines with the carbon, and produces a very fetid, oily substance, which communicates a disagreeable odour to the whole quantity of gas. Arsenic and sulphur also combine with a small quantity of hydrogen. The gas may be entirely freed from these foreign bodies, and may be rendered inodorous by allowing it to remain for some time in contact with caustic potassa, which absorbs the oily matter and the combination of sulphur and hydrogen, and subsequently with the perchloride of mercury, or corrosive subli- mate, which absorbs the combination of arsenic and hydrogen. In order to obtain rea- dily this result, the gas is made to pass through two long tubes, curved in the shape of the letter U (fig. 143), filled with pieces of pumice-stone, those in tube C saturated with a concentrated solu- tion of caustic potassa, p and in D with a solution of the chloride of mercu- ry. The hydrogen leaves this apparatus mixed only with aqueous vapour. We are often required to operate on dry gases. They are col- lected, in that case, not over water, but in a mercurial cistern. These cisterns are generally cut out of marble or some solid stone : the smaller ones are of porcelain or cast-iron. They are made of such a shape as to require the least possible quantity of mer- cury, and still to give, in places, depth sufficient for manipulation. Figs. 144 and 145 represent two vertical sections of mercurial troughs of marble; fig. 144 giving the longitudinal, and fig. 145 the transverse section in the plane xy of fig. 144. The line zu marks the level of the mercury. Fig. 142. Fig. 143. HYDROGEN. 95 The bell-glasses in which the gases are “ then collected must be previously well dried. In order to dry a bell- glass or a bottle, it is heated over some coals, turning it in every direction to give it a uniform tempera- ture : and, in addi- tion, air is constantly driven into the interior with a common bellows, to the nozzle of which a glass tube sufficiently long to penetrate to the bottom of the glass has been affixed. The bell- glass is filled with mercury, and inverted in the mercurial, precisely as in the water cistern. In order to dry the gas collected in the glass, wTe introduce some powerful absorbent of moisture, as a piece of melted chloride of calcium, and let it remain for several hours. At other times, the gas is dried before it is collected, and, for this purpose, is made to pass through a long tube E (fig. 143), filled with pieces of chloride of calcium. Gases may be likewise perfectly dried by means of concentrated sulphuric acid, a substance extremely absorbent of moisture, and which gives off no sensible vapour at the ordinary temperature. Pumice-stone, previously prepared and saturated with this acid, is introduced into one of the curved tubes. As this stone frequently contains small quantities of chloride, which by contact with sul- phuric acid, disengage chlorohydric acid, which would mix with the gaS) it is saturated with sulphuric acid, and calcined in an earthen crucible. The chlorides are thus completely decomposed, and changed into sulphates. §74. The inflammation of the explosive mixture of oxygen and hydrogen, or of hydrogen alone, in contact with the air, is not only effected by a lighted taper or an electric spark; it likewise takes place in a cold, in the presence of certain substances, the principal of which is platinum sponge.* If we throw a piece of platinum sponge into an eprouvette containing a mixture of 2 parts of hydrogen and 1 of oxygen, an explosion will instantly ensue. If we project a jet of hydrogen in the air, upon the sponge, this substance becomes incandescent and the gas inflames. The action of the sponge, in this case, has not as yet been clearly explained: but advantage has been taken of it to construct a machine for obtaining fire instantly by means of hydrogen gas. Fig. 144. Fig. 145. * The name of platinum sponge is given to the spongy mass of metallic platinum obtained by decomposing certain combinations of platinum by heat. 96 ELEMENTS OF CHEMISTRY. § 75. We are acquainted with two combinations of hydrogen and oxygen. The first combination, the protoxide, is simply water.* COMBINATIONS OF HYDROGEN AND OXYGEN. PROTOXIDE OF HYDROGEN, OR WATER, HO. §76. We have seen (§71) that hydrogen, burning in the air, produces water ; but in order to make the experiment conclusive, the gas must previously be perfectly dried, for without this pre- caution it might be alleged that the Avater arose from the moist gases, and from the solution of which the temperature always rises during the reaction. The apparatus is then arranged as in fig. 146. By holding a tubulated bell-glass slightly inclined, above the flame, the water formed by the combustion trickles down the sides of the glass, and may he collected in a saucer, in any quan • tity we may desire. §77. Pure Avater is tasteless and inodorous: under slight pres- sure it is colourless, but when this pressure is greatly increased, i \ assumes a very decided greenish tinge. Water becomes solid in the intense cold of winter. The tern perature at which this change takes place is marked 32° on Fahrer heit’s thermometer, f If a vessel filled with broken ice or snow b o placed in a warm room, the ice soon melts, and, as soon as the fusion, has commenced, a thermometer in the \Tessel marks constantly tbi* same temperature, until the last portions of ice have disappeared. This constant temperature has been selected as one of the fixed points of the thermometer. Water may, hoAvever, be reduced be- low zero A\Tithout becoming solid, which occurs when it is allowed to get cold in a vessel perfectly quiescent. It has been knoAvn to Fig. 146. * Water was considered by the ancients as one of the four elements of nature. It was only toward the close of the eighteenth century that it was ascertained to be composed of hydrogen and oxygen. Priestley first observed that, when hydrogen is burned in an earthen vessel, at the expense of atmospheric air, or of oxy- gen, a certain quantity of water is deposited on the sides of the vessel. But the composition of water has only been incontestably proved by the almost simulta- neous discoveries of Watt, Cavendish, and Lavoisier. f It is the zero, 0°, of the centigrade and Reaumur’s thermometers. HYDROGEN. 97 descend to 10.4° F. without congealing: but if the vessel he agitated, or, better still, if a foreign body be introduced therein, icicles immediately form, and the temperature ascends to 32°, and preserves this degree until the whole of the water is solidified. A similar phenomenon occurs in the fusion of all substances. The transformation of fluid water into ice is, therefore, an actual crystallization, which, however, but rarely gives rise to appreciable crystals : they are aciculge dovetailed into each other and producing a continuous transparent mass. Crystalline forms may, however, be sometimes recognised in the small icicles which form in muddy water. When the temperature of the air is below 32°, the moist- ure is precipitated in the form of snow or hoar-frost. Each flake of snow is the union of many crystals. With a good lens, the elementary crystals, which are regular 6-sided, elon- gated prisms, may be re- cognised, grouped in stars around a centre, so as always to form angles of 60° and 120°. Nos. 2, 3, 4, 5, 6, 7, 8 of fig. 147, represent some of the most simple of these groupings. Hoar-frost frequently appears in less complicated forms, and we sometimes find perfectly regular hexa- hedral spangles (No. 1). The crystalline form of ice belongs, therefore, to the hexagonal system. Water increases in volume by congelation, so that the density of fluid is greater than that of solid water. Fluids and solids in- crease in volume, expand, by raising their temperature: water is an exception, for a few degrees above 32°. Between 32° and 39J°, water, instead of expanding, contracts; about 39J°, it pre- sents a minimum of volume, and, consequently, a maximum of density. Above 39J°, to as high a degree as has been observed, it expands regularly. It has been agreed to assume as unity the density of water at 39J°, and compare with it that of other solids or fluids.* The density of ice is therefore represented by 0.94. The force with which water expands in congelation is irresistible, and sufficient to burst the thickest bomb-shell. Very tenacious but porous stones frequently split during the winter, when the water contained in their pores is frozen. §78. Water readily assumes the gaseous state: the tempera- ture at which this change takes place depends on the pressure of the atmosphere. The second fixed point of Fahrenheit’s ther- mometer is marked 212° (100° of the centigrade), the tempera- Fig. 147. * In England and the United States, we generally assume water at 60° as the unit of density.—J. C. B. 98 ELEMENTS OF CHEMISTRY. ture at which water boils under the pressure of 29.92 inches (760mm) of mercury. The temperature at which this ebullition takes place diminishes writh pressure; thus, in the vacuum of an air-pump, water will boil under a superstratum of ice. Water assumes the aeriform state when the temperature is over 212°, and the pressure less than 29.92 bar. We shall subsequently see how we may ascertain experimentally the weight of a certain volume of this vapour, and compare it with an equal volume of at- mospheric air at the same temperature and under the same pres- sure : this relation is called the density of the vapour of water (steam). If we ascertain its numeric value for temperatures above 212°, and successively increasing, we shall find that, from about 266°, this relation remains sensibly constant for all the higher temperatures, and that it is represented by the fraction 0.622. We shall therefore adopt this value for the density of the vapour of water. We shall hereafter define, in the same manner, the densi- ties of other vapours. Water gives off an appreciable vapour into the air: the formation of this vapour is the more abundant as the air is drier, that is, is less saturated with the vapour of water, and the temperature is higher. It is then said that water evaporates in the air. The air always contains a certain quantity of the vapour of water. It is very near its point of saturation in rainy w'eather and in winter; and, on the contrary, very far from it during the hot days of summer. Certain substances possess the property of absorbing the water of the air, even when it is not saturated, and of dissolving in this water. These substances are said to be deli- quescent, such as chloride of calcium, potassa, etc. On the con- trary, other substances containing water, readily part wTith a portion of it to the surrounding air, if the latter is not saturated, and fall into powder: these are called efflorescent. Sulphate of soda belongs to this list. It is evident that no substance is efflo- rescent, in an atmosphere saturated with moisture, and that all soluble bodies are, on the contrary, deliquescent. It sometimes happens, however, that substances effloresce by ab- sorbing the moisture of the air. This occurs in crystallized or melted substances, which have an affinity for water, and form with it new deliquescent combinations. Melted sulphate of soda, ex- posed to a damp atmosphere, absorbs water and falls into powder. §79. The most limpid river and spring water is not pure: this can be always ascertained by evaporation, a residuum remain- ing in the vessel. Ham-water is nearly pure, but, as it generally falls on roofs before being collected, it always dissolves a small quantity of foreign bodies. Water is purified by distillation. As large quantities of distilled water are used in the laboratory, this process is conducted on a large scale, in an apparatus called a still. The still (fig. 148) is composed of a copper kettle A, set in HYDROGEN. 99 Fig. 148. a brick furnace, and to which is fitted a head or capital B, ter- minated in a curved pipe bed, which leads to a worm, contained in a large cylinder of metal or wood, jpqrj, kept filled with water.* The extremity of the worm terminates at a outside of the cylinder. The water to be distilled is introduced into the kettle at t. As the water in the cylinder, which serves as a refrigerator, is neces- sarily heated by the condensation of the vapour in the worm, it must be occasionally renewed. This is most easily effected by means of a reservoir containing cold water, which is slowly carried by the outer tube TT' to the lower part of the cylinder. In this way, the cold water is always at the bottom, and the heated water escapes by the tube o at the upper part. The supply of cold Fig. 149. * The head, pipe, and worm should be of solid tin.—J. C. B. 100 ELEMENTS OF CHEMISTRY. water may be so regulated that the heated water shall escape at nearly the boiling point, and be reconducted into the kettle, thus avoiding unnecessary waste. We are sometimes required, in the laboratory, to distil very volatile fluids, the vapour of which must be refrigerated, so as to occasion no loss. We then use one of the apparatuses (figs. 149 and 150). In fig. 149, the balloon A contains the fluid to be distilled. The glass tube abc serves as a worm: it is held in a tin tube or cooler DE, by corks which should close water-tight: the end of the tube enters a bottle to receive the distilled liquid. The cooler has a funnel at d through which cold water enters, and at / a curved tube by which the heated water escapes. In fig. 150, the fluid to be distilled is introduced into a retort, the neck of which enters a larger tube luted to the refrigerator. When we have but a small quantity of liquid to distil, we may contrive an apparatus, merely with glass tubes and corks, as re- presented in fig. 151. §80. Water dissolves a great number of solid and liquid substances, which, generally speak- ing, dissolve in greater quantities as the tem- perature is more ele- vated ; so that if a hot saturated solution of these substances be al- lowed to cool, a portion of the substance crystallizes. To obtain the rest of the solution, the water must be evaporated. This is done by placing the solu- tion in a porcelain saucer over hot coals, or, better still, an alcohol Fig. 150. Fig. 151. HYDROGEN. 101 lamp. The operation requires care when we desire not to lose the slightest quantity of the mat- ter in solution, as is the case in chemical analyses. The so- lution should not be heated to ebullition, because the bub- bles of vapour which form on the heated bottom of the saucer burst on the surface, and infallibly throw out from the saucer small quantities of the solution. Fluids are frequently evaporated in a rvater-bath (fig. 152): the porcelain saucer containing the solution to be evaporated, is placed upon another larger copper one partly filled with water and heated by an alcohol lamp. Sometimes, we do not put any water into the copper vessel: the porcelain saucer is then heated in an air-bath which causes a very regular evapora- tion. Lastly, in laboratories where there are many solutions to evaporate at once, all the saucers are put in the same sand-bath, heated by wood or coal. It frequently happens that evaporation is required to be per- formed slowly and at a low temperature. The saucer is then placed on a larger glass capsule, containing con- centrated sulphuric acid, and the whole is covered with a bell-glass (fig. 153). The sulphuric acid absorbs the moisture of the air as fast as it is ab- stracted from the solution. The evaporation is more rapid if the saucers are placed under the receiver of an air-pump from which the air is ex- hausted. § 81. Water dissolves gases equally. The so- lubility of the same gas in water increases with the diminution of temperature and pressure ex- erted on the solution by the undissolved portion of the gas. When a certain volume of wTater is surrounded by a limited atmosphere of gas, the water dissolves such a portion of it, that this portion of gas, occupying a volume equal to that of the fluid, possesses an elastic force which is the same constant fraction of the pressure exerted by the undissolved gas on the solution. This fraction is entirely independent of the absolute value of the pressure: we shall suppose it to be ~ for nitrogen, and for oxygen. Thus, wlien 1 litre of water is surrounded by an unli- mited atmosphere of oxygen, it dissolves a portion of the gas such, that this gas, occupying the volume of 1 litre, will have the density proper to it under a pressure of h; li representing the Fig. 152. Fig. 153. 102 ELEMENTS OF CHEMISTRY. pressure of the undissolved oxygen gas on the liquid. If, in a second experiment, the pressure of the undissolved gas is 4 5 the litre of dissolved oxygen will have the density proper to it under the pressure of ,7. j-. Hence its absolute weight will be 5 times smaller in the second case than in the first. Where water is in contact with an atmosphere formed by the mixture of two or more gases, it dissolves of each a quantity pre- cisely equal to that which it would have dissolved if in contact with an atmosphere of this gas alone, exerting a pressure equal to the fraction of the total pressure, proper to it in the gaseous mixture. Thus water, in contact with the air, dissolves a quantity of nitrogen equal to that it would dissolve if in contact with an atmosphere consisting wholly of this gas, exerting a pressure equal to | of that of the atmosphere, that is, 4*t, and a quantity of oxygen equal to that which it would dissolve if in contact with an atmosphere of pure oxygen exerting a pressure 5 times less than that of the atmosphere. Consequently, 1 litre of water dissolves, on contact with the air, 1 litre of oxygen with the density proper to it under a pressure of - h: and 1 litre of nitrogen with the density proper to it under a pressure of 4 4 h. If we wish to bring these gases under the ordinary pressure of the atmosphere, it must he remembered that the volumes of the gases are inversely as the pressure exerted upon them. Consequently, 1 litre of water dissolves, on contact with the air, a fraction 4 • 5 of a litre of oxygen: and a fraction 4 4 °f a litre °f nitrogen, and, therefore, a total volume of gas represented by The whole volume of the gas dissolved may be easily ascertained by means of the following experi- ment. Fill a glass balloon (fig. 154) entirely with water; fill also the discharging tube; then insert the stopper a into the neck of the bal- loon : the displaced water escapes by the tube, and the apparatus is perfectly filled with Avater. Pass the curved end of the tube under a bell-glass filled Avith mercury, and apply heat to the balloon. When the temperature of the water approaches 40° or 50°, little bubbles are seen to disengage from the sides of the balloon. If the A\-ater be made to boil, for a feAV minutes, the vapour drives the dis- engaged air into the bell-glass. We measure the volume of air thus collected, and compare it to that of the Avater from Avhich it arose. § 82. Water combines with a great number of substances. With powerful acids it acts the part of a feeble base; and, on the con- trary, that of a feeble acid with poAverful bases. Fig. 154. HYDROGEN. 103 Water combines with a great number of salts when these are crystallized in their aqueous solutions; the same salt frequently combines with very different proportions of water, according to the temperature at which crystallization takes place. § 83. Analysis of Water.—Let us now ascertain the relative proportions in which hydrogen and oxygen combine to form water. To do this, we introduce into the same bell-glass, over mercury, the well-known volumes of these gasses, and apply heat to the mixture. The two gases combine in determinate proportions and form water, which condenses on the sides of the bell-glass. As that gas which is in excess does not entirely disappear, we mea- sure the remainder and ascertain the volumes of the two gases which have combined. To perform this experiment, we must procure bell-glasses marked into equal divisions and intended to measure gases. Such glasses may be bought, but it is better to divide them our- selves, when wre wish to be very exact. We select a bell of very pure glass, of 1 or 2 centimetres to f inch) diameter internally, and 2 or 3 decimetres (8 to 12 inches) in length. It is placed verti- cally, the closed end downward. We then make a measure or gauge A (fig. 155) with a piece of closed tube, the edges of which are very accurately ground, so that the mouth of the tube may be exactly closed with the small plate of polished glass B. The gauge is then over-filled with mercu- ry, the excess of which is driven off by the glass plate, or obturator, applied to its mouth. This measure is poured into the bell-glass to be divided, and the air-bubbles which adhere to its sides are carefully removed. This being done, a mark is made on the glass at the level of the mercury, a second measure poured in, another mark made, and so on. It is evident that the spaces between the marks on the glass correspond to equal capacities: and if the bell-glass be not too irregular, it may be granted that they preserve the same diameter in each of the spaces. These spaces may be made to vary in size, according to that of the gauge. After this preliminary gauging of the bell-glass, it is to be emptied, and covered with a thin coat of the common liquid var- nish of the copper engravers. The bell-glass is then placed in a dividing machine, and, by means of an iron point, we mark, on the coat of varnish, divisions arranged in such a manner that each space between two consecutive marks of the gauging shall con- tain the same number of equal divisions. A larger mark is made at each fifth division, to facilitate its reading, and figures are marked at every tenth. The divisions are then painted with hydro- fluoric acid in solution. This acid possesses the property of dis- solving glass: and consequently attacks the surface of the bell- glass from which the varnish has been removed, and leaves the divisions engraved thereon. Fig. 155. 104 ELEMENTS OF CHEMISTRY, When we purchase graduated bell- glasses, they should be verified, if re- quired for exact experiments. This is easily done by means of small gauges resembling those with which we have just measured the bell-glass, or by pouring into the glass quantities of mercury which have been accurately weighed. It is evident that, if the glass be correctly graduated, the vo- lumes occupied by the quantities of mercury should be proportional to their weight. The bell-glass being accurately gra- duated, we introduce a certain volume of hydrogen, exactly weighed, taking care to plunge the glass into the mer- cury until the level of this metal is the same within and without. It is more convenient to make this measure- ment as in fig. 156, in a glass eprouvette, in which we can more readily level the mercury and read the division. A certain quantity of oxygen gas is then introduced into the same bell-glass. The mixture is then introduced into an apparatus called a eudiometer, and which is so arranged that an electric spark can be passed into its interior. The eudiometer (fig. 157) is composed of very thick glass, having at its upper part an iron mounting a passing through the thickness of the tube, and hermetically cemented to the opening. On the side of the glass, at b, a second hole is bored, in which is cemented a strong iron wire, which pene- trates nearly to the upper mounting. The outer extremity of this wire is hooked. The eudiometer, filled with mercury, is inverted over the mercurial trough and the mixture of the gases introduced. The surface of the glass is rubbed several times with a hot rag. The mixture is then exploded by means of a galvanic battery applied at a and b. At the moment of combustion, a great quantity of heat is disen- gaged, producing considerable dilatation of the gases, and there- fore the gaseous mixture should only half fill the eudiometer, else a part of it would infallibly be projected from the tube. This loss is prevented by closing the opening of the eudiometer with a valved stopper A. At the moment of explosion, there is an increase of elastic force in the apparatus, and the dish i is closely applied to the surface of the stopper, so that nothing can escape. As soon as the heat is dissipated, which soon occurs, the water condenses ,Fig. 156. Fig. 157 HYDROGEN. 105 in liquid drops on the sides of the eudiometer, and then occupies a volume 2000 times less than that of the gases from which it sprang. The tension in the apparatus is then lessened, the valve i rises, and the mercury enters the eudiometer. If the gases disappear entirely, it is a proof that the quantities introduced were exactly in the proportion proper to form water: wThich happens if we have introduced exactly 1 volume of oxygen and 2 volumes of hydrogen. In general, one of these gases is in excess, the gaseous residuum is then passed into the graduated bell- glass, measured, and its return ascertained by the application of a lighted taper; if it burns, the residuum is hydrogen. Suppose that we introduce into the eudiometer 100 measures of hydrogen, 75 measures of oxygen : we shall find, after combustion, 25 of oxygen. Therefore, 100 of hydrogen have combined with 50 of oxygen, or 2 volumes of hy- drogen and 1 of oxygen. The same experiment may be made over the water-trough, but then we cannot ascertain the nature of the product of combustion. When we use the water-trough, the mountings of the eudiometer should be of brass. There is also a water-eudi- ometer (fig. 158) which is readily used. It is composed of a thick glass cylinder AB, intended to contain the mixture: this cylinder is fitted below into a brass mounting BC, having a stop- cock at S. A funnel C allows the introduction of the gas. The glass cylinder communicates above with a second funnel D, into which we put water. A stopcock R permits or cuts off the communication. A graduated glass tube EF is screwed to the bottom of the cup D. Lastly, at v, the metallic mounting A is perforated by a hole, into which a glass tube has been cemented, traversed by the metallic rod t, which is thus isolated from the metallic mounting and nearly approximated to it internally. The use of the apparatus is easily understood: when we open the cocks R and S, and plunge the eudiometer into the water-cistern above the cup D, it fills with water: we close the cock R and raise the eudiometer. Measure in the gra- - duated tube EF the hydrogen and oxygen, and introduce the mixture into the eudiometer through the funnel C. The mixture is then exploded as before, by means of the button t, and at the moment of explosion the cock S must be closed to prevent the escape of the gas. Eig. 158. 106 ELEMENTS OF CHEMISTRY. The remaining gas is now to be measured: this is easily done in the graduated tube EF. To do this, we fill the tube with wa- ter, and invert it in the cup D, where it is screwed. By opening the cock R, the gas will pass into the tube EF. To measure it, we unscrew the tube, and plunge it into the water-cistern, so as to establish a coincidence of level externally and internally. The greatest difficulty of eudiometric analyses performed over mercury is owing to the extravasation of the gases: but it may be avoided by using eudiometers divided into equal parts. In order to construct such eudiometers, we select a glass tube closed at one end, of above 10 or 15 millimetres (J to \ inch) in diameter internally, and 1 or 2 millimetres (.04 to .08 inch) in thickness. We bore two small holes a and b in the tube (fig. 159) by means of a small steel drill in a turning-lathe, keeping the spot moist- ened with turpentine. We can thus pierce the tube without any risk of breaking it. We then cement to these holes pieces of platinum wire, which are brought nearly to touch each other inside of the glass. We may also solder the pla- tinum wire to the glass, by means of an enameller’s lamp. This is preferable when the walls of the glass are not very thick. The tube is then divided into equal divisions. The walls of this eudiometer being generally thinner than those of the ordi- nary eudiometer, it is prudent not to hold it in the hand at the moment of explosion, but to fix it in a Support (fig. 160). We may also use another ar- rangement, which has the ad- vantage of requiring only a small quantity of mercury. The eudi- ometer has the shape of a tube curved like the letter U. In order to fill it with mercury, it is made to assume the position of fig. 161, and then that of fig. 162: the closed leg A remains filled with mercury. The two gases are introduced by passing into the open leg B the discharging tubes of the gasometers, and causing them to ascend into the leg A. The volumes of the gases intro- duced, and of the residuum after combustion, are measured in the eudiometer itself, by carefully bringing the Fig. 159. Fig. 160. Fig. 161. Fig. 162. HYDROGEN, 107 mercury to the same level in both legs, which is readily done by abstracting or adding the metal with a pipette. We shall describe hereafter, in a chapter devoted to the analysis of gaseous mixtures, an eudiometric apparatus more perfect than those just explained, and which furnishes us with very exact results. § 84. Water therefore results from the combination of 2 volumes of hydrogen and 1 of oxygen: hence, we can easily deduce the composition of wrater in weight, since we know the densities of these two gases. In fact, 1 volume of air weighing 1.0000, 1 volume of oxygen weighs 1.1056 2 “ hydrogen “ 2 x0.0692 = 0.1384 The water produced weighs 1.2440 In order to obtain the quantity of hydrogen and oxygen which forms 100 grammes of water, we make the proportions 1.2440 : 1.1056 : : 100 : x, whence x = 88.87. 1.2440 : 0.1384 : : 100 : y, whence y = 11.13 ; therefore, 100 parts of water contain 11.13 hydrogen, 88.87 oxygen, moo When 2 volumes of hydrogen combine with 1 volume of oxygen, what is the value of the vapour of water resulting from the com- bination ? If the 2 volumes of hydrogen, combining with 1 of oxygen, formed only a single volume of vapour of water, the den- sity of this vapour would be 1.244. But direct experiment has given, for this density, a value one-half less, that is, 0.622: therefore, 2 volumes of hydrogen combining with 1 of oxygen, have produced, not 1, but 2 volumes of vapour of water. § 85. We cannot avoid calling the attention of the student to the simplicity of the relations presented by the volumes of the two combining gases, and the vapour of water resulting from their combination, instead of the complicated and infinitely variable relations which might have occurred. This is not a fortuitous circumstance, peculiar to the case under consideration. We shall also recognise very simple relations in the combinations of the other elementary gases. The study of such combinations has discovered this law1 of nature: When two elementary gases comf bine, their volumes have to each ether very simple numerical ratios, and the volume of the resulting compound, considered in the gaseous state, bears also a very simple ratio to the sum of the volumes of the gases which entered into the combination. * Discovered by M. Gay-Lussac. 108 ELEMENTS OF CHEMISTRY. § 86. Another method, still more exact than the eudiometer, has been employed to determine directly the weight of hydrogen and oxygen which combine to form water. Several metallic oxides, heated in a current of hydrogen gas, give off their oxygen, and are reduced to a metallic state. This oxygen, combining with the hydrogen, forms water which can be weighed. The loss in weight of the metallic oxide, gives the weight of the oxygen entering into the water. The difference between the two weights gives the hydrogen. It is necessary to use in this experiment pure and perfectly dry hydrogen: it is prepared by means of the apparatus described (§ 73), and represented by ABODE (fig. 163). The oxide of cop- per is introduced into a strong glass balloon F, with two necks. This balloon communicates with another G, intended to collect the Fig. 163. greater part of the water formed in the experiment: it is suc- ceeded by a tube II, filled with pumice-stone soaked in concen- trated sulphuric acid, and which absorbs the last portions of water. Before making the experiment, we weigh with the greatest nicety the balloon F, empty and very dry; then the same balloon with the oxide of copper perfectly dried. The difference between the two weights gives that of the contained oxide. The balloon G and the tube H are also weighed. The apparatus being arranged, the hydrogen gas is slowly generated, and continued for a long time, in order completely to drive the air out of the apparatus. When it is completely filled with hydrogen, the balloon F is heated by an alcohol lamp. The combustion of the hydrogen with the oxygen of the oxide of copper soon commences, and the water trickles down the sides of the balloon G: the last particles of water formed condense in the tube H, which the hydrogen in HYDROGEN. 109 excess must traverse before passing out into the air. The experi- ment is continued until the oxide of copper is completely reduced to the state of metallic copper. The balloon G is then allowed to cool, in the midst of the current of hydrogen; then the portion of the apparatus to the left is separated from the caoutchouc a. The balloons GF and the tube II are then filled with hydro- gen, and if weighed in this state, the difference between their weights before and after the experiment would depend, not only on the substances which they have condensed during the reaction, but also on the excess of weight of the air which originally filled the apparatus, over the hydrogen which has replaced it. The apparatus must therefore be restored to its primitive condition, and again be filled with atmospheric air. For this purpose, we secure, by means of caoutchouc, the extremity / of the tube II (fig. 163) to the tube s of fig. 164. This tube communicates with the upper part of the aspirator V, filled with water. At I is a tube filled with pumice-stone, which prevents the vapour of the water in the jar V from penetrating into the tube H, and increasing its w'eight. By opening the stopcock r, the water flows out, and is re- placed by air which enters at a (fig. 163), is deprived of its moisture in the tube E filled with pumice-stone soaked in sul- phuric acid, traverses the apparatus FGH, and drives out the hydrogen from it. If we maintain a nearly regu- lar current of air, it will be sufficient to cause the tube to descend into the water to a certain distance above the level whence the water flows ; the jar then acts the part of a Mariotte’s jar, and the discharge is nearly regular, so long as the level of the water does not reach the end of/he tube. We weigh separately, first the balloon F, then the receiver G, with the tube H. The difference between the weight of the balloon F, containing the oxide of copper, before the expe- riment, and its weight when containing the metallic copper, gives the weight of the oxygen in the water. The increased weight of the receiver G and the tube H gives the weight of the water formed. The most exact experiments made in this way, have shown that 100 parts of water contain, Fig. 164. Hydrogen 11.11. Oxygen 88.89. 100.00. § 87. In the experiment just described, as well as that perform- ed in the eudiometer, we ascertain the composition of water by 110 ELEMENTS OF CHEMISTRY. finding the volumes or weights of the separate elements which enter into it: we thus make what is called the synthesis of water. But we frequently ascertain the composition of compound bodies by an inverse method. These bodies are decomposed, so as to ascertain the weight of their elements, either by really isolating these elements, or uniting them in combinations of which the com- position is known. This process is called analysis. We have described (§ 68) an experiment by which water is de- composed, by passing its vapour through a porcelain tube heated to redness and containing metallic iron. If, in this experiment, wre measure the volume of hydrogen gas disengaged, and from this measure deduce the weight of this gas : if, on the other hand, we ascertain the weight of the oxygen which combined with the iron, by weighing the latter before and after the experiment,—we shall have obtained by analysis the composition of water. But this experiment is not sufficiently exact. The composition of water may be determined exactly, by ana- lysis, by means of the voltaic pile. If we plunge the two poles of a buttery terminating in platinum wire into water slightly acidu- lated with sulphuric acid, we shall see small bubbles of gas along each wire. These gases may be collected in separate bell-glasses, and we shall find that the gas disengaged at the positive pole is oxygen, and that collected at the negative pole is hydrogen, and that the volume of the latter is precisely double that of the oxygen. The experiment is generally performed in the apparatus repre- sented in fig. 165. The bottom of a wine-glass is pierced with two very small holes, through which the platinum wires are passed. To close them completely some melted mastic is poured into the glass. The glass is filled with acidu- lated water, and a small graduated bell-glass is placed over each wire. In order to effect the decomposition of the water, it will be enough to bring the platinum wires in communi- cation with the two wires of the battery. The addition of a small quantity of sulphuric acid renders the water a better conductor of electri- city, and consequently facilitates its decompo- sition. The synthetic or analytic method is used for ascertaining the composition of bodies, accord- ing as one or the other mode appears more ap- plicable to the case. § 88. We frequently express the composition of water in another manner. Instead of inquiring how much hydrogen and oxygen are in 100 parts of water, we ask how much hydrogen is required to form water with 8 parts of oxygen, and say, Fig. 165. HYDROGEN. 111 8 of oxygen combine with 1 of hydrogen and form 9 of water. The quantities 8 of oxygen and 1 of hydrogen are called equiva- lent quantities, or chemical equivalents; and we have agreed to assign as the equivalent of water the number 9, which is the quantity of Tyater containing the quantities 8 of oxygen and 1 of hydrogen. In the same manner, if we consider these bodies in the gaseous state, 1 volume of oxygen is equivalent to 2 volumes of hydrogen in the formation of water; and we say that the equivalent of oxygen in volume is 1 volume, and the equivalent of hydrogen is 2 volumes. From the above definition, the equivalent of the vapour of water is therefore 2 volumes, since it requires 2 vo- lumes of vapour of water to give 1 of oxygen and 2 of hydrogen. We shall adopt the letter 0 to express the equivalent of oxygen, that is the weight 8 of oxygen, and the letter H to express the equivalent of hydrogen, or its weight 1. The equivalent of water, that is the weight 9 of water, will be represented by. HO. Thus, the characters, II, 0, and HO recall not only the nature of the bodies they represent (§ 54), but also the determinate weight of those bodies, or their equivalents. Lastly, the composition of water is expressed in another manner which deserves to be mentioned, because it is used by many chemists. It is admitted that bodies are formed of molecules, indivisible by mechanical means, and which are called atoms. Let us sup- pose that, when two bodies combine, an atom of one of these bodies unites to 1, 2, 3, 4, 5... atoms of the second, or 2 atoms of the first with 3, 5, 7, 9... of the second. The law of the combination of gases according to simple proportions, a law demonstrated by experiment, will merely be a consequence of the preceding hypo- theses, if we admit that the number of atoms contained in equal volumes of the different gases bear to each simple proportions. Let us advance the most plain hypothesis, and admit that equal volumes of all the elementary gases contain the same number of atoms. Experiment has shown that 1 volume of oxygen combines with 2 of hydrogen to form water: we can therefore say that 1 atom of oxygen combines with 2 of hydrogen to form 1 atom of water. But the proportions between the ponderable quantities of oxygen, hydrogen, and water, as ascertained by experiment are as the numbers 8:1:9; we may therefore say that the propor- tions between the weights of the atom of oxygen, the atom of hydrogen, and the atom of water are those of the numbers 8 : J : 9, or even, absolutely, that the weight of the atom of oxygen, or the atomic weight of oxygen is 8 hydrogen \ water 9 112 ELEMENTS OF CHEMISTRY. If we adopt the characters II and 0 to represent the atomic weight of hydrogen and oxygen, it is evident that the atomic formula of water will be II20. The double atom of hydrogen is often represented by the cha- racter H. The formula of water is then HO. Many chemists represent the atoms of oxygen by an equal number of points placed above the character which expresses the substance com- bined with the oxygen : thus, they write water'll. We shall exclusively adopt the notation of equivalents in the present work. BINOXIDE OF HYDROGEN, HO,. § 89. Hydrogen can combine with a quantity of oxygen greater than that necessary to form water. The second combination has received the name of binoxide of hydrogen, or oxygenated water} We have seen (§64) that by heating the peroxide of manganese with concentrated sulphuric acid, the peroxide is brought to a state of protoxide, which combines with the sulphuric acid, and the oxygen is given off. Other peroxides undergo similar decomposi- tion, when cold, and in contact with dilute acids: but then the oxygen which is freed is not given off, but remains in combination with the water: this is the case with the peroxides of barium, strontium, and potassium. The peroxide of barium is used for the preparation of oxygenated water. This peroxide is rubbed with water in a porcelain mortar, so as to form a liquid paste: and this paste is gradually added to a mixture of 1 part of ordinary chlorohydric acid, and 3 parts of water, contained in a porcelain capsule, and constantly stirred with a glass rod. The peroxide of barium dissolves Avithout the disengagement of any gas: chlo- ride of barium, water, and oxygen, Avhich remains in combination with the water, are formed. ( Oxygen.. Barium Oxygen. Binoxide of hydrogen. Binoxide of Barium. Protoxide of barium or baryta " "Water. Chloride of barium. Chlorohydric acid Hydrogen. t Chlorine ... The substances brought into contact are the binoxide of barium, of which the formula is Ba02, and chlorohydric acid, which we write HC1. The water of the hydrate of the binoxide of barium is separated in combination with one-half of the oxygen of the binoxide, and consequently in the state of binoxide of hydrogen, which dissolves in the surrounding water. The products of the reaction are the chloride of barium BaCl, and the binoxide of hy- drogen which, as we shall presently see, we should write H02. We may therefore express the reaction by the following equation: Ba02 + HC1 = BaCl + II02. 1 Also called peroxide of hydrogen, and discovered by Thenard in 1818. HYDROGEN. 113 When the chlorohydric acid is nearly saturated by the baryta, we pour into the solution sulphuric acid, which precipitates the barium in the state of insoluble sulphate of baryta, and the chloro- hydric acid is again formed in the liquid. Chloride of barium.. Chlorine. Barium.. Water... f Oxygen... t Hydrogen.. >Baryta.. Chlorohydric acid. Sulphate of baryta. Sulphuric acid BaCI + HO + S03 = BaO,S03 + HC1. Toward the close, the sulphuric acid is added dropwise, in order not to be in excess. The sulphate of baryta is separated by a fine filter, and we obtain a liquid identical with the original acid liquid, except that it contains a certain quantity of binoxide of hydrogen. We can treat this fluid like the original acid fluid, and dissolve it in an additional quantity of binoxide of barium, until it is saturated with chlorohydric acid, and then again precipitate the baryta by sulphuric acid. After this second operation, the acid solution contains twice as much binoxide of hydrogen as after the first. When these operations have been repeated frequently enough, we obtain a liquid well charged with binoxide of hydrogen, but which contains chlorohydric acid, of which it must be freed. To do this, we add, gradually, some sulphate of silver. Chloride of silver, which precipitates, is formed; and sulphuric acid, which is dis- solved in the liquid. Chlorohydric acid... Chlorine... Hydrogen. Oxygen. Water. Chloride of silver. Sulphate of silver. Silver Sulphuric acid. Ag0,S03 + HC1 = AgCl + S03 + HO. The sulphuric acid is precipitated, in its turn, by a solution of baryta, which is added dropwise, so as to use only the quantity absolutely necessary. The liquid is again filtered, and placed under the receiver of the air-pump, above a large capsule contain- ing concentrated sulphuric acid. We may thus obtain the binoxide of hydrogen in a state of great concentration, and even of entire purity. A condition essential to the success of the experiment is to keep the vessel containing the acid liquid in ice, whilst we dissolve the binoxide of barium in it, in order that the fluid may not become heated, which would decompose a great portion of the binoxide of hydrogen. The precipitates of sulphate of baryta which are suc- cessively separated, contain a considerable quantity of fluid: we must squeeze them carefully in a cloth, so as to lose as little fluid as possible. It is also well to add, from time to time, a few 114 ELEMENTS OF CHEMISTRY. drops of chlorohydric acid, to replace that which is lost in all these manipulations. The experiment may be much simplified as follows:—After having, for the first time, saturated the solution of chlorohydric acid with the binoxide of barium, we add an additional quantity of concen- trated chlorohydric acid; then a second dose of binoxide of barium, which gives an additional quantity of binoxide of hydrogen and chloride of barium. By exposing the solution to a very low tem- perature, a great portion of the chloride of barium crystallizes : it is separated by pouring the fluid into another vessel. We again add chlorohydric acid, then the binoxide of barium, and so on. We thus obtain a fluid highly charged with binoxide of hydrogen, and never containing more than the quantity of chloride of barium which it can hold in solution at a very low temperature. This quantity is not great, if we take care, at the close, to plunge the solution into a freezing mixture of pounded ice and sea-salt, in which the temperature falls to 14°. In order to separate the chloride of barium which remains in the fluid, we add, gradually, the sulphate of silver, which precipitates, at once, the chlorine in the state of chloride of silver, and the barium in the state of sul- phate of baryta. These precipitates are separated, and the fluid is evaporated under the receiver of an air-pump. § 90. The binoxide of hydrogen, reduced to its maximum of concentration, is a colourless fluid, of a syrupy consistence, and possessing a peculiar odour. Its density is 1.453. It has never been solidified at any temperature. This fluid is not very fixed, and decomposes spontaneously at a temperature of 59° to 68°. When heated, its decomposition is very rapid, and sometimes takes place with an explosion. The binoxide of hydrogen dissolved in water is more fixed, and does not decompose until the liquid is heated to 104° to 122°. The ready decomposition of the binoxide of hydrogen by heat, renders its analysis very simple. We weigh a certain quantity of the binoxide, dissolve it in water, boil the solution, and collect the oxygen which is given oft'. Now, it will be remembered that this quantity of oxygen is precisely equal to that which exists in the quantity of water arising from the decomposition of the binox- ide, and which is formed by subtracting the weight of the oxygen collected from the weight of the binoxide submitted to analysis. This analysis is performed in the apparatus represented in fig. 166. The solution of the binoxide of hydrogen is put into the small flask A, to which is fixed a curved tube bed, of which the curved end descends into a cylinder C full of mercury, but so that the end d of the tube may be above the level of the mercury. Before fitting the cork to the neck of the small bottle, we pass over the tube cd a graduated tube B, which descends into the test- glass, until the tube d nearly reaches its top: the tube B is held HYDROGEN 115 in this position by the support S. The cork is then fitted, and the level of the mercury within and without the tube exactly adjusted, which is readily done by ab- stracting or adding, with a pi- pette, a small quantity of mercury in the test-glass C: lastly, we note to what division the mercury rises. The balloon is heated: as the oxygen disengages, the bell-glass is raised, so as to maintain an equal pressure within and without. When the wrater has boiled for a few moments, decomposition is completed. The apparatus is al- lowed to assume the ordinary temperature, the level of the mercury is established, and the divi- sion it has reached marked down: the increase of volume of gas in the bell-glass represents the volume of disengaged oxygen. We have just seen that the binoxide of hydrogen produces, when decomposed by heat, quantities of water and oxygen such that the oxygen disengaged is precisely equal to that which exists in the water which has become free. Now, water consists of 1 equivalent of hydrogen and 1 of oxygen, and we write it HO: the binoxide of hydrogen is therefore considered as consisting of 1 equivalent of hydrogen and 2 of oxygen, and its chemical formula should be II02. Solutions of the binoxide of hydrogen being more fixed when they contain some hydrochloric acid, a small quantity of this acid is generally allowed to remain, when we wish to preserve them. The binoxide of hydrogen parts with its oxygen readily to a number of substances, converting metallic oxides into peroxides. It bleaches the tincture of litmus like chlorine. A drop on the skin makes a white mark. § 91. The solution of binoxide of hydrogen in contact with cer- tain substances exhibits some very remarkable phenomena. With gold, platina, and silver finely divided, or certain metallic oxides, such as the peroxide of manganese, peroxide of lead, etc., it de- composes with effervescence by giving off oxygen, whilst the sub- stances whieh affected the decomposition undergo no change. These substances acted by their presence, but did not enter che- mically into the reaction. This mysterious aetion is called action of presence or catalysis; we shall find it again in several phe- nomena. It is proper to remark that substances act, in these cases, with more energy in proportion to their division, for the oxygen only disengages from the surface. Fig. 166. 116 ELEMENTS OP CHEMISTRY. If we add a few drops of sulphuric acid to oxygenated water which is decomposing from the presence of silver or of peroxide of manganese, the evolution of the gas is immediately arrested, hut reappears if the acid is saturated by a base. Salts do not effect the decomposition of oxygenated water. Metallic oxides easily reduced, as the oxides of silver, gold, and platinum, exhibit a very remarkable phenomenon with oxygenated water; which is not only decomposed, hut the oxides themselves part with their oxygen and return to the metallic state. The ready decomposition of oxygenated water by the peroxide of manganese furnishes a simple method of determining by ap- proximation the richness of a solution of binoxide of hydrogen. We fill a small graduated bell-glass with mercury, and, with a pipette, pass a small quantity of the solution to the top of it. We mark the number of divisions it occupies, and introduce some finely divided peroxide of manganese, wrapped in tissue-paper. Decomposition ensues as soon as the powder reaches the fluid. The volume of the oxygen disengaged, compared with the volume of the solution which has produced it, gives the richness or strength of the fluid. 117 NITROGEN, OR AZOTE.* Equivalent = 175.0. § 92. We have seen that atmospheric air supports combustion only by means of the oxygen it contains. When the oxygen of the air has been absorbed by the combustible substance, there remains a gas in which all combustion is immediately extinguished. This gas is nitrogen, or azote. We float on the surface of the water of a cistern (fig. 167) a large cork, on which is placed a small porcelain capsule con- taining a bit of phosphorus, to which fire is communicated by a taper, and the capsule is immediately covered with a large bell-glass, immersed a short distance in the water. Com- bustion continues in the confined volume of air, until the oxygen has entirely disappeared in consequence of its combination with phosphorus. From this com- bination results phosphoric acid, which dissolves in the water. When the gas has cooled, after the extinction of the phosphorus, we find that its volume has considerably decreased, and is reduced to about £. If we require only a small quantity of nitrogen gas, we may deprive the air of its oxygen by means of phosphorus at the ordi- nary temperature. It is sufficient to allow a stick of phosphorus to remain for twenty-four hours in a bell-glass filled with air, over the pneumatic cistern. Copper, heated to redness, also deprives air very perfectly of its oxygen. A current of pure nitrogen can readily be procured from the gasometer described (§ 60), by introducing some copper turnings into a hard glass tube ef (fig. 168), one of the ends of which, e, is made to communicate with the tube c of the gasometer, and to the other end / a discharging tube is fitted, which allows the gas to be collected. As atmospheric air always contains a small quantity of carbonic acid, and is moreover saturated with water in the gasometer, if we wish to obtain perfectly pure nitro- gen, it is necessary, before it reaches the tube filled with copper turnings, to pass it first through a tube T containing pumice-stone soaked in caustic potassa, which absorbs the carbonic acid, and a Fig. 167. * The name nitrogen (which generates nitre) has been given to this gas, because it forms an acid with oxygen, nitric, also called azotic acid, which, combining with potassa, forms the nitrate of potassa, commonly called nitre, or saltpetre. 118 ELEMENTS OF CHEMISTRY. Fig. 168. second tube T' filled with pumice-stone imbued with sulphuric acid, which absorbs the water. The glass tube ef containing the cop- per, is arranged on a small sheet-iron furnace, which permits of its being raised to a red-heat: the tube is wrapped with a sheet of foil, to prevent its losing its shape. Nitrogen is often obtained, in the laboratory, quite as pure, by another method—the decomposition o£ ammonia by chlorine. Am- monia is a compound of hydrogen and nitrogen : chlorine combines with hydrogen to form chlorohydric acid, which, in its turn, com- bines with the undecomposed ammonia, to form the chlorohydrate of ammonia, which remains in solution in the wTater. The nitrogen, being set free, is disengaged. The flask (fig. 169) contains a mixture of peroxide of manganese and chlorohydric acid: the chlorine disengaged in this reaction passes into a tubulated bottle, half filled with a solution of ammoniacal gas in water : it instantly loses its yellow colour, and an infinity of little bubbles of nitrogen, which may be collected when the atmospheric air has been entirely expelled from the apparatus, escape from the fluid. This experiment is free from danger so long as the ammonia- cal solution contains an excess of ammonia : but, if the disengage- ment of chlorine be continued after the ammonia has been entirely changed into a chlorohydrate, the chlorine acts on the chlorohy- drate of ammonia, and gives rise to an extremely dangerous com- pound, which we shall meet again under the name of chloride of Fig. 169. NITROGEN. 119 nitrogen. This substance appears under the form of yellow oily drops, and its formation should be carefully avoided, as it is one of the most explosive substances known. We also obtain very pure nitrogen, and in large quantities, by boiling in a flask a concentrated solution of the nitrite of am- monia, which decomposes into water and nitrogen. The composi- tion of nitrite of ammonia is represented by the formula NHS II0,N03: it contains the elements of 4 equivalents of water and 2 of nitrogen. In fact, we have NH3II0,N03=4H0+2N. § 93. Nitrogen is a colourless, inodor- ous and tasteless gas, which thus far has never been liquefied under any pressure. Its density is 0.9713, that is, something less than that of the air. A lighted taper is instantly extinguished in it (fig. 170). Animals cannot live in nitrogen, and \ they perish for want of oxygen, which is indispensable to respiration; hence it has received the name azote from some che- mists (from » privative, and life). Nevertheless, this gas exerts no delete- rious influence on their organs, since f of the atmosphere consist of it. Water dissolves a very small quantity of nitrogen, about T||-3 of its volume: or, in other words, 1 litre of wrater dissolves 25 cubic centimetres, or 1 kilogramme of water dissolves 0sm.031 of nitrogen. Fig. 170. ATMOSPHERIC AIR. § 94. Atmospheric air consists essentially of a mixture of oxygen and nitrogen, in proportions identical throughout the globe. It contains in addition, a very small quantity of carbonic acid gas and a variable quantity of vapour of water. Air contains, more- over, but in scarcely appreciable quantities, some other gases or vapours, arising from the decomposition of animal and vegetable matter. § 95. We will describe the various methods by which the com- position of atmospheric air may be exactly determined.* This * Air was considered by the ancients as one of the four elements of nature. This opinion reigned undisturbed until toward the close of the eighteenth century. Lavoisier first proved, incontestably, that air was a mixture of two gases, pos- sessing different properties, and nearly determined their proportions. The fol- lowing is the experiment which led this illustrious and unfortunate chemist to this result (Traite elementaire de Chimie, tom. I. p. 35, ed. 2d): “ I took a matrass containing about 36 cubic inches, with a very long neck, 6 or 7 lines interior diameter, and bent it as in fig. 171, so that it could be set in a furnace MN, whilst the end 0 of the neck opened under a bell-glass PQ in a mei’curial cistern RS. I introduced into the matrass 4 ounces of very pure mer- 120 ELEMENTS OF CHEMISTRY. analysis consists of two operations performed separately. The object of the first is to determine the carbonic acid and the water; cury; then, sucking with a siphon which I passed under the bell-glass PQ, I raised the mercury to LL. I marked this height carefully by pasting a strip of paper over it, and observed exactly the barome- ter and thermometer. “ Things being thus pre- pared, I lighted the lire in the _ furnace MN, and kept it up | for nearly 12 days, so that the mercury was heated nearly to the boiling point. “Nothing remarkable oc- curred on the first day: the mercury, though not boiling, was constantly evaporating : it coated the inside of the vessels with drops, at first very small, but which gradually increased, and, when they had acquired a certain size, fell spontaneously to the bottom of the vessel, and joined the balance of the mercury. On the second day, I saw swim- ming on the surface of the mercury small red particles, which, for 4 or 5 days, increased in number and size, after which they ceased enlarging, and remained absolutely in the same state. After 2 days, seeing that the calcination of the mercury (oxidation of the mercury) progressed no longer, I extinguished the fire, and allowed the vessels to cool. The volume of air, contained as well in the matrass as in its neck and under the empty part of the bell-glass, reduced to a pressure of 28 inches, and to the temperature of 19°, was, before the operation, of about 55 cubic inches. After the operation, the pressure and temperature being the same, there remained only 42 to 43 inches: there was consequently a diminution of volume of nearly one-sixth. On the other hand, having carefully collected the red particles which formed, and separated them as much as pos- sible from the liquid mercury with which they were coated, they were found to weigh 45 grains. “ The air which remained after this operation was reduced to f of its volume by the calcination of the mercury, and was no longer fit for respiration or com- bustion ; for animals perished in it in a few moments, and a candle was as rapidly extinguished as if plunged into water. “ I took the 45 grains of red matter which had formed during the operation, and introduced them into a small glass retort, to which was adapted an apparatus calculated to receive the liquid and aeriform products which might separate: having lighted the fire in the furnace, I observed that as the red matter became heated, its colour became more intense. When the retort subsequently approached incandescence, the red matter gradually began to lose its volume, and in a few minutes entirely disappeared: at the same time, 41 £ grains of liquid mercury collected in the receiver, and 7 or 8 cubic inches of an elastic fluid much more fitted to support life and combustion passed under the bell-glass. “Having introduced a portion of this air into a glass tube of an inch in dia- meter, and plunging a candle therein, it burned with a dazzling flame; and charcoal, instead of burning quietly, as in ordinary air, burned with a flame and sort of decrepitation, like phosphorus, and a brilliant light which the eye could hardly endure. “A little reflection on this experiment will show us that the mercury, by calcining, absorbs the salubrious and respirable portion of the air, and that the remaining portion is a kind of mephitis, incapable of supporting animal life and combustion. Atmospheric air is, therefore, composed of two elastic fluids of dif- ferent, and, as it were, of opposite natures. “ This important truth is proved by recombining the two elastic fluids thus Eig. 171. NITROGEN. 121 and of the second, to determine the proportions of hydrogen and nitrogen in the air, when freed from its carbonic acid and aqueous vapour. Fig. 172 represents the apparatus by means of which we ascer- tain very accurately the quantity of carbonic acid and vapour of water which exists in the atmosphere. Fig. 172. A cylindrical vessel V, of galvanized sheet-iron, containing 50 to 100 litres (11 to 22 gallons), is supported by a tripod over a large tub capable of holding all the water contained in the cylinder Y. This cylinder has, at its lower part, a stopcock r, furnished with a tube about a decimetre (4 inches) in length, and curved at its end. Two tubes, a and b, are attached to the vessel. In the central tube a, we fasten hermetically, by means of a metallic stopper and soft wax, a metallic tube ad open at both ends: this tube is curved at c, and has a stopcock s. In the lateral tube b, we introduce a ther- mometer T, the bulb of which should descend toward the middle of the vessel V. The capacity of the vessel Y may be very exactly ascertained. To do this, we take a balloon (fig. 173) of the capacity of about 10 litres (2|- gallons), on the neck of which is engraved a horizontal line a; the balloon is filled with water to this line, and weighed. separately obtained, that is, the 42 inches of mephitis or non-respirable air, and the 8 cubic inches of respirable air, we thus recompose an air resembling that of the atmosphere, and nearly as fitted for combustion, respiration, and the cal- cination of metals.” Lavoisier adds that the proportion of respirable gas formed by his experiment, is probably a little too feeble, because he could not combine it perfectly with the mercury. 122 ELEMENTS OF CHEMISTRY. The vessel is then emptied very accurately of water, and again weighed, and the difference gives the weight of the water. It is easy to show that if the balloon be several times filled and emptied in the same manner and at the same temperature, we shall always obtain within a small fraction the same weight P of water. The vessel Y is completely filled with water at the same temperature, the thermometer T adjusted, and the tube ad. The stopcock s being opened, the cock r is opened, and water allowed to flow into the balloon (fig. 173) as far as the level a; the cock r is then closed, and the balloon emptied precisely in the same manner as it had been gauged. This operation is repeated until the vessel V is en- tirely emptied. We thus find that the balloon has been filled, a certain number of times n, and, in the last operation, if it is not completely filled, the water it contains is weighed. Suppose there remains a weight of water p: it is evident that the vase Y contained a weight of Avater represented by nV+p. If the Avater were at the temperature of +39J°, the weight wP-fp, in kilogrammes, would represent the capacity Y of the vessel in litres. But this Avater is generally at a temperature t, at which it possesses a density somewhat less than at 39|° : this density 8, for any temperature t, is found in all Avorks on physics; the capacity of the vase Y in litres will therefore be represented by y _ nVJrP 8 ’ In order to determine the quantities of carbonic acid and vapour of water which exist in the air, Ave fill the vase V with water, and attach to the tube c a series of tubes, A, B, C, D, E, F. The tubes A, B, E, E, are filled with large pieces of pumice-stone soaked in concentrated sulphuric acid; the tubes C, D, with pumice-stone soaked in a concentrated solution of caustic potash: lastly, to the last tube A, we adapt a long tube fg, which passes out into the external atmosphere Avhich we are about to analyze. The curved tubes, containing the pumice-stone, are closed at both ends with good corks pierced by smaller curved tubes, as in fig. 172. The corks should be covered Avith sealing-AA’ax, which renders them very smooth. We are thus more certain of a hermetical closure; and the corks not being exposed to the air, cannot change in weight, by absorbing or giving off moisture during the experiment. The tubes are joined together by means of small caoutchouc tubes, strongly tied on Avith silk thread. The two tubes A and B are weighed together; and likewise the three tubes, C, D, and E. It is unnecessary to weigh the tube F, as it remains attached to the apparatus, and is of no use, except to prevent the vapour of Avater disengaged from the vessel Y from passing into the tube E. Fig. 173. NITROGEN. 123 The apparatus being thus arranged, the water is allowed to flow from the vessel V, which is called an aspirator. This discharge can only take place so long as bubbles of air reach this vessel by the tube ad: the discharge of the water will, moreover, have an equal velocity, because it will take place under the pressure of the column of water comprised between the level xy of the lower orifice and the level x'y' of the orifice of the tube ad. In fact, the tube ad is entirely filled with air, and communicates freely with the atmosphere, from the connection of the tubes A, B, C, D, E, F; consequently, on the whole level stratum x'y' which passes through the orifice d, there is a pressure equal to that of the external atmosphere. In the stratum xy, the pressure which expels the water is equal to the pressure of the atmosphere, increased by the pressure produced by the column of water between the levels xy and x'y'. The pressure which opposes this discharge is that of the external atmosphere: the discharge will therefore take place under the pressure produced by the column of water comprised between the levels x'y' and xy, and will be the more rapid for a uniform opening of the cock r, as the column of water between xy and x'y' is greater. The flow of water taking plaee only under the pressure of the column comprised between the levels xy and x'y', it is evident that this flow will be strictly constant whilst the level of the w'ater in V is above the stratum x'y'. This is not the case with the air: it will enter more rapidly as the level of the water descends in the vessel Y. Let us suppose that this level has reached the stratum x"y", the pressure on the stratum x'y' is equal to that of the external atmosphere ; at an indefinite moment it equals the elastic force of the gas in the upper part of the vessel y, and, in addition, the weight of the liquid column comprised between the levels x"y" and x'y'. Thus, by supposing the vessel to be perfectly cylin- drical, so that the level of the water may descend regularly, by reason of the constant discharge of the fluid, the air which will enter the apparatus during a minute will go on increasing: for, it must not only fill the vacuum, always the same, made by the discharge of the water, but likewise must constantly increase the elastic force of the internal air, so that this force, added to the pressure of the liquid column comprised between the planes x"y" and x'y', and which always goes on diminishing, equals the pressure of the external atmosphere at the level x'y'. Absolute regularity of the current of air which traverses our apparatus is not indispensable to the success of the present experi- ment; we must, however, call attention to this circumstance, for this regularity is necessary for other experiments, and it is proper to show that we do not obtain it by the arrangement just described. The external air, therefore, traverses, before reaching the vase y, the series of tubes A, B, C, D, E, F„ In the two tubes A and 124 ELEMENTS OF CHEMISTRY. B it deposits its moisture; in the tubes C and D its carbonic acid. But as the gas arriving in the latter tubes is completely dry, and as the solution of caustic4potassa gives off a sensible quantity of vapour of water, we have placed, after the tubes C and D, the tube E, filled with pumice-stone soaked in sulphuric acid, which retains this small quantity of water. When the aspirator is perfectly empty, we mark the height I! of the barometer and the temperature t of the thermometer T. The curved tubes are detached; A, B weighed together, as like- wise C, D, E. The increased weight of these two systems of tubes gives in A and B the quantity of vapour of water, and in C, D, E, that of carbonic acid, existing in the atmospheric air which has traversed the apparatus. We must now ascertain the weight of this air from the data of the experiment. The volume of air filling the aspirator is V; but this air is satu- rated with vapour at the temperature t. Let us designate by/ the maximum elastic force of the vapour of water at this temper- ature t. The elastic force of the dry air which has entered the apparatus is H—/: a quantity of atmosphere has therefore entered our apparatus such that it occupies, after having entirely parted with its vapour of water and its carbonic acid, a volume Y, at a temperature t, and under a pressure II—/. The weight P of this air, dried and deprived of carbonic acid, is therefore P-V.1-.S9M. Let us suppose that the weight of the carbonic acid found is p, and the weight of the vapour of water is p'; we shall conclude from our experiment that a weight V+p+p' of atmospheric air, under the conditions in which we have analyzed it, contains p of carbonic acid, and p' of vapour of water: and we may calculate by a simple proportion the quantities of carbonic acid and water found in 100 parts of this atmospheric air. It is important that the pumice-stone in the tubes should be in large fragments, and merely wetted with the oil of vitriol, in order that an excess of this liquid may not accumulate at the lower part of the curved tubes. The external air ought to pass freely through all these tubes: for, otherwise, at the end of the experiment, the air which fills the aspirator Y might have an elastic force much inferior to that of the external air. We turn upward the pipe terminating the stopcock r, in order that, after the discharge, the curved part may remain filled with water, and prevent the entrance of air into the vessel Y. Experiment has shown that free atmospheric air contains quan- tities of carbonic acid, varying from 4 to 6 ten thousandths. The quantity of vapour of water is much more irregular, owing to tem- perature and its state of saturation. § 96. Let us now suppose the air to be deprived of its carbonic NITROGEN, 125 acid and its vapour of water, and see how we shall ascertain the proportions of oxygen and nitrogen it contains. This may be done by several methods, of which we shall describe the most perfect. Many substances absorb the oxygen of the air, even at ordi- nary temperatures. It is therefore sufficient, in order to analyze the air, to introduce a certain quantity of air into a graduated bell-glass, measure this volume very accurately, under given con- ditions, introduce the absorbing substance, and allow it to remain in the bell-glass until the volume of the gas no longer decreases sensibly, and lastly, to measure again with great exactness the remaining volume, which must be pure nitrogen. Phosphorus is the absorbing substance best adapted to this purpose. The experiment is performed as follows: Melt some phosphorus under water, and then run it into bullet- moulds, always under water at about 104°. Introduce into the cavity of the mould, whilst the phosphorus is yet fluid, a platinum wire twisted into a curl at the end. The mould is then plunged into cold water to solidify the phosphorus, and we have a small ball of phosphorus firmly fixed to the end of the platinum wire. This being done, we introduce into a graduated bell-glass placed over mercury, a certain volume of air, which is carefully mea- sured. The inside of the bell-glass must still be somewhat damp. Although it may have been carefully wiped, and no drops of water be visible, the air intended for analysis will be saturated with moisture by the small quantity of water given off by the sides of the bell-glass. Let t be the external temperature, H the height of the barometer, / the tension of the vapour of water corresponding to the tempera- ture t, and which will be found in a small table annexed to this work. The volume Y of the gas already observed would be, were it dry, at the temperature of 32°, and under the pressure of 0m.760, V 1 H-f » • 1+0.00367.t 0.760 The ball of phosphorus is introduced into the gas (fig. 174), which is easily done by means of the platinum wire to which it is attached, and allowed to remain until the gas no longer diminishes in volume. This sometimes requires more than twenty-four hours. Absorption proceeds more rapidly by placing the bell-glass in the sunshine. When the absorption is completed, the phosphorus is withdrawn, and the volume of gas remaining measured, after it has ac- quired the temperature t' of the surrounding air. Let us suppose that this volume is V7, the barometric pressure H': lastly, that the elastic force of the saturated vapour of water at the temperature t' is f: the volume occupied by this air, deprived of its Fig. 174. 126 ELEMENTS OF CHEMISTRY. moisture, at the temperature of 32° and under the normal pressure of 0m.760, will be V' i n >—f ' 1+0.00367.t'' 0.760 ’ This, therefore is the volume of nitrogen found in a volume V. i+o,oo367of dry atmospheric air: whence may be immedi- ately deduced the volume of nitrogen and oxygen in 100 parts of atmospheric air. The air may likewise be analyzed by employing substances which do not absorb oxygen at the ordinary temperature, but w'hich, when strongly heated, combine actively with this body. The experiment may also be arranged so as to weigh at the same time the oxygen which has combined with the absorbing substance, and the nitrogen which remains free. By performing the experiment in the following manner, we may obtain great accuracy (fig. 175): ah is a glass tube difficult of fusion, filled with metallic copper, and arranged over a long sheet- iron furnace, so that it may be heated throughout its whole length. The stopcocks r and r' are fitted to the ends of this tube. The extremity a of the tube is brought into communication with a balloon V holding about 20 litres (5 galls.), having a stopcock u : and the extremity b communicates with an apparatus ABC. The apparatus A, figured on a larger scale in fig. 176, is intended to absorb the carbonic acid of the air. This apparatus, called Liebig's potassa bulbs, from the illustrious chemist who contrived it, consists of three bulbs, b, c, d, arranged on the same axis, and two bulbs a and e placed above, and communicating with the first by narrow tubes. A concentrated Fig. 175. Fig. 176. NITROGEN. 127 solution of potassa is introduced, so as to entirely fill the three lower bulbs. If, then, we slowly exhaust the air by the tube g, the ex- ternal air enters at/, and traverses the solution of potassa, passing successively through the bulbs; lastly, in order to reach the bulb e, it must pass through a new column of solution of potassa. The gas therefore remains much longer in contact with the potassa than if it were to traverse a straight and unbroken column of fluid, and consequently'frill be in the most favourable conditions for the absorp- tion of carbonic acid. The tube B (fig. 175) is filled with pieces of pumice-stone soaked in a concentrated solution of caustic potassa: it is intended to absorb the last portions of carbonic acid gas which might have escaped from the apparatus A. Lastly, the tube C, filled with pumice-stone soaked in sulphuric acid, completely desiccates the air. This being done, the tube ab is exhausted as perfectly as possible, and the stopcock r and r' closed. This tube, when exhausted of air, is weighed, which weight is represented by p. The balloon V is also weighed under the same circumstances: let its weight be P. The apparatus is then put in order and the tube ab heated to redness. The stopcock rr is then opened: the external air enters the tube ab after having traversed the series of tubes ABC, which deprive it of its carbonic acid and watery vapour : this air gives off its oxygen to the heated metallic copper, and the nitrogen re- mains isolated. The stopcock u of the balloon is opened, and the stopcock r very slightly, so that the gas enters very slowly the balloon V. The rate of its passage can moreover be estimated by the bubbles which traverse the bulbed receiver A: the bubbles or gas should go over one at a time. When the passage of the bubbles becomes slower, which necessarily happens when the dif- ference between the elastic force of the gas in the balloon and that of the external air diminishes, the stopcock r is further opened. At the end of the operation, it is opened completely. As soon as the gas ceases to form, the stopcocks r, r' and u are closed, the coals removed, and the apparatus taken to pieces. The balloon is weighed : let V be its weight: P'—P is evidently the weight of the nitrogen which has entered it. Weigh the tube ab: let p' be its weight: p'—p will be the weight of the oxygen which has combined with the metallic copper, increased by the quantity of nitrogen in this tube. This last quan- tity is easily ascertained by again making a vacuum in the tube, and finding its weight p"; p'—p" is then the nitrogen which has been withdrawn by the air-pump, and p"—p the quantity of oxygen combined with the metallic copper. We therefore find a weight of nitrogen (P'-P )+(p'-p"), and a weight of oxygen p"~p, 128 ELEMENTS OF CHEMISTRY. forming a weight of dry atmospheric air deprived of its carbonic acid, represented by (P'-P ) + (p'-/0+(/'“P)=(P'-p) + (p'-p)- It will, therefore, be easy to ascertain, by a proportion, the weights of oxygen and nitrogen which enter into 100 parts by weight of atmospheric air; and as we know the densities of oxygen and nitrogen, we may equally deduce the composition of the air by volume. § 97. Great care must be taken in weighing the balloon Y, if we wish to be very exact. This operation is necessarily done in the air : now, we know that a body immersed in a fluid loses of its weight a portion equal to the weight of the fluid it has displaced. The volume of air displaced by the balloon is the same in both cases : if, therefore, the air were of the same density in both cases, the difference P'—P would not be affected by this circumstance, and would give exactly the weight of the nitrogen which has en- tered the balloon. But if the air has changed between the two weighings, in consequence of variations of temperature or baro- metric pressure, the quantity of air displaced will not be the same, and the difference P'—P will no longer correctly represent the weight of nitrogen in the balloon. It is difficult to calculate the correction in the value of P'—P, but we can experiment so as to guard against this cause of error. Glass balloons, and in general all large vessels, should be weighed by hooking them beneath the dishes of the balance (fig. 177). Fig. 177. NITROGEN. 129 Instead of balancing the balloon hooked beneath one of the dishes, by means of ordinary weights placed in the other, it is balanced by a second balloon hermetically closed, and exactly resembling the first. This second balloon is hooked to the other dish of the balance, so that it floats in the same stratum of air as the first. The two balloons displacing the same volume of air, it is evident that all variations which occur in the air affect them exactly in the same manlier, and that the difference of weight P'—P between these two weights will be independent of their variations. It now remains to us to point out how to arrange two balloons which displace exactly the same volume of air. In order to do this, we ascertain exactly the volume of air dis- placed by the balloon A, which is to serve for the experiment. For this purpose, this balloon is completely filled with water, and it is weighed immersed in water having exactly the same tem- perature as that which fills it. The apparent weight of the bal- loon filled with water is so slight that it may be ascertained by hooking it beneath one of the dishes. We withdraw the balloon from the water and weigh it again, but in the air, after having wiped it dry. For this second operation, we use a strong ordinary ba- lance. The difference between the two weights will be evidently the weight of water displaced by the external volume of the balloon. A second balloon B, having nearly the same capacity as A, is selected, and the weight of water its external volume displaces as- certained as before. Let us suppose that the external volume of the second balloon is rather smaller than that of the first furnished with a stopcock : we fasten to the neck of the balloon B, with com- mon cement, a brass mounting terminating in a hook intended to suspend the balloon beneath the balance. Let us suppose that the weight of water displaced by this mounting, added to the weight we previously found for the water displaced by the external vo- lume of the balloon B, be less by n grammes than the weight of water displaced by the balloon A : it will be sufficient to append to the balloon B a small glass tube closed at both ends, which exactly displaces n cubic centimetres of water. If the balloon B is, Avith its mounting, much lighter than the balloon A, we introduce into it, before closing it hermetically, a quantity of mercury sufficient to balance the balloon A with a very small addition. Fig. 177 represents the two balloons hooked under the dishes of a Fortin’s balance. The balance should be con- tained in a closet of thin wood, to shield it from currents of air. We are thus certain that both float in strata of air of the same temperature, and that they are not unequally influenced by the presence of the experimenter. The oscillations of the balance may also be observed, from a distance, with a spy-glass. § 98. Atmospheric air may be also very exactly analyzed by means of the eudiometer. 130 ELEMENTS OF CHEMISTRY. We introduce into the eudiometer, previously wiped dry, a cer- tain volume Y of atmospheric air: the temperature is t, the ba- rometric pressure H, and the elastic force of the saturated vapour is / at the temperature t. The volume of dry air will be, there- fore, at 32°, and under the pressure 0m.760, tr 1 . —~\T . » • l+0.00367.t 0.760 Y 0 A volume of hydrogen gas is then introduced somewhat less than that of the air, and we measure anew the volume V' of the gas: the temperature and pressure 'will not have sensibly changed in the interval, and we shall have the same values for t, H, and /. But let us suppose, for greater generalization, that these quanti- ties have become t', II', and f: the volume of the dry gaseous mixture would be, at 32°, under the normal pressure of 0m.760, y/ . H ~jf/=Y/ . V • 1 +0.003670.760 T o Y'0—V0 will therefore be the volume of dry hydrogen under nor- mal conditions. An electric spark is passed through; the oxygen of the air burns a volume of hydrogen double of itself, and the product of the combustion condenses in the state of liquid water, of which the volume is of no importance with relation to the volume of the gases which have produced it. When the eudiometer acquires the same temperature as the surrounding air, the volume of the re- maining gases is weighed. Let us suppose that this volume be Y", the barometric pressure H", the temperature t" and/", the elastic force of saturated vapour corresponding to the temperature t": the volume of the dry gaseous mixture will be, at 32°, and under the pressure 0m.760, y// 1 v • 1 + 0.00367X" 0.760 r o Y"0—V'0 is therefore the volume of the dry oxygen and hydro- gen gases, under normal conditions, which have combined. yn yt 8 3 0 will be the volume of oxygen, V"-—V' 2 0 g—~ will be the volume of hydrogen. We conclude, hence, that a volume Y0 of atmospheric air con- yn yf y// yt tains a volume - —0 of oxygen, and a volume Y0— — 3—0 of nitro- gen. The eudiometric analysis of the air gives very exact results when this analysis is carefully conducted. But, when we desire very great exactness, it is best to employ a eudiometer of peculiar construction, such as was mentioned in § 83; and which we will describe in the fourth part of this course, when treating of the analysis of compounds and gaseous mixtures. NITROGEN, 131 It has been ascertained, by a great number of analyses, that atmospheric air contains, on an average, in volume Oxygen 20.90 Nitrogen 79.10 100.00 or in weight, Oxygen 23.10 Nitrogen 76.90 100.00 The constitution of the air collected in various localities, and at different heights in the atmosphere, affords scarcely any sensible variations. It is very easy to collect a small quantity of air in any given locality, by means of small tubes, drawn out at both ends, and con- taining 30 or 40 cubic centimetres (2 or 2\ cub. in.). They are to be filled with air by a bellows, and the points closed by the flame of an alcohol lamp. The air contained in these tubes may be preserved for an indefinite time and analyzed in the laboratory by the eudometric process. § 99. The great constancy observed in the constitution of the air, has led some chemists to regard atmospheric air, not as a mix- ture of oxygen and nitrogen, but as an actual chemical combina- tion of these gases. We shall give the principal reasons which prove this opinion to be erroneous, and that oxygen and nitrogen are merely mixed in atmospheric air. Experiment has shown that two gases always combine in simple ratio of volumes. Now, the simple proportion which most closely approaches the direct analyses of the atmospheric air is the fol- lowing : £ of oxygen or oxygen 20.00 f of nitrogen or nitrogen 80.00 100.00 The discrepancy between these numbers and the results of analysis cannot be attributed to the error of experiment, since analyses of the air, made in various ways, have always led to the same result. Heat is always disengaged in the combination of two gases: now there is no appreciable change of temperature when we mix oxygen and hydrogen; and if these gases are mixed in the propor- tions constituting the air, we obtain a gaseous mixture absolutely identical with our atmosphere. But the most convincing proof that air is a simple mixture of oxygen and nitrogen, is in the manner in which atmospheric air 132 ELEMENTS OF CHEMISTRY. behaves with water. We have seen (§ 81) that water wThich has been for a long time in contact with air always contains a certain quantity of gas in solution, and have described the mode by which this gas may be separated and collected. If the atmospheric air be a compound of oxygen and nitrogen, the gases dissolved in water should present the same composition as the atmosphere, and contain 20.90 '. oxygen, 79.10 nitrogen. If, on the contrary, air is only a simple mixture of these two gases, as oxygen and nitrogen are not equally soluble, the compo- sition of the dissolved gases will be different from that of the air, and may even be calculated by the rule pointed out (§ 81). Let us admit, for the sake of clearness, that air is formed of \ oxygen and f nitrogen, the fractions of solubility being for oxygen and i for nitrogen, the gases will be found dissolved in water in the proportions f,7, of oxygen, fir of nitrogen; or, 7,=0.046, 7=0.025, we shall have in the dissolved gases, Oxygen i.0.046 0.0092 31.5 Nitrogen f.0.025 0.0200 68.5 0.0292 100.0. Now, the direct analysis of this gaseous mixture extracted from water has given y Oxygen 32.0 Nitrogen 68.0 llO. Which agrees, as nearly as possible, with the composition we have calculated, founded on the law of solubility of the gases, and the presumption that the atmospheric air is a mixture of nitrogen and oxygen. COMPOUNDS OF NITROGEN AND OXYGEN. § 100. We are acquainted with five definite combinations of nitrogen and oxygen: 1. The protoxide of nitrogen; 2. The deutoxide of nitrogen ; 3. Nitrous or azotous acid; 4. Hyponitric or hypazotic acid; 5. Nitric or azotic acid. NITROGEN, 133 To which the following formulae have been assigned: 1. Protoxide of nitrogen NO or AzO ; 2. Deutoxide of nitrogen N02 “ Az02; 3. Nitrous acid N03 “ Az03; 4. Hyponitric acid N04 “ Az04; 5. Nitric acid NOs “ Az05. § 101. Nitric acid is obtained by heating saltpetre or nitrate of potassa with concentrated sulphuric acid. Nitric acid being more feeble and volatile than sulphuric, is separated from its com- bination and passes over in distillation. Nitrate of potassa is also called nitre, whence azotic acid was originally called nitric. This .tame is in very general use at this day, although it does not har- monize with the rules of our chemical nomenclature. Nitric acid has not been yet obtained free from water, or anhy- drous.* The most concentrated contains 14 per cent, of water: it has a density of 1.522, and boils at 186.8°. If a small quantity of water be added to this acid, and the mixture be distilled, the iirst portions which pass over contain more real acid than the fluid which remains in the retort. If we observe a thermometer plunged into the boiling fluid, wTe will see the temperature continually rise, until it reaches 253.4°. Here it remains stationary, and the fluid which distils presents a constant composition: it contains 40 per cent, of water. If, on the contrary, we add considerable water to the most con- centrated acid, and distil this new mixture in a tubulated retort pro- vided with a thermometer, this instrument will at first mark about 212°, but the temperature will gradually rise to 253.4°, and remain stationary until the end of the distillation. The first portions are nearly pure water: the succeeding contain a greater quantity of acid; so that the fluid in the retort becomes more and more con- centrated, until it contains only 40 per cent, of water. Now, experiment has proved that all homogeneous compounds which are not decomposed by ebullition, boil at a constant temperature under the same pressure. When a liquid thus presents a constant tem- perature during the distillation it undergoes in consequence of boiling under the same pressure, it is regarded as homogeneous, and is said to be a compound of definite proportions. The liquid acid, formed of 60 per cent, of pure nitric acid, and 40 of water, presents the characters of a compound with defi- nite proportions. The density of this acid is 1.42. NITRIC ACID, NOs or AzOs. * M. Deville has since succeeded in preparing anhydrous nitric acid by passing dry chlorine over dry nitrate of silver. It forms transparent, colourless crystals, of a right rhombic form, fusing at 85°, boiling at 113°, and decomposing near the latter point. 134 ELEMENTS OF CHEMISTRY. In the first hydrate of nitric acid, the proportion of the oxygen of the water to the oxygen contained in the pure acid is as 1 to 5; its formula is therefore NOs+HO. In the second hydrate, this proportion is as 4 to 5, and the formula is NOs+4HO. §102. The first hydrate N05+H0 congeals at—58°. When pure it is colourless, but soon turns yellow when exposed to the light. This agent decomposes nitric acid, leaving oxygen and hyponitric acid N04, which remains dissolved in the undecomposed acid. Nitric acid N05+H0 is therefore a not very stable com- pound : it is easily decomposed by heat, for a few successive dis- tillations decompose a large proportion of it. If the vapour of nitric acid be driven through a highly heated porcelain tube, the acid is completely decomposed into nitrogen and oxygen. If the tube be less heated, the products of decomposition are oxygen and hyponitric acid. When we endeavour to deprive nitric acid N05+H0 of the water it contains, it is decomposed into oxygen and nitrous acid: which happens when it is distilled with four times its weight of concentrated sulphuric acid or with anhydrous phosphoric acid, both of which have great affinity for water. Nitric acid N05+H0 has a marked affinity for water: it be- comes heated when mixed with this fluid, and fumes in a moist atmosphere. This last property has given to this hydrate the name of fuming nitric acid, and depends on the monohydrated nitric acid N05-f-H0 possessing a greater tension of vapour, at equal temperatures, than nitric acids containing larger proportions of water. It therefore follows that when the fumes of the mono- hydrated acid reach the damp air, and there combine with an addi- tional quantity of water, the more hydrated acid cannot remain entire in the state of an invisible vapour in the air, and a consi- derable portion of it precipitates in the form of mist. The second hydrate NOs+4HO is much more fixed than the first: it is neither decomposed by the influence of light alone, nor by repeated distillations. By distilling it with about its weight of concentrated sulphuric acid, f of its water may be removed, and the first hydrate N05+H0 then passes by distillation. It is proper not to use a great excess of sulphuric acid, for a consider- able portion of nitric acid, would be decomposed. § 103. Nitric acid is easily decomposed by many substances, to which it yields a portion of its oxygen. Carbon and sulphur de- compose it at the boiling temperature: and many metals at the ordinary temperature. It is an active agent of oxidation daily employed in the laboratory. NITRIC ACID. 135 Nitric acid, at its maximum of concentration, being much less fixed than the more dilute acid, ought to be a more energetic agent of oxidation. It is so, in fact, as regards the majority of substances: thus, sulphur, phosphorus, and carbon are much more rapidly acted on by the first hydrate N05+I10, than by the more dilute acids. The contrary, however, obtains with many metals : thus, iron and tin, which are readily acted on by slightly dilute nitric acid, exhibit but little reaction in this acid at its maximum of concentration; but this reaction becomes very energetic when a small quantity of water is added. Nitric acid destroys the majority of animal substances ; it stains the skin yellow, and also imparts this hue to wool. Advantage is taken of this property in dyeing. § 104. Nitrogen and oxygen may combine under the influence of the electric spark, so as to produce nitric acid, the presence of water, or, better still, of water and a powerful base together, being ne- cessary to produce the effect. To prove this, we arrange a curved tube (fig. 178) filled with mercury, so that the two open ends may be plunged into two separate vessels filled with mercury. We introduce into the upper part of the tube a small quantity of air and solution of potassa: and, lastly, establish a communication with the mercury in one of the vessels and the plate of an electrical machine, which is steadily turned, whilst the other vessel communicates with the earth by means of a small iron chain. We thus pass through the tube a series of electric sparks, which effect the combination of the nitrogen and oxygen. After the passage of a great number of sparks, the alka- line solution contains a certain quantity of nitrate of potassa. § 105. We have said that nitric acid was obtained by the distil- lation of saltpetre with sulphuric acid. In this process there are several circumstances worthy of remark. Potassa forms two combinations with sulphuric acid; one neutral and the other acid. The latter contains twice as much sulphuric acid as the first. The neutral combination is anhydrous; its formula is therefore K0,S03: the acid combination contains, on the contrary, a certain quantity of water, with which it does not part under 392°: its formula is K0,2S03-fH0, written thus (no’soV ’ *s considered, i*1 last case, as a double salt formed by the combination of the neutral sulphate of potassa K0,S03 with the sulphate of water H0,S03. If we add to one equivalent of saltpetre, K0,N05, two equiva- lents of monohydrated sulphuric acid 2(S03-fH0), we may form Fig. 178. 136 NITROGEN. (no so3) N05-|-H0, or H0,N05, that is, the bisulphate of potassa and monohydrated nitric acid, which in fact takes place, and distillation will separate the acid. The following are the most suitable proportions for the success of the operation: 100 nitrate of potassa.. 46.61 potassa, 53.39 nitric acid, 79.1 sulphuric acid, 17.7 water, 96.8 sulphuric acid which will give 62.29 of monohydrated nitric acid. But, if we add only one equivalent of concentrated sulphuric acid H0,S03 to one equivalent of nitrate of potassa K0,N05, the reaction becomes much more complicated; only J an equivalent of nitrate of potassa is then decomposed, giving J an equivalent of monohydrated nitric acid, J(NOs+IIO), which distils over, and there remain in the retort a 4 equivalent of acid sulphate of /KO SO \ potassa, i(jiQ’go3/> \ equivalent of undecomposed nitrate of potassa, J(K0,N05). If we increase the temperature, there is a reaction between the acid sulphate of potassa and the undecom- posed nitrate of potassa: neutral sulphate of potassa is formed, and, consequently, a \ equivalent of monohydrated nitric acid becomes free: but as the temperature at which monohydrated acid then forms is sufficient to decompose it, we only obtain reddish brown vapours, and no nitric acid. In the laboratory, fuming nitric acid is obtained by placing in a glass retort equal parts of nitrate of potassa and sulphuric acid: the acid should be introduced by means of a long-necked funnel (fig. 179), so that it may not touch the sides of the neck of the retort; without this precaution, during the distillation, a small quantity of sulphuric would be mixed with the nitric acid. The neck of the retort is introduced with a matrass (fig. 180) which is corked by a continuous current of cold water. No corks should he used in the construction of the apparatus, for nitric acid attacks cork very readily, and the latter might even take fire therefrom. Fig. 179. Fig. 180. NITRIC ACID. 137 In the first stage of the reaction, reddish vapours form, arising from the decomposition of the first portions of nitric acid which become free. These necessarily come into contact with a large quantity of concentrated sulphuric acid which has not yet reacted; they must therefore decompose into nitrous vapours and oxygen. By the application of a proper degree of heat, the greater part of the nitric acid distils over without alteration. The close of the operation is announced by copious reddish vapours filling the re- tort : the distillation must then be stopped and the product con- densed in the receiver separated. This new appearance of nitrous vapours is easily explained: nearly the whole of the nitrate of potassa is decomposed, and, in order that the sulphuric acid may react on the last portions of this salt, it is necessary that the mat- ter in the retort should assume a certain fluidity, which is given only by great elevation of temperature, sufficient, in all cases, to decompose the last portions of nitric acid which become free. The acid collected is not pure: it is coloured yellow by the dis- solved nitrous acid, and may also contain a small quantity of sul- phuric acid introduced during distillation. To purify it, it must be shaken with a small quantity of finely powdered nitrate of lead, and then distilled in a retort; the first portions containing the nitrous acid being collected, the receiver is changed, and the pure nitric acid collected. The operation should be arrested before all the acid is distilled, for the last portions may contain some nitrous acid, generated because the sides of the retort, being no longer bathed by fluid, may become so heated as to decompose the nitric acid. In manufactories, the glass retort is replaced by a cast-iron cylinder (figs. 181 and 182) closed in by two flat plates, which are adjusted by means of bolts. Two of these cylinders are arranged alongside of each other in the same furnace, so that both ends are in a line with the front wall of the furnace. The anterior end has, toward the top, a tube d (fig. 182) introduced into a curved adapter by which the va- pours are led into the first three-mouthed stone-ware receiver. Two of these re- ceivers are placed side by side, each communicating with one of the two con- nected cylinders. These receivers also communicate with each other, by means of a curved tube of stone-ware uniting two of their mouths. Their third mouth corresponds with a series of two-mouthed receivers placed in a series.* Fig. 181. * Two-mouthed jars are more generally employed, the acid fumes merely passing over the surface of the water, where a large proportion is absorbed in the first jars, and the remainder is taken up before it leaves the series.—J. C. B. 138 NITROGEN. Fig .182. The back plate of the cylinders being removed, the proper quan- tity of saltpetre is introduced, and the plate replaced. The concen- trated sulphuric acid is poured in by a cast-iron funnel E (fig. 183), fitted to a tube c, which is then closed by a stopper of earthen-wrare. When the cylinders are charged, the joints are luted with clay, and they are heated as regularly as possible. The operation being terminated, the back plate of the cylinder is taken off, and the sulphate of potassa removed by iron scrapers. The acid condensed in the first receivers is necessarily impure, and contains a considerable quantity of sulphuric acid. This im- pure acid is used in the manufacture of sulphuric acid, as we shall see hereafter. The succeeding receivers contain the acid of com- merce. This acid is more or less concentrated: it contains a cer- tain quantity of nitrous acid, and frequently some chlorine, arising from the impurity of the nitre used in the process. The last re- ceivers contain a very weak acid. The receivers are not empty at the beginning of the operation. The first generally contains the very dilute acid solution formed in the last receivers of a preceding process, and which thus acquires the strength required in commerce. In the last, on the contrary, pure water is introduced, in order to obtain a complete condensa- tion of the nitrous vapours. § 106. The nitric acid of commerce is sufficiently pure for the greater part of the uses of the laboratory. We sometimes, how- ever, require a very pure acid, as in analytical researches. Now, as the acid of commerce generally contains some chlorine and sul- phuric acid, it may be purified by agitating it with a small quan- tity of a concentrated solution of nitrate of silver, and then dis- tilling it in a glass retort, in an apparatus resembling that of fig. 180.* § 107. Analysis of Nitric Acid.—In order to ascertain the Fig. 183. * A tolerably pure acid, t. e. free from chlorine, may be obtained by simply heating the strong acid gently, whereby the chlorine passes off together with some nitric acid.—J. C. B. NITRIC ACID. 139 quantity of nitric acid contained in an acid diluted with water, we proceed in the following manner:—We weigh accurately 10 grammes (about 150 grs.) of this acid in a glass flask containing about 200 cubic centimetres (12 cubic inches), and then add a certain quantity of water. We weigh, also very accurately, 100 grammes (1500 grs.) of very dry and finely powdered oxide of lead, and pour this oxide into the flask. The oxide of lead combines with the nitric acid, and the water becoflies free. The water can then be driven off by heat. The last operation demands some caution: the flask must be kept inclined, as in fig. 184, so that nothing may be projected without the vessel. When the matter appears dry, we continue the heat, and introduce as far as the centre of the flask a glass tube fas- Fig. 184. tened to the nozzle of a bellows. By blowing gently, the current of air drives off the last portions of the vapour of water. Care must he taken not to heat the flask too much, lest the nitrate of lead he decomposed, which is indicated by the appearance of reddish vapours. When the flask has cooled, it is weighed, and as we know the weight of the empty flask, we deduct from it the weight P of the oxide of lead and anhydrous nitric acid. P—100 is, there- fore, the quantity of anhydrous nitric acid contained in the 10 grammes of dilute acid. This process is founded on the circumstance of the oxide of lead being an anhydrous base, and the nitrate of lead not containing any water in combination. The weight of the oxide of lead added should also be greater than that which would form, with nitric acid, a neutral nitrate; for, otherwise, the nitric acid would not be entirely retained, and a portion would be volatilized. § 108. The composition of anhydrous nitric acid is determined as follows: We first begin by ascertaining the weight of nitric acid con- tained in a known weight of crystallized neutral nitrate of lead. To do this, we weigh exactly 10 grammes of the oxide of lead, and pour upon it a quantity of nitric acid, such that, after the complete transformation of the oxide of lead into a nitrate, there shall re- main an excess of free acid. It is evaporated and perfectly dried. This latter operation may be done in a small glass balloon (§ 107), by which the quantity of water contained in the hydrated acid is 140 NITROGEN. ascertained. The neutral nitrate of lead remains alone: it is weighed: let P be its weight; P—10 is therefore the weight of nitric acid contained in a weight P of neutral nitrate of lead. We thus find that 10 grammes of nitrate of lead contain Oxide of lead 6^.738 Nitric acid 3gm.262 10gm.000. We then take a tube ab (fig. 185) of very strong glass, of about 60 centimetres (24 inches) in length and 12 millimetres (| inch) in diameter, closed at one end: we place at the bottom about 10 grammes of bicarbonate of soda, and, above, a few centimetres in Fig. 185. length of metallic copper. Again, we weigh very exactly 10 grammes of nitrate of lead, which are introduced into the tube ab, immediately above the layer of metallic copper: and, lastly, the tube is filled with copper turnings. We fit to the open end a, by means of a cork, a curved tube acd, which plunges into a small mer- curial tub V, and arrange the tube ab over a sheet-iron furnace which allows us to heat its whole length. The tube ab is filled with air, which must be expelled. To effect this, Ave apply heat to the closed end of the tube; the bicarbonate of soda parts with a portion of its carbonic acid, which expels the air through the mercury. We can readily ascertain if the air is entirely driven out, by collecting some of the gas in a bell-glass, and observing if it is perfectly absorbed by a solution of potassa. If this absorption is complete, it is evident that the air has been entirely expelled and replaced by carbonic acid. We then remove the coals which heated the bicarbonate of soda, and heat to redness all the anterior part of the tube containing the metallic copper. We then place some coals near the part con- taining the nitrate of lead, so as to slowly decompose this salt, and collect the gases evolved in a large bell-glass C, over the mer- cury, to the top of which we have passed a certain quantity of a concentrated solution of potassa. The volatile products arising from the decomposition of the nitrate of lead pass over the heated PROTOXIDE OF NITROGEN. 141 copper, which seizes upon their oxygen, and the nitrogen alone reaches the bell-glass. When the nitrate of lead is entirely decomposed, the tube re- mains filled with nitrogen, which must be also driven into the bell- glass. For this purpose, we again heat the extremity b of the tube, which still contains some undecomposed bicarbonate of soda. This salt again gives off carbonic acid, which drives all the nitro- gen out of the tube. The carbonic acid which reaches the bell- glass at the same time with the nitrogen, is absorbed by the alka- line solution: so that, at the close of the experiment, we find in the glass all the nitrogen arising from the decomposition of 10 grammes of nitrate of lead. We then measure, exactly, the gas collected. To do this, we transfer it to a graduated bell-glass over the pneumatic cistern, and carefully measure its volume satu- rated with the vapour of water, after having levelled the water in the bell-glass with the general level of the cistern. Suppose V to represent the cubic centimetres occupied by the gas: t its temperature: / the elastic force of the vapour of water at t°: H the height of the barometer at the moment of measuring the gas. The number V0 of cubic centimetres, occupied by the gas at the temperature of 32°, and under the normal pressure of 0m.760, will be y =y l . E—f. Y O ’ • 1. 0. 00367. t 0.760 If this volume were air, it would weigh Y0. 0gm.001293. But, as it is nitrogen gas, which weighs less than air in the ratio of -9713, the weight of nitrogen will be 1.0000 ° ° p=Y0. 0gm.001293. 0.9713== Y0. 0gm.001256. We infer from this experiment that 10 grammes of nitrate of lead, or 3gm.262 of anhydrous nitric acid, contain 0gm.845 of ni- trogen. . We hence conclude that 100 of nitric acid contain Nitrogen 25.93 Oxygen 74.07 ioOo Or in volume, 1 volume of nitrogen, which weighs 0.9713 2J of oxygen 2.7640 making..... 3.7353 In fact, from the proportion 3.7353 of nitric acid : 0.9713 of nitrogen :: 100 of nitric acid : x, 142 NITROGEN. we find #=25.99, nearly the proportion of nitrogen found by ex- periment in 100 of nitric acid. All the other combinations of nitrogen with oxygen are easily obtained by the decomposition of nitric acid under given condi- tions. PROTOXIDE OF NITROGEN, NO. § 109. When nitric acid acts on a metal, the protoxide or deut- oxide of nitrogen is evolved, according to the nature of the metal. Zinc dissolves in dilute nitric' acid, disengaging a mixture of protoxide and deutoxide of nitrogen ; but if we allow this gase- ous mixture to remain some time in contact with damp zinc or iron filings, the deutoxide of nitrogen is decomposed and changed in protoxide, yielding a portion of its oxygen to the metal. The protoxide of nitrogen can be. much more readily prepared. We heat the nitrate of ammonia in a small glass retort (fig. 186), Fig. 186. provided with a curved tube: the substance at first melts, then boils, and disengages a large quantity of gas, which may he col- lected either over mercury or water. The retort must be gradu- ally heated, so as not to disengage the gas too rapidly. The nitrate of ammonia disappears and is changed into protoxide of nitrogen and water. The formula of nitrate of ammonia is NH3 H0,N05: by heat, it is changed into 2 equivalents of protoxide of nitrogen, 2NO, and 4 equivalents of water, 4HO. We have, in fact, NH3H0,N05=2N0+4H0. § 110. The protoxide of nitrogen is a colourless, inodorous, and tasteless gas, of a density of 1.527. It liquefies at 32° under a pressure of about 30 atmospheres. It solidifies at 148° below zero. It undergoes no change by contact with the air. An incandes- cent coal continues to burn in this gas with a bright light, as in oxygen. A taper having some burning points is rekindled when PROTOXIDE OF NITROGEN. 143 plunged into the protoxide of nitrogen, and burns with a very brilliant flame. This property, distinctive of oxygen, may cause it to be confounded with the protoxide of nitrogen. Sulphur, if burning feebly, is extinguished when plunged into a vessel filled with protoxide of nitrogen: but when the burning surface is of some extent, its combustion is very rapid. Phosphorus burns in the protoxide of nitrogen with a very bril- liant white light. We shall not be surprised that the combustion of substances is more energetic in the protoxide of nitrogen than in atmospheric air, when we remember that the one-half of the volume of the former, and only one-fifth of that of the latter is oxygen. But, in atmospheric air, the oxygen and nitrogen are merely mixed, whilst, in the protoxide of nitrogen, they are combined: the combustible body must therefore be in conditions under which it can destroy this combination; and that it may continue to burn in protoxide of nitrogen, its temperature must generally be elevated. We have seen that atmospheric air supported animal life only from the oxygen it contains. The phenomenon of respiration ap- pears to consist essentially in a sort of combustion of the organic matters by oxygen, a combustion which evolves carbonic acid and vapour of water. The essential functions of respiration can be carried on equally well in an atmosphere of protoxide of nitrogen ; for many animals can live several hours in this gas. However, a prolonged continuance in this gas will give rise to disturbance sufficient to produce death. The protoxide of nitrogen, inhaled by man, produces a species of intoxication, accompanied, it is said, by agreeable sensations. It was found to possess this property at an early period of its dis- covery, and hence received the name of exhilarating gas. When this experiment is made, the gas should be perfectly pure, as it often contains some chlorine, which would violently affect the respiratory organs. The chlorine is owing to the nitrate of am- monia sometimes containing small portions of chlorohydrate of ammonia. We have said that the protoxide of nitrogen liquefied at 32° under a pressure of 30 atmospheres. The liquid protoxide of nitrogen may be obtained by compressing the gas in a strong metallic reservoir surrounded by ice, by means of an air-pump. By opening the stopcock of the reservoir, after having inserted it, a portion of the liquid reassumes the gaseous state, but cools the rest to such a degree that it does not volatilize, and even assumes, in part, the solid state, forming a white snow. The liquid part may be col- lected in a tube, and kept in this state for more than half an hour. When a metal is plunged into this fluid, it produces a noise similar to that resulting from the immersion of red-hot iron in 144 NITROGEN. water. Mercury produces the same effect, and rapidly congeals, forming a metal resembling silver in its physical properties. Potassium, which rapidly decomposes the gaseous protoxide of nitrogen under the influence of heat, is not changed by the con- tact of the liquid protoxide. Carbon, sulphur, phosphorus, and iodine belong to the same category. The temperature of liquid protoxide of nitrogen at the ordinary atmospheric pressure is very low, and supposed to be —148°. It descends much lower when placed beneath the receiver of an air-pump, which is rapidly ex- hausted, a portion of it then congealing into a white snow. If we place in the protoxide, which is evaporated in the vacuum of an air-pump, a small tube hermetically sealed and containing some liquid protoxide, the latter freezes and forms a perfectly limpid solid mass. § 111. The protoxide of nitrogen is easily analyzed as follows : A certain given volume of gas is measured in a graduated glass, placed over mercury, and introduced into a tube curved as in fig. 187. A piece of potassium, fastened to a wire, is passed into the curve of the tube, and heated by an alcohol lamp. Energetic com- bustion ensues, the potassium decom- - poses, the protoxide of nitrogen seizes upon its oxygen, and sets free the nitrogen. At the moment of decom- position, the glass must be firmly held in the hand, lest it might be projected from the cistern. When the tube has cooled, the gas is again passed into the graduated glass, and its volume will be found to be unchanged by decomposition. We hence conclude that the protoxide of nitrogen contains exactly its volume of nitrogen. If we deduct from the weight of a volume 1 of the protoxide of nitrogen or from the density of this gas =1.527 the weight of a volume 1 of nitrogen or its density =0.972 there remain 0.555 very nearly equal to or 0.5528, or the half of the density of oxygen gas. 1 volume of protoxide of nitrogen therefore contains 1 volume of nitrogen 0.972 J “ of oxygen 0.552 17524 If we make the proportion 1.524 : 0.972:: 100 : x, Fig. 187. DEUTOXIDE OF NITROGEN. 145 x will be the weight of the nitrogen contained in 100 grammes of protoxide of nitrogen: we then have Nitrogen 63.77 Oxygen 36.23 imoo § 112. The analysis of the protoxide of nitrogen may also be made in the 'eudiometer, by means of hydrogen gas. Suppose that we have introduced into the eudiometer 100 measures of protoxide of nitrogen 150 “ of hydrogen Total... 250 Let us pass an electric spark through, and again measure the volume of gas : we shall find it reduced to 150 measures; 100 have therefore disappeared. If the nitrogen and oxygen were merely mixed, instead of being combined with condensation, we might deduce the composition of the gas from the volume which has dis- appeared : but that is impossible, and we must ascertain directly the quantity of hydrogen which has served to burn the oxygen of the protoxide. This quantity will be known, if we know how much hydrogen remains in the 150 divisions of gas after the explosion. To ascertain this volume, we will introduce into the eudiometer 50 divisions of oxygen, making in all 200, and pass an electric spark. After the explosion there remain only 125 divisions of gas; 75 have therefore disappeared, formed of hydrogen and oxygen in the proportions constituting water, that is, 50 of hydrogen and 25 of oxygen. Thus, in the 150 parts of gas which remained after the first electric spark, there were 50 parts of hydrogen, and consequently 100 parts of nitrogen. Now, as we have introduced, from the first, 150 parts of hydrogen, and only find 50, 100 parts have been burned by the oxygen of the protoxide of nitrogen: 100 parts of this gas, therefore, contain 100 parts of nitrogen 50 “ of oxygen. § 113. This compound is obtained by dissolving metals in nitric acid properly diluted. We generally use copper or mercury. Copper affords pure deutoxide of nitrogen, provided the tempera- ture he not allowed to rise too high during the reaction, and the acid be sufficiently diluted. The operation is effected in the same apparatus as that used for making hydrogen gas. Copper turnings are placed in the bot- tom of a two-mouthed bottle A (fig. 188), and covered with a DEUTOXIDE OF NITROGEN, NOa. • 146 NITROGEN. layer of water. A discharging- tube is fitted to one of the mouths a, and to the other l a straight tube terminating in a funnel, acting as a safety-tube, and through which the nitric acid is slowly and gradually added. The gas may be col- lected over mercury or water. Water dissolves fa of its volume. Very pure deutoxide of nitro- gen may be obtained by heating the nitrate of potassa KO,NOs with a solution of the protochloride of iron FeCl, in an excess of chlorohydric acid. Fig. 188. 6FeCl+K0,N05+4HCl=N0a+3(Fe,Cy+KCl+4H0. To make this preparation, we take two equal volumes of chloro- hydric acid: we heat one with iron filings, to change it into pro- tochloride of iron, and add it to the other volume of acid. The nitrate of potassa is then treated with this mixture. § 114. The deutoxide of nitrogen is a colourless gas which has hitherto borne the greatest degree of pressure without liquefaction. Its density is 1.039. When mixed with the air, it immediately gives off reddish va- pours, absorbing, in this case, oxygen, and changing into hypo- nitric acid: the vapours have a strongly acid reaction. The deutoxide of nitrogen has of itself no acid reaction, as is easily shown by the following experiment. We collect some deut- oxide of nitrogen in a glass, over mercury, and pass into the glass some tincture of litmus, which preserves its blue colour. But, if we introduce some bubbles of oxygen, the tincture is red- dened immediately. A taper presenting some points of ignition does not inflame when plunged into this gas ; but an incandescent coal burns with great brilliancy. Phosphorus may be melted in deutoxide of nitrogen without taking fire; whilst, in the air, this always happens. But inflamed phosphorus continues to burn longer and with more brilliancy in this gas than in the open air. The light may be compared to that of burning phosphorus in oxygen. Burning sulphur is extinguished in the deutoxide of nitrogen. The deutoxide of nitrogen is therefore a less active agent of combustion than the protoxide; and yet, for the same quantity of nitrogen, it contains twice as much oxygen. This shows that the nitrogen and oxygen are combined with much more energy in DEUTOXIDE OF NITROGEN. 147 the deutoxide than in the protoxide, since more powerful affinities are necessary to effect the decomposition. The deutoxide of nitrogen is absorbed by a solution of the sul- phate of the protoxide of iron, and then acquires a very deep brown-colour. This reaction may be used to separate the prot- oxide from the deutoxide of nitrogen. The deutoxide of nitrogen is largely dissolved in concentrated nitric acid, but there is a reciprocal decomposition: the deutoxide takes from the nitric acid a portion of its oxygen, and the two substances pass into the state of hyponitric acid. The liquid assumes a brown hue, which is deeper and deeper in proportion to the forma- tion of hyponitric acid. When nitric acid is more diluted with water, it is more fixed, and a smaller quantity of acid decomposes. Lastly, when nitric acid is greatly diluted, it is no longer decomposed by the deutoxide of nitrogen. These solutions of hyponitric acid in nitric acid more or less concentrated present very various colours. With monohydrated nitric acid, the liquid is brown: with a more diluted acid, it is yellow. Acid of a density of 1.35 is green; that of 1.25 becomes clear blue; and, lastly, acid of a density less than 1.15 is colour- less. This experiment is generally made in the following manner. To a large two-mouthed bottle (fig. 189), in which the deutoxide of nitrogen is produced, a series of three-mouthed bottles are fitted, Fig. 189. and arranged as in the wood-cut. This apparatus has received the name of Woolf's bottles.* In the first two bottles we place the * We frequently adopt, in our chemical apparatus, tubes arranged in a peculiar manner, and called safety-tubes. Their objeet is to prevent explosion and the mixture of the fluids contained in the various vessels composing it. Theory of the safety-tubes.—Let us suppose a flask A (fig. 190) in which there is an evolution of chlorine gas, by the reaction of chlorohydric acid on the peroxide of manganese: let B be a test-glass filled with a solution of potassa, on which we desire the chlorine to act, and for which it has a great affinity. We lead the chlorine by the discharging-tube abc to the bottom of the test-glass B. As long as the chlorine is freely furnished by the flask A, the operation goes on regularly, and bubbles of gas pass through the solution of potassa. The elastic force of the gas, in the balloon A, equals the pressure of the external atmosphere on the solution uf potassa, increased by the pressure of a column of 148 NITROGEN. most concentrated nitric acid; in the third, nitric acid more diluted, having a density of 1.45 ; in the fourth, acid of 1.35; in the fifth, acid of 1.25; and lastly, in the sixth, acid of 1.10. The first bottle is originally of a brown colour; but, as the solution of potassa equal in height to the distance between the level of the fluid in the test-glass and the extremity c of the discharging-tube. The pressure of the external atmosphere is measured by the height H of a column of mercury which is in equilibrium with it, or in other words, by the height of the mer- cury in the barometer. The pressure of the column a' of a solution of potassa may be expressed by a co- lumn of mercury which would produce an equivalent pressure. If we designate by x the height of this column, by d! and $ the densities, compared with water, of the solution of potassa and of the mercury, we shall evidently have 6! xS=a'd' whence x=.a'—. The elastic force of the gas in the interior of the flask will be therefore ex- d' pressed by a column of mercury of which the height is II-f-a' y. Let us suppose that the evolution of chlorine ceases in the flask A, either because the quantity of chlorohydric acid is exhausted, or because the flask has become too cool. The solution of potassa contained in the discharging-tube be, continues to absorb the chlorine contained in the flask A: the elastic force of the gas in the apparatus gradually diminishes, and the constant pressure of the atmosphere on the fluid in the test-glass will drive this fluid into the tube be. If the operator be present, he may save the experiment by quickly uncork- ing the flask A; but if he be absent, the solution of potassa will soon ascend to the top of the discharging-tube, and, the absorption of the chlorine by the potassa continuing, the greater part of the solution of potassa may pass into the flask A. Absorption is then said to have taken place, and the experiment fails. It is impossible that an accident of this nature can occur, if we adjust to the flask A, and in the same cork, a curved tube efg having a bulb u, as represented in fig. 191. Into this tube is poured a small quantity of the same fluid as that in the flask, which in this experiment would be chlorohydric acid. When the operation goes on regularly, and the gases are disen- gaged at the extremity c of the discharging-tube, the elastic force of the internal gas is represented by H-j-a' The chlorohydric acid will therefore as- s cend in the leg fg of the curved tube, until the column, elevated above the level of the fluid in the bulb u, the height of which we represent by h, equals the elastic force of W-\-a'lL ’ diminished by the pressure H of the atmosphere, for this latter pressure is also exerted on the top of the column h. If d represent the density of the chlorohydric acid compared with water, a column of mercury exerting the same pressure as the column h of chlorohydric acid would be expressed by li—. We should therefore have t ,, ,, h —=a —- whence h=a'—.. Fig. 190. Fig. 191. DEUTOXIDE OF NITROGEN. 149 deutoxide of nitrogen constantly carries water which condenses in this first bottle, the acid it contains changes colour successively. The second bottle assumes a brown tinge; the third becomes yel- low ; the fourth, blue; the sixth is colourless. Let us now suppose that the disengagement of the gas ceases, and that, in consequence of the absorption of the chlorine by the solution of potassa, the elastic force of the gas in the flask A becomes less than that of the atmosphere: it will be seen, that if the various parts of the apparatus are properly propor- tioned, no absorption of the solution of potassa into the flask A is to be feared. In fact, as the elastic force of the gas in this flask becomes less than that of the atmosphere, the solution of potassa will rise in the tube be ; but, at the same time, the chlorohydric acid will descend in the leg fg of the curved tube. If the fluid reaches the lowest point / before the solution of potassa reaches the sum- mit b of the discharging-tube, the atmospheric air will enter by the curved tube and prevent the interior elastic force from being weakened. Absorption will therefore be impossible, and the operation cannot fail of success. The bulb u of the curved tube is intended to prevent, by its great relative capa- city, the level of the fluid from rising high in leg fe, in consequence of the intro- duction of the fluid previously contained in the leg fg: the air therefore passes into the apparatus, where the internal elastic force has become very slightly in- ferior to that of the atmosphere. This bulb is also useful, because it contains the quantity of fluid necessary to completely fill the leg./y, when the elastic force of the internal gas becomes much greater than that of the atmosphere. More- over, this elastic force cannot increase in- definitely : it cannot surpass the pressure of the external atmosphere by a quantity greater than that which balances the fluid column contained in the leg fg: for then this column would be projected from the tube, and the internal gas would commu- nicate freely with the atmosphere. This last circumstance frequently oc- curs in the experiment we have selected as an example. The tube be is often closed by the deposit of crystallized matter formed by the reaction of the chlorine on the potassa. The gas continuing to be disengaged in the flask A, its elastic force continually increasing, if it find no other exit, as in fig. 190, this force will soon be sufficient to burst the flask. The addition of the curved tube removes the danger: and it is therefore with great justice called a safety-tube. This tube has still another use. It enables us to add, as required, portions of hydrochloric acid, without un- corking the flask. When the vessel in which the gas is generated is a retort with but one opening, we use a discharging-tube, to which is attached a tube shaped like the letter S, as in fig. 192. This tube then acts as a safety-tube only; it cannot be used for the introduction of the fluid necessary for reaction. This arrangement is called Welter's tube, from its inventor. We can place in this tube any liquid which exerts no chemical action on the gas. The vessel used for chemical reaction is often a two- mouthed bottle (fig. 193), as in the preparation of hydro- gen gas and the deutoxide of nitrogen. We then use, Fig. 192. Fig. 193. 150 NITROGEN. § 115. The analysis of the deutoxide of nitrogen is made by potassium in a curved bell-glass, in the same way as that of the protoxide. After the decomposition, we find the volume of gas reduced by one-half. Thus 1 volume of deutoxide contains a J volume of nitrogen. as a safety-tube, a simple straight tube, surmounted by a funnel, and its lower end passing below the surface of the liquid. Let us now suppose that we wish to pass the same gas, successively, through a series of bottles containing solutions, different or identical, which can absorb it. We use the arrangement represented in fig. 194. A is the flask in which the Fig. 194. chlorine is generated, and the three-mouthed bottles B, C, D, E, contain the solu- tions intended to absorb the gas. Let us suppose that the evolution of the chlorine be such that the bubbles of gas traverse the fluid of the four bottles, and let us inquire what is the elastic force possessed by the gas in each of these bottles. The pressure of the atmosphere is freely exerted, by the tube o open at both ends, on the surface of the fluid contained in the bottle E. The gas in the bottle E has therefore an elastic force equal to that of the external atmosphere, which we suppose represented by a column H of mercury. In the bottle D, the pressure on the surface of the fluid is equal to the pressure H of the bottle E, increased by a column of mercury balancing the column a!'" of the liquid E, which the gas should depress in the discharging-tube, to escape by the opening c"". If d"" represent the density, compared with water, of the fluid E, the column of mercury balancing the column a"" of the liquid E is expressed d"" by a"" The elastic force of the gas in the bottle D is therefore expressed by S d"" H 4-a"" iP In the bottle C, the pressure on the surface of the fluid is equal to the pressure d"" II-|-a"" of the gas in the bottle D, augmented by the column of mercury balancing the column a"" of the fluid D, which the gas must depress to pass from C into D. This column of mercury is expressed by a"' ~, if d'" be the density of the fluid D. The elastic force of the gas in the bottle C is therefore d'" d"" H-j-a'" ——}-a"" —• 0 0 The pressure on the surface of the fluid in the bottle B is equal to the pressure d’” d"” H-f-a'"— of the gas in the bottle C, increased by the column of mer- cury balancing the column a” of the fluid C. This column of mercury is expressed DEUTOXIDE OF NITROGEN. 151 Deducting from the density of the deutoxide = 1.039 0 Q79 one-half the density of nitrogen — 0.486 2 there remain 0.553 d" by a"—, if d" represent the density of the fluid C compared with water. The elastic force of the atmosphere of the bottle B is therefore Lastly, the elastic force of the gas in the flask A, is equal to the elastic force d" d!" d”" of the bottle B, that is to say H-j-a"—-j-a'"——, increased by a column of mercury which balances the column a' of the fluid B. This column of mer- cury is expressed by a,'—, a' being the density of the fluid B, compared with J* water. The elastic force of the gas in the flask A is therefore expressed by d’ d" d!" d"" H+a'7+«"7-K"T+«""T- Thus, when the gas passes freely through the fluids of the bottles B, C, D, and E, we shall have a3 the elastic forces of the gases : In E H D H + a"" — S Jt" jmt C H 4- a'" i 6 vols. chlorohydric > acid Chlorohydrate > 0f • ammonia. 6 yols. ammonia 4NH3+3C1=3(NH3,HC1)+N. This experiment is made in the following manner: Pour into a 168 NITROGEN. long tube closed at one end, a solution of chlorine in water, so as to .fill of the tube, and completely fill it with a solution of am- monia. Close the opening of the tube with the finger, and invert it. The solution of ammonia, which is lighter, ascends in the tube, and bubbles of nitrogen are immediately disengaged. Advantage is sometimes taken of this reaction in the laboratory, in the pre- paration of this gas: we have described the process previously (§ 92). Ammonia presents, in contact with many substances, very curious reactions, which are, in general, too complex to be mentioned here. We shall return to them hereafter. ADDITION TO THE COMBINATIONS OF NITROGEN WITH OXYGEN. Anhydrous Nitric Acid, NOs. § 124 bis. We said (§101) that anhydrous nitric acid had not hitherto been procured in an isolated state. This important sub- stance has just been obtained, and is prepared by treating very dry nitrate of silver heated to 122° or 140° with chlorine: this com- pound is changed into chloride of silver, and white prismatic crys- tals of anhydrous nitric acid are deposited on the cold sides of the apparatus. The oxygen of the oxide of silver is disengaged, as well as nitrous vapours and the oxygen arising from the decom- position of a portion of the nitric acid. Anhydrous nitric acid melts at 85°, and boils at about 115.° At a temperature slightly elevated above its boiling point, it de- composes into oxygen and hyponitric acid. 169 SULPHUR. • Equivalent S=16 (200 0=100). § 125. Sulphur is a substance found very abundantly in nature, sometimes isolated, and sometimes in combination with a great number of metals. Sulphur, when alone, is sometimes found per- fectly pure and in very regular crystals : but most frequently it is intimately mixed with earthy substances. We shall soon see how it is separated from these substances and obtained in the two states found in commerce. Sulphur may be obtained in three states. At the ordinary tem- perature it is solid: if heated above 232°, it melts, and furnishes a very limpid canary-yellow fluid : the pieces of unmelted sulphur remain at the bottom of the vessel, thus proving that sulphur in- creases in bulk or dilates in passing from the solid to the liquid state. Water presents the contrary phenomenon, ice being lighter than water: the latter, passing from the solid to the liquid state, contracts instead of dilating. Sulphur passes suddenly from the liquid to the solid state, without becoming doughy, and is thus most favourable for crystallization. The phenomenon of crystalliza- tion of sulphur may be readily observed in a glass tube. When the temperature falls to about 232°, the particles of sulphur, in solidifying, will be seen to form needles, which start from one side and shoot through the liquid mass. New needles are implanted in the crystals already formed, and so on, until the whole mass is solidified. If we do not wait for the completion of the solidifica- tion, and pierce the solid crust which forms on the surface, the remainder of the fluid may be poured off, and the crystals exposed. We may, in this way, obtain beautiful crystals by melting a quan- tity of sulphur in an earthen vessel, and cooling it very slowly. When a solid crust has formed on the surface, it is perforated, and the balance of the liquid poured oif. When the mass has become cold, we separate carefully the upper crust, keeping the vessel in- verted, for fear of injuring the crystals which line the sides. These crystals are long brilliant prisms, of the same colour as the liquid sulphur. The predominating form of the crystals of sulphur obtained by fusion is an oblique prism with a rhombic base, in which the prin- cipal axis is inclined at an angle of 85° 54' to the base, and the obtuse angle of the base is 90° 32'. This form belongs to the fifth system of crystallization. Sulphur may be crystallized at a low temperature, by dissolving 170 SULPHUR. it in a volatile liquid. The sulphuret of carbon is the most appro- priate. If we expose a solution of sulphur in sulphuret of carbon to the air, the liquid evaporates rapidly, and the sulphur, not finding sufficient sulphuret of carbon to keep it in solution, is slowly deposited, in the midst of the liquid, in regular crystals, differing totally from those which form in melted sulphur. We remarked upon this point in the introduction (§ 39). Sulphur, crystallized by solution, presents exactly the same form and appearance as the natural article sometimes found in very large and perfectly pure crystals. The most ordinary form of these crystals is that of fig. 70 : the predominating form is a right octahedron with a rhombic base, belonging to the fourth system of crystallization (fig. 66). The fracture of these crystals is vitre- ous and conchoidal. Their density is 2.07. The crystals which are deposited in the melted sulphur are transparent and slightly elastic ; but they soon, lose these proper- ties and become opaque and friable. They are then of a clearer yellow. We assigned the cause of this change (§ 39). It sometimes happens, however, that by dissolving in sulphuret of carbon sulphur which has been recently melted, the liquid left to spontaneous evaporation deposits at the same time crystals belong- ing to both systems. It is easy to distinguish the right octahe- drons with rhombic bases, which generally predominate, and the oblique prisms with rhombic bases. The mixture becomes much more evident when we leave the crystals untouched for several days. The octahedric crystals remain transparent and preserve their colour, whilst the oblique prismatic crystals become opaque, friable, and of a yellow straw-colour. We cannot here (as we have done in § 39) explain the dimorphism of the sulphur by the dif- ference of temperature at which crystallization took place, since the two incompatible forms have been developed in the same me- dium. It is probable that these two forms have some reference to the two states of ordinary sulphur and soft sulphur, of which we are about to speak; for, by dissolving soft sulphur in sulphuret of carbon, we obtain a greater number of prismatic mingled with the octahedral crystals. Melted sulphur is perfectly limpid, and of a clear yellow: if it is further heated, its colour becomes deeper, and at the same time it loses its fluidity. At 320° it flows with difficulty, and its colour passes from yellow to browrn. At 392° it is so tenacious that the vessel containing it may be inverted without its escaping: its colour is then of a deep brown. If the temperature is carried still higher, the sulphur recovers its fluidity, still preserving its brown colour. Lastly, at 750° it boils, and may be distilled. Distillation is performed in a glass retort, furnished with a receiver. The sul- phur is introduced into the retort, which is heated by coals. The sulphur at first melts, and then passes through the various stages SULPHUR. 171 we have indicated, and lastly boils. The vapour is driven into the beak of the retort, when it first condenses in the form of a very fine powder, which is called flowers of sulphur. But the distilla- tion continuing, the temperature rises in the beak, and soon exceeds 232°, the degree at which sulphur melts, and the vapours condense only in the fluid state. If the sulphur subjected to distillation contains non-volatile foreign substances, they remain in the retort. The colour of the vapour of sulphur is yellowish-brown, and its density has been found to be 6.654. If we heat sulphur in a crucible to a temperature higher than 392°, and then pour it out, in a small stream, into an earthen vessel filled with cold water, wTe obtain a spongy, brown, soft and elastic mass, which retains its softness for some time: it then gets harder, and in a few days the sulphur assumes its ordinary hard- ness, but its colour continues deeper. Soft sulphur becomes hard in a few minutes if, instead of allowing it to remain at the ordinary temperature, it is heated to about 212° : the change ensues sud- denly, with a spontaneous evolution of heat: for soft sulphur heated to 212°, elevates its temperature to 230°. Sulphur is a combustible substance, and burns with a bluish flame, giving out the well-known suffocating smell, which needs no description. The sulphur combines with the oxygen of the air, and gives rise to a gaseous compound, sulphurous acid gas. § 126. Native sulphur is frequently found in volcanic countries: it impregnates the ashes of certain extinguished craters called sol- faterras. But it is principally found in irregular masses in the midst of the bituminous marls, strata of gypsum, and lime, found in the chalky formation. The mines of Sicily, which are of this cha- racter, are the most important in the world, and furnish nearly all the sulphur consumed in the arts. This native sulphur is merely mixed with earthy matter. The rich ores are heated in large kettles until the sulphur melts: the earthy matter falls to the bottom of the kettle: the sulphur is taken out with ladles and poured into sheet-iron vessels, from which, when cool, it is easily removed. It is thus exported, under the name of crude sulphur. The earthy residuum taken from the kettle is then subjected, with the poorer ores, to distillation. The sulphurous earth is placed in earthen pots (fig. 202) holding about 5 gallons. At the Fig. 202. 172 SULPHUR. upper part of these pots is an opening, which is closed during the operation, and by which the pots are charged and emptied. A bent earthen tube conveys the distilled sulphur into other similarly shaped pots, which act as receivers. At the bottom of these pots is an aperture, which is occasionally opened to allow the sulphur to run into buckets of water. The pots containing the sulphurous earth are placed in two rows on a long furnace, called a galley furnace. This first distillation is very imperfectly performed ; the sulphur thus obtained containing 10 or 15 per cent, of earthy matter. Hence it is still called in commerce crude sulphur. The crude sulphur is subjected to a second distillation conducted with great care. The apparatus (fig. 203) in which it is per- Fig. 203, formed, consists of a cast-iron kettle CD, which acts as a retort, and a large chamber of brick-work, serving as a receiver. The kettle is placed over a furnace, at K. Formerly, a cast-iron door served to charge the kettle with sulphur, and to extract the resi- duum, but, in modern times, the distillation is continuous. The va- pour of sulphur which rises from the kettle, is conveyed by the pipe D into the chamber A, where it condenses in the form of very fine powder, which is the flowers of sulphur. The chamber is furnished with valves s which allow the internal heated air to escape, and prevent the entrance of the external air. SULPHUR. 173 With this apparatus, we may obtain either flowers of sulphur or roll sulphur. The vapour of sulphur, by condensing, heats the chamber, which soon attains a temperature greater than 232°, so that the sulphur cannot condense in the solid state, but remains liquid on the floor of the chamber. If, therefore, we wish to ob- tain flowers of sulphur, the chamber should be made as large as pos- sible, and the distillation be occasionally suspended, in order to allow the walls to cool. If, on the contrary, we wish to obtain liquid sul- phur, we operate with a small room, and do not interrupt the process. In order to charge the kettle C, it was formerly necessary to remove the door, which was a serious inconvenience, and frequently caused explosions by the mixture of the atmospheric air with the highly-heated vapour of sulphur. This danger is now avoided by placing outside of the furnace a second kettle M, which is heated by the hot air of the furnace before it reaches the chimney. This kettle communicates with the first by means of a pipe v. The kettle M is charged with crude sulphur, which melts in it and deposits a portion of the foreign matter, so that the sulphur enters the kettle C already purified by a kind of decantation. The flowers of sulphur are withdrawn from the chamber, after the operation, through a lateral door. The melted sulphur flows out hv a small gutter r (fig. 203), of which the aperture o is closed by a stopper: it is received in moulds of pine wood (fig. 204), moistened but well wiped, and takes the shape of conical sticks: this is the stick or roll sulphur of commerce. By cooling in the moulds, the sulphur first crystallizes at its periphery, and then gradually toward its axis: it also un- dergoes some contraction, as is manifested by the cavity filled with confused needles, which are always seen in the end of the stick occupying the upper part of the mould. Flowers of sulphur almost always exhibit a slight acid reaction with litmus paper, which is due to the presence of a very small quantity of sulphuric acid, and can be removed by repeated washings. Fig. 204. COMBINATIONS OF SULPHUR WITH OXYGEN. § 127. Sulphur forms a great number of compounds with oxygen. Seven of these are now well ascertained, and they are all acid, viz.: 1. Hyposulphurous acid SaOa* 2. Trisulpho-hyposulphuric acid SsOs 3. Bisulpho-hyposulphuric acid S405 4. Monosulpho-hyposulphuric acid S3Os 5. Sulphurous acid S03 * The names given by Berzelius to the above are, for 1, Dithionous; 2, Penta- thionic; 3, Tetrathionic; 4, Trithionic ; and 6, Dithionic acids.—J. C. B. 174 SULPHUR. 6. Ilyposulphuric acid Ss05 T. Sulphuric acid S03 We shall commence with the study of sulphurous acid, because this substance is used in the preparation of nearly all the other compounds of sulphur and oxygen. Sulphurous Acid, S02. § 128. Sulphurous acid is formed when sulphur is burned in oxygen or in the open air. In the laboratory, several processes are used in its preparation. A mixture of 6 parts of powdered peroxide of manganese, and 1 part of flowers of sulphur, is heated in a small glass retort (fig. 205): the sulphurous acid gas is made to traverse a small washing-bottle, which retains a small quantity of the sul- phur volatilized by the heat and carried over by the current of gas. In this experiment, the sulphur burns at the expense of a portion of the oxygen of the peroxide of manganese: sul- phurous acid gas, which is the product of combustion, is disengaged, and the protoxide of manganese remains in the retort. Sulphurous acid is also obtained by decomposing sulphuric acid by a metal which removes a portion of its oxygen, but which does not decompose water when in contact with active acids. Mercury or copper is used for this purpose. The more oxidable metals, such as iron or zinc, would at the same time decompose the water always contained in concentrated sulphuric acid, and sulphurous acid gas and hydrogen would be disen- gaged together. The mer- cury or copper-turnings are placed in a flask (fig. 206), concentrated sulphuric acid added, and heated with a few coals. The gas must be pass- ed through a washing-bottle containing some water, which absorbs the vapour of sulphuric acid. If we wish to obtain the gas perfectly dry, wre adapt to the bottle a tube containing chloride of calcium. This gas must be collected over mercury, as it is very soluble in water. § 129. Sulphurous acid is a colourless gas, having a smell re- Fig. 205. Fig. 206. SULPHUROUS ACID. 175 sembling that of burning sulphur. It acts energetically on the respiratory organs, exciting cough and a sensation of suffocation. Its effects are not dangerous when breathed in small quantities. Its density is 2.24T. Sulphurous gas liquefies, under the ordinary pressure, at about 14°. It is therefore easy to prepare it in a liquid state in the laboratory, it being merely necessary to pass the gas well dried through a bulb A (fig. 207), placed in a refrigerating mixture of ice and sea- salt, or of ice and chloride of calcium. When the bulb is sufficiently filled, the ends a and b of the tube are closed by the blowpipe. If we prefer preserving the liquid sulphurous acid in glass tubes, we take tubes closed at one end and elongated in the middle, as in fig. 208 ; the upper part A forming the funnel. The acid being poured into this funnel, the first drop which enters the cavity B volatilizes and expels the air, so that if we afterward plunge B into the refrigerating mixture, the vapours of sulphurous acid are condensed in it, and it is filled with liquid acid. The tube is filled to three-fourths of its capacity, and then sealed by the blowpipe at Chlorohydric acid. Sulphuric Acid, S03. §133. We have seen (§ 131) that sulphurous acid dissolved in water absorbed oxygen from the air, and was changed into sul- phuric acid. This transformation is easily effected by energetic oxidizing substances, such as concentrated nitric acid. If a current of sulphurous acid gas be passed through concentrated nitric acid heated to ebullition, the sulphurous is entirely condensed in the form of sulphuric acid, and the nitric passes into the state of hyponitric acid. Sulphuric acid may be also obtained by heating sulphur with nitric acid; but a long time is required to completely oxidize the sulphur. By these two processes, we obtain a mixture of sulphuric acid, nitric acid, and water. This mixture is distilled in a glass retort: nitric acid more or less mixed with water passes over at first; the temperature in the retort gradually rises, and reaches at last 617°. It then remains stationary, and there passes over a very acid homogeneous fluid, composed of sulphuric acid and water, which is called concentrated sulphuric acid ox oil of vitriol. We shall now study its properties. § 134. Concentrated sulphuric acid is a fluid of oleaginous con- sistence, the density of which at 59° is 1.843 ; it boils at 617°. It SULPHURIC ACID. 179 is inodorous. The tension of its vapour is not appreciable at the ordinary temperature; and, in fact, we may allow two saucers to remain for several days under the receiver of an air-pump, the one containing concentrated sulphuric acid, and the other a solution of chloride of barium, without the latter becoming cloudy. Now, if sulphuric acid gave olf any sensible vapour, this vapour, coming in contact with the solution of the chloride of barium, would decom- pose it, and form an insoluble sulphate of baryta, which would precipitate in the form of a white powder. Concentrated sulphuric acid congeals at —30°. Sulphuric is one of the most powerful acids known, for it deeply reddens litmus, even when diluted with 1000 times its weight of water. Assisted by heat, it expels the majority of the acids from their compounds. This last property depends both on the activity of the acid and its property of boiling only at a very high tem- perature ; and it is from this latter reason that sulphuric acid, when heated, drives off the chlorohydric and nitric acids. But it is again expelled, under the influence of heat, by phosphoric and boracic acids. These acids are nevertheless weaker than sulphuric acid at the ordinary temperature ; but they are also much less volatile. The distillation of concentrated sulphuric acid in a glass retort is a dangerous operation, on account of the commotion produced in the liquid when boiling, and which is sometimes sufficient to elevate the retort, and thus render it liable to break when falling down again on its support. The ebullition becomes more regular if bits of platinum wire be introduced into the retort. The bubbles of vapour are then disengaged not on the sides of the vessel, but at the ends of the wire. However, the ‘distillation of sulphuric acid can only be safely carried on in glass retorts, by heating the liquid on the sides and not from the bottom. The retort is placed in a circular wire-furnace, as in fig. 210, the coals being arranged Fig. 210. around it, and the bottom remaining free. In order to prevent the vapours from condensing on the top of the retort, it is covered with a sheet-iron hood A, which rests on the furnace, and is cut 180 SULniUR. out to allow the passage of the beak of the retort. The ebullition of the fluid then takes place against the sides of the retort, without any very violent commotion. Concentrated sulphuric acid has a powerful affinity for water. It readily absorbs the vapour of water contained in the air, and hence its frequent use in the laboratory to dry gases. Its affinity for water is such, that it frequently causes the formation of water in organic substances, at the expense of the oxygen and hydrogen they contain. In this way it carbonizes the corks of the bottles in which it may be kept. Cork, like the greater part of vegetable substances, is a compound of carbon, hydrogen, and oxygen. Con- centrated sulphuric acid removes a portion of the hydrogen and oxygen to form water, with which it unites, and the carbon forms with the balance of the hydrogen and oxygen a brown substance, giving the cork the appearance of having been charred. When concentrated sulphuric acid is poured into water, the acid falls through the liquid, like a syrup, to the bottom of the vessel, and forms a distinct layer, which slowly dissolves in the superna- tant water; but, if the fluids are agitated, they dissolve immedi- ately with a great evolution of heat. It is dangerous to pour water into concentrated sulphuric acid. A portion of the water, uniting with the acid, disengages a great quantity of heat, which may instantly convert the remainder of the water into vapour, and consequently project the acid from the vessel. We should therefore mix these liquids by pouring the acid in a small stream into the water, while stirring the latter. Concentrated sulphuric acid, when brought into contact with ice or snow, melts it immediately. The affinity of the acid for water causes the melting of the ice, and the latter, passing into the liquid state, absorbs a great quantity of heat, which it can only receive from the mixture. On the other hand, the combination of sul- phuric acid with water evolves heat. The temperature will therefore be elevated or depressed, as either of these effects may predo- minate. If we shake together rapidly 4 parts of concentrated acid with 1 part of pounded ice, the temperature will rise to 100°: but if we mix 1 part of acid with 4 of ice, the temperature will frequently fall to —4°. § 135. The composition of sulphuric acid may be ascertained in the following manner: Weigh exactly, in a small glass flask, 5 grammes of sulphur, on which highly-concentrated nitric acid is poured. By heating it moderately, the sulphur is changed into sulphuric acid, which remains mixed with the excess of nitric acid and water. When the sulphur has entirely disappeared, boil for some time; the nitric acid and a portion of the water are disengaged, and there remains in the matrass only a mixture of water and sul- phuric acid. In order to find the actual quantity of sulphuric SULPHURIC ACID. 181 acid in the mixture, it is to be combined with an anhydrous base, Avith which it will form an anhydrous sulphate. The base, Avhich is generally chosen, is the protoxide of lead, which can be easily obtained pure. A certain quantity of this oxide is Aveighed, 50 grammes, for example (the quantity should be greater than is required to saturate the acid), and poured into the flask; the sulphuric acifl combines with a portion of the oxide of lead, sulphate of lead is formed, and the Avater set free. The Avater is driven off by heating the flask, and the latter completely desic- cated by bloAving in a current of air with a bellows arranged as in (§ 107). ' The flask is weighed after cooling, and we find a weight of 62.5 Deduct the quantity of oxide of lead added 50.0 The Aveight of the sulphuric acid remains 12.5 Five grammes of sulphur therefore produce 12gm.50 of sulphuric acid. By operating in this way, a small quantity of the sulphuric acid may be lost during ebullition; for, above 212°, the tension of the vapour of this acid is very sensible. The composition of sulphuric acid may be ascertained in another Avay, free from this source of error. Five grammes of sulphur are again transformed into sulphuric acid by means of nitric acid; but the operation is performed in a small glass retort furnished Avith a receiver. The small quantity of sulphuric acid Avhich escapes is then condensed in the receiver. When the transformation of the sulphur into sulphuric acid is completed, instead of driving off the excess of nitric acid by heat, which Avould occasion a small loss of sulphuric acid, we add water, and pour into the liquid, heated to boiling, a solution of chloride of barium. A precipitate of sulphate of baryta is formed, completely insoluble in water, which is collected on a small filter, carefully washed Avith boiling Avater, until the Avash-water is no longer clouded by sulphuric acid. The filter is then dried, and calcined in a small platinum crucible. The sulphate of baryta is thus per- fectly dried; the filter is burned, leaving only some ashes AArhich are too trifling to be noticed, if the filter is small. The crucible is Aveighed, the matter it contains emptied, and it is again weighed. The difference represents the weight of the sulphate of baryta. This Aveight will be 36gm.45 : noAv, experiment has shoAvn that 100 parts of sulphate of baryta contain Sulphuric acid 34.29 Baryta 65.71 100.00 consequently, 36.45 of sulphate contain 12.50 of sulphuric acid. 182 SULPHUR. The following is a third method of ascertaining, by synthesis, the composition of sulphuric acid. We find in nature the sulphuret of lead PbS, perfectly pure and well crystallized, and called galena by mineralogists. We weigh, in a flask, a small quantity of this sulphuret reduced to a very fine powder (10 grammes for example), and treat it with concen- trated nitric acid, which changes it into the sulphate of the oxide of lead PbO,S03. We know that the transformation is perfect when the gray metallic powder of the sulphuret of lead is entirely changed into a white powder. It is then evaporated to dryness, and the residuum in the flask dried, as above (§ 107).* We then find that the 100 grammes of sulphuret of lead produce 12.676 of sulphate of lead; the increase of weight represents the oxygen absorbed by the sulphur and the lead, to change the former into sulphuric acid, and the latter into oxide of lead. We shall see, hereafter, that, in all the neutral sulphates, the proportion of oxygen contained in the base is J of that contained in the acid : consequently f. 2gm.676=2gm.007 represents the quantity of oxygen absorbed by the sulphur to transform it into sulphuric acid. The composition of the sulphate of lead is easily determined by synthesis. We weigh in a platinum crucible 10 grammes of oxide of lead, and pour upon it an excess of sulphuric acid, which changes the oxide into a sulphate. The excess of acid is then driven off by heating the crucible over an alcohol lamp to redness. We again weigh the crucible, after cooling, and obtain the weight of the sulphate of lead. This weight will be 13gm.585: we hence conclude that 10 grammes of oxide of lead combine with 3gm.585 of sulphuric acid; or, in other words, that the sulphate of lead is composed of Sulphuric acid or 26.39 Oxide of lead 10gm.000 73.61 100.00 12gm.676 of sulphate of lead consequently contain 3gm.345 of sul- phuric acid, which itself contains 2gm.007 of oxygen. We therefore arrive at this final result, that 3gm.345 of sulphuric acid contain 2gm.007 of oxygen and lgm.338 of sulphur, or that anhydrous sulphuric acid is formed of Sulphur 40.00 Oxygen 60.00 100.00 Or, if we refer this compound to the weight 16 of sulphur, which represents its equivalent, Sulphur 16 Oxygen 24 40 SULPHURIC ACID. 183 which corresponds to 1 equivalent of sulphur and 8 equivalents of oxygen: the formula of anhydrous sulphuric acid is therefore S30, and its equivalent 40. The composition of sulphuric acid may he equally established by analysis. If we pass concentrated sulphuric acid in vapour through a porcelain tube heated to redness, it separates into water which becomes free, and a mixture of sulphurous acid gas and oxygen. We find that these gases are exactly in the proportion of 2 volumes of sulphurous acid gas and 1 of oxygen. Now, 2 volumes of sulphurous-acid gas contain J of a volume of vapour of sulphur and 2 volumes of oxygen; sulphuric acid therefore con- tains : 4 of a volume of vapour of sulphur, weighing; 2.218 8 “ oxygen 3.318 which gives for the percentage composition of sulphuric acid, Sulphur 40.06 Oxygen 59.94 100.00 This composition differs but little from that obtained by synthesis. It is, however, proper to remark that the analytic method just described is less exact than the synthetic methods previously ex- plained. It requires, in fact, 1st. The measurement in volume of two gases; a measurement which is always inaccurate, especially for sulphurous acid gas which departs so remarkably from the law of Mariotte : 2d. It rests on the density of the vapour of sulphur, of which we only know the approximate value, because its experi- mental appreciation is accompanied by very great difficulties. § 136. Concentrated sulphuric acid, which has hitherto alone occupied our attention, is not an anhydrous acid: it contains a certain quantity of water, which it is important to ascertain ex- actly. We weigh, in a small flask, 100 grammes of very pure and fine powdered protoxide of lead, and pour on it carefully, by means of a pipette, a given quantity of acid which we wish to analyze. (This quantity should always be less than that necessary to convert the whole of the oxide of lead used into a sulphate.) We again weigh the flask, and find its weight to be P: the in- crease of weight (P—100) gives the quantity of concentrated acid to be experimented on. We add a small quantity of water, to assist the combination of the sulphuric acid with the oxide of lead; then evaporate the water and dry, as has been stated (§ 107). By again weighing the balloon, we find a weight P', consisting of the 100 grammes of oxide of lead and the anhydrous sulphuric acid contained in the weight (P—100) of concentrated acid: (P—P') therefore represents the weight of water contained in this acid. 184 SULPHUR. We find, in this manner, that 100 parts of concentrated sulphuric acid contain 18.3 of water and 81.7 of acid. If we refer this composition to the weight 40 of red sulphuric acid, which represents its equivalent, we find Sulphuric acid 40 Water 9 Concentrated sulphuric acid 49 These numbers give, for the percentage composition, Sulphuric acid 81.G4 Water 18.36 100.00 Now, 9 represents precisely 1 equivalent of water (§ 88); there- fore, concentrated sulphuric acid contains 1 equivalent of water and 1 equivalent of dry sulphuric acid, and its formula should he written S03+II0, or S03,IIO. The equivalent of concentrated sulphuric acid is 49. Monohydrated sulphuric acid is not the only compound of definite proportions that sulphuric acid can form with wTater. If we add to concentrated sulphuric acid a wreight of water equal to that which it already contains, we obtain a second hydrate S03+2H0, which crystallizes, in large crystals, at a temperature approaching 32°. We know that crystallization always announces a definite combination. These crystals continue so long as the temperature does not rise above 44° or 46°. In the laboratory, we frequently see these crystals in the bottles of the sulphuric acid of commerce, for it is rarely at its maximum of concen- tration, and, during the winter, a portion separates in the state of the crystallized hydrate S03-f2II0. When we mix water and concentrated sulphuric acid, the volume of the mixture is always less than the sum of the volumes of the fluids mixed : wre then say that there is contraction. Let v repre- sent the volume of the concentrated acid, v' that of the wTater, and Y the volume of the fluid after mixing, the fraction is called the fraction of contraction. This value of this fraction is smallest in the mixture of sulphuric acid and water corresponding to the formula S03+3II0, and has induced chemists to look upon this hydrate as a third definite compound of sulphuric acid and water. If we heat to ebullition the various hydrates of sulphuric acid in a tubulated retort furnished with a thermometer, we find that the hydrate S03+II0 is the only one presenting a uniform boiling point: the other hydrates give off their wTater, and the temperature of the boiling point rises successively till it reaches 617°, which is the boiling point of the concentrated acid. The acid S03+II0 is therefore the only hydrate which distils without change. SULPHURIC ACID. 185 § 13T. A peculiar sulphuric acid is prepared in the arts, and known under the name of Grerman sulphuric acid, or Nordhausen acid. This acid, the preparation of which will be explained here- after, consists of a solution of anhydrous sulphuric acid in mono- hydrated acid S03-|-H0. If Nordhausen sulphuric acid be care- fully heated in a glass retort, it separates into anhydrous sul- phuric acid, which is disengaged in the form of vapour, and mono- hydrated acid, which remains in the retort. If this vapour be collected in a small long-necked matrass, cooled by a refrigerating mixture, it condenses in long, white, brilliant needles, forming masses resembling asbestos. Anhydrous sulphuric acid melts at about 77°, and boils between 86° and 95° ; its vapour is colourless. It has a powerful attraction for water, so that if a small quantity of it be thrown into this liquid, a sound is produced resembling that of plunging a red-hot iron in water. The combination of anhydrous sulphuric acid with water disengages a great quantity of heat, and hence it follows, that when the anhydrous sulphuric acid comes in contact with the water, a high temperature is developed, which vapourizes the contiguous particles of water, but the vapour is immediately condensed by the adjacent strata of cold water. The production of vapour, followed by immediate condensation, is the cause of the hissing, which also occurs when a highly heated body, red-hot iron, for instance, is plunged into water. If we let fall a drop of water into a bottle containing anhydrous sulphuric acid, an explosion ensues with the production of light. Anhydrous sulphuric acid exposed to the air gives off dense white fumes. Its vapour, at the ordinary temperature, possesses considerable tension; for it then closely approximates the tem- perature of 95°, at which it boils under the ordinary pressure of the atmosphere. On the other hand, the vapour of monohydrated sulphuric acid S03,H0, has scarcely any sensible tension under the same circumstances. It therefore follows, that if anhydrous sulphuric acid be exposed to the air, it will give off vapour copi- ously, which immediately combines with the vapour of water of the atmosphere, forming a hydrated acid, which is precipitated in the form of mist. The fumes given off in the air by monohy- drated nitric acid were explained in the same manner (§ 102). The same is true of all other substances, gaseous or volatile, which fume when exposed to the air. § 138. Anhydrous sulphuric acid may be obtained immediately, by strongly heating the bisulphate of soda Na0,2S03, which parts with one-half of its sulphuric acid at a temperature not suf- ficiently elevated to decompose the acid. Three parts of recently calcined, and consequently anhydrous, neutral sulphate of soda, are mixed with two of concentrated sul- phuric acid, and gradually heated to a dull red. The substance swells at first, losing water, and then melts, when it is cast into 186 SULPHUR. plates, which are broken, and the pieces immediately introduced into an earthen retort, furnished with a receiver cooled by means of ice. By being carefully heated, one-half of the sulphuric acid distils in the anhydrous state, and condenses in the receiver. The residuum in the retort is neutral sulphate of soda, and may be again treated with ordinary sulphuric acid, thus serving inde- finitely for the preparation of anhydrous sulphuric acid. Anhydrous sulphuric acid may also be obtained by passing a mixture of sulphurous acid gas and oxygen through a tube con- taining platinum sponge and heated to redness. The oxygen and sulphurous acid gases, which do not act on each other when passed through a heated tube, combine, on the contrary, if the tube con- tain very finely-divided platinum ; and yet the metal undergoes no change during the experiment. We have here again an instance of that mysterious and hitherto unexplained influence which cer- tain bodies exert by their presence on chemical combinations or decompositions,—an influence which has received the name (§ 91) of action of presence, or catalysis. When the Nordhausen acid is cooled below 32°, it deposits crys- tals belonging to a hydrate containing less water than the mono- hydrated sulphuric acid, and having a formula 2S03-f IIO. The sulphates of the various bases act very differently under the influence of heat. Sulphates containing very powerful bases, as potassa, soda, baryta, or lime, are unchanged, except at the highest temperature. The sulphates formed by more feeble bases, as the metallic oxides, are decomposed at a temperature more or less elevated. Generally speaking, sulphuric acid, in such a case, is decomposed into sulphurous acid and oxygen. A portion of this latter gas frequently combines with the metallic oxide, caus- ing it to pass into a higher state of oxidation. The sulphates formed by some peroxides, the peroxide of iron, for example, decompose at so low a temperature that the sulphuric acid may escape without decomposition. On this last property is founded the preparation of the Nordhausen sulphuric acid. We obtain incidentally, in several metallurgic processes, princi- pally in the treatment of the ores of copper, large quantities of the sulphate of the protoxide of iron, called in commerce green vitriol. The formula of this salt is FeO,SQ3+7HO. Subjected to heat, the sulphate of iron loses at first 6 equiva- lents of water; the 7th is disengaged only at a more elevated tem- perature. If heated still higher, the protoxide of iron changes into a peroxide at the expense of the sulphuric acid, by absorbing a quantity of oxygen equal to one-half of that which the protoxide already contains; one-half of the sulphuric acid is thus decomposed SULPHURIC ACID. 187 and changed into sulphurous acid, which is disengaged, and there remains a subsulphate of the peroxide of iron Fe203,S03. This reaction is represented by the folloAving equation: 2(Fe0,S03)=S02+Fe203,S0s: Fe303 is the formula of the peroxide of iron. If we raise ,the temperature still higher, the sub-sulphate of the peroxide of iron is, in its turn, decomposed, the sulphuric acid becoming free, and the peroxide of iron remaining. The sulphate of the peroxide of iron still retains a little water at the moment of its decomposition, so that the sulphuric acid which is disengaged is not completely anhydrous. In the Ilartz mountains, where fuming sulphuric acid, called Nordhausen sulphuric acid (from the little village where it is depo- sited for transportation), is chiefly made, the vitriol is heated on a plate exposed to the air, until it loses the greater part of its water. It is then placed in retorts of earthen-ware A (fig. 211) arranged in 8 rows in a galley furnace, each furnace containing 120. It is heat- ed with wood, until the sulphuric acid begins to pass over, which is easily known by the dense vapours it produces in the air. To the vessels A, which act as retorts, are then adapted vessels B nearly resembling them, but smaller, and serving as receivers. Ordinary concentrated sulphuric acid is put in these receivers, being much cheaper than the fuming acid, and is not considered to be changed into Nordhausen acid until it has received the products of four successive distillations. It is then composed of nearly of anhydrous sulphuric acid, and § of monohydrated acid. A similar acid may he prepared in the laboratory, by introducing into a stone retort, the peroxide of iron of commerce, known under the name of colcothar, moistening it with concentrated sul- phuric acid, and then distilling. The first products are not col- lected, as they contain too much water, while the last are very rich in anhydrous sulphuric acid. § 139. The preparation of commercial monohydrated sulphuric acid, sometimes called English sulphuric acid or oil of vitriol, is founded on the following reactions, which we have previously indicated : 1st. The deutoxide of nitrogen N02, in contact with an excess of air, is changed into hyponitric acid N04; 2d. Hyponitric acid, in contact with a small quantity of water, is changed into monohydrated nitric acid and nitrous acid, Fig. 211. 2N04+H0=N05,H0+N03. 188 SULPHUR. 3d. Nitrous acid N03, in contact with a large quantity of water, is changed into hydrated nitric acid and deutoxide of nitrogen, 3N03+nH0=N05+nH0+2N03. Consequently, hyponitric acid, in contact with a large quantity of wrater, is changed into hydrated nitric acid and into deutoxide of nitrogen, 6N04+nH0=4N05+nH0+2N02: 4th. Sulphurous acid S03, in contact with hydrated nitric acid N054-nII0, is changed into sulphuric acid, while the nitric is transformed into hyponitric acid. S0fl+N0fi+nH0=S03+nH0+N04. The following experiment explains all the reactions which occur in making oil of vitriol by the English method: We introduce, at the same time, into a large balloon A (fig. 212) Fig. 212. filled with air, and the sides of which are moistened, 1st, sul- phurous acid gas obtained by heating some copper with concen- trated sulphuric acid in a flask, and, 2dly, deutoxide of nitrogen produced in the bottle C, by causing dilute nitric acid to act upon copper. The deutoxide of nitrogen, mixing with the air of the balloon A, combines with the oxygen, and changes into hyponitric acid N04, which, under the influence of the moisture of the balloon, changes, in its turn, into hydrated nitric acid and deutoxide of nitrogen. The nitric acid formed reacts on the sulphurous acid, which it changes into sulphuric, and changes itself into hyponitric acid, which is again decomposed by contact with water into nitric acid and deutoxide of nitrogen. The newly-formed deutoxide of SULPHURIC ACID. 189 nitrogen, coming into contact with the oxygen of the air, changes into hyponitric acid; and this succession of remarkable reactions continues indefinitely. So that, as long as any oxygen remains in the balloon, the same deutoxide of nitrogen may change an in- definite quantity of sulphurous into sulphuric acid. This result may be obtained by passing into the balloon, through one of its four tubes, a .slow current of oxygen, which will replace that dis- appearing in consequence of the reaction. The deutoxide of nitrogen may be replaced, in this experiment, by any more oxygenated compound of nitrogen, as hyponitric or nitric acid. But, in order that all these circumstances may combine, there must be a large proportion of vapour of water in the balloon. That which would be disengaged from the moist sides at the ordi- nary temperature not being sufficient, it is necessary to heat the bottom of the balloon. When there is less water, the reaction is indifferent. Let us sup- pose that there is no water in our balloon; the sulphurous acid and hyponitric gases then act slowly on each other; but we have seen (§ 132) that when the two substances in the liquid state are mixed in a tube, which is then hermetically sealed, they combine after some time, forming a crystallized compound, which is a hydrate of the preceding compound, N03,2S03. This hydrate is constantly formed in the balloon, and deposited on its sides in the form of small crystalline tufts, if the balloon be not heated. The same crys- tals also frequently form in the manufacture of sulphuric acid on a large scale, and have been called crystals of the leaden chambers. They should, however, only be considered as accidental, and their formation avoided ; for, if they do not afterward meet with water to decompose them, they dissolve in the sulphuric acid, the purity of which they change, as it thus retains a portion of nitrous acid, which would have served to change an additional quantity of sul- phurous into sulphuric acid. In the manufacture of sulphuric acid on a large scale according to the English method, the balloon of our experiment is replaced by one or more large wooden chambers C (fig. 213) lined with sheet-lead, closely soldered. The sulphurous acid is prepared by burning sulphur in atmospheric air, the combustion taking place in a furnace A, on a large pan of sheet-iron. The furnace is surmounted by a dome and a large flue in mason-work, which conducts the gas into the leaden chamber. The oxygenated compound of nitrogen is the deutoxide of nitrogen, nitrous vapours, or nitric acid. In some manufactories, nitrate of potassa is placed in a small cast-iron pot in the pan containing the burning sulphur. This pot becomes thus highly heated, the sulphurous acid reacts on the nitrate of potassa, transforms it into a sulphate, and deutoxide of nitrogen is disengaged, which enters 190 SULPHUR. the leaden chamber, mixed with sulphurous acid and an excess of atmospheric air. In order to produce the reaction which will form sulphuric acid, it is sufficient to inject into the chamber, jets of steam under high pressure from the boiler B. The hydrated sulphuric acid falls in the form of rain upon the floor of the chamber. The quantity of nitrate of potassa used is about A that of the sulphur. An opening at the upper part of the chamber, provided with a valve s, gives exit to the remaining gases. These gases should be deprived, as completely as possible, of sulphurous acid and oxide of nitrogen, and to do this, several conditions must be fulfilled: 1st. The proportions of nitre and sulphur burned must be pro- perly regulated. 2d. The quantity of steam injected must be proportioned to the quantity of gas on which we operate, for if it be too small, reaction takes place with difficulty, many crystals of the leaden chamber are produced, which cause a loss of the nitrous products, and im- pair the purity of the sulphuric acid. If the quantity of steam is too great, we obtain a very dilute sulphuric acid, which requires considerable expense to be brought to a proper degree of strength: 3d. The leaden chambers should be made as large as possible: in order that the gases may remain in them for a long time; and they should be so arranged as to effect a perfect mixture of the gases. To do this, they are divided into several compartments, separated by leaden partitions pierced with holes at the bottom; or several chambers are arranged in succession, communicating by leaden pipes. One or several jets of steam are projected into each of the chambers, and regulated by stopcocks. Sulphur is some- times burned in several chambers, so as to generate sulphuric acid at several points at once. Registers, properly arranged, allow us to graduate the quantity of atmospheric air admitted into the apparatus. In many manufactories, nitric acid is substituted for the deut- oxide of nitrogen. Sulphur alone is burned in the furnaces; the mixture of atmospheric air and sulphurous acid enters the first chamber, which is small, and in which the foreign substances car- ried over by the current of gas are deposited. A leaden pipe Fig. 213. SULPHURIC ACID. 191 conveys the mixed gases into a second chamber, into which nitric acid is steadily poured. This acid, contained in vessels placed outside, is made to fall on porcelain saucers arranged like a fountain, and immediately beneath the orifice of the pipe which conducts the mixture of sulphurous acid and air. The current of hot gas vapourizes the nitric acid, at the same time that its sul- phurous acid .decomposes it. The gases, intimately mixed, reach several large leaden chambers successively, where the chief reac- tion takes place, amid jets of steam projected at various points. Small openings are made in the walls of these rooms, which allow us to inspect them, and ascertain if the gaseous mixture contains a proper quantity of nitrous vapour. The flow of nitric acid is governed by this knowledge. Iron pyrites is at present substituted in a few manufactories for sulphur, that is to say, a sulphuret of iron FeSz, found abun- dantly in many places, and consequently cheaper than sulphur. The pyrites will burn in a furnace previously heated, and its sulphur be converted into sulphurous acid. But the sulphuric acid thus obtained always contains some arsenious acid, arising from the metalline arseniurets which almost always accompany iron pyrites. The manufacture of sulphuric acid by the English method has greatly advanced in latter years, the apparatus having been im- proved, the production of the article doubled, and the propor- tion of the nitre used much diminished. Fig. 214 represents a section of the apparatus now used. (We have supposed its various parts to be arranged in a line, to render the wood-cut more intelligible, though it is not generally the case in large esta- blishments.) A, A' are two furnaces coupled together, in which the sulphur is burned, one of them, A', being seen in section so as to show its internal arrangement. The sulphur burns on a large sheet-iron plate, and the heat produced by the combustion is used to furnish the quantity of steam necessary for reaction in the leaden chambers. For this purpose, a boiler Y is placed in each furnace, immediately over the pan on which the sulphur burns, and a pipe aa'a" conducts the steam into the different chambers. The two furnaces communicate with the same chimney bb', which should be at least 6 or T metres (20-24 feet) in height, so as to give the gas an ascending force sufficient to drive it through the various parts of the apparatus. The chimney bb' conveys the mixture of sulphurous acid gas and atmospheric air into a leaden drum BB, in which are arranged small inclined shelves of lead. A continuous current of oil of vitriol, properly regulated, and strongly charged with nitrous products, is made to fall on the upper shelf, from the vessel R. As the sulphuric acid flows along the shelves, and collects on the bottom of the drum, a portion of 192 SULPHUR. Fig. 214. the nitrous products reacts on the sulphurous acid, which it con- verts into sulphuric, while the remainder is disengaged in the state of gas in the gaseous mixture of sulphurous acid and atmo- spheric air. From the drum BB, the gases are conveyed by the cast-iron pipe c, into a small leaden chamber C, containing about 100 cubic metres (360 cubic feet), and called the denitrificator. At the very origin of the pipe c, a jet of steam, under high pressure, is driven into the chamber C, to furnish the water necessary for the reaction of the nitrous gas, oxygen and sulphurous acid. The sulphuric acid produced falls on the floor of the chamber C. The gases are then conveyed, by the pipe d, into a second chamber D, of nearly the same size as the first. In front of the orifice of the pipe d, an earthen-ware vessel like a cascade-fountain is placed, on the top of which a continuous and properly regulated stream of nitric acid is poured. (This acid is contained in vessels outside of the chamber, and not represented in the figure.) The nitric acid is decomposed ; sulphuric acid is formed, and the nitrous gas produced in the reaction mixes with the sulphurous gas and SULPHURIC ACID. 193 atmospheric air. The sulphuric acid thus obtained is highly charged with nitrous compounds ; it falls on the floor of the room D, and thence flows, through a small pipe, into the room C, where it comes into contact with gases containing a large quantity of sulphurous acid, which takes from it its nitrous products. For this purpose, the floor of the chamber D is somewhat higher than that of C. The gases are then conveyed by the pipe e into a large chamber E, where the reaction of the sulphurous, nitrous and oxygen gases chiefly takes place, because the gases remain there for some time. Jets of steam are projected into this chamber at several points. The sulphuric acid produced falls upon the floor. At the same time, the denitrified sulphuric acid of the chamber C, of which the floor is somewhat higher than that of E, is brought in. Some- times, instead of one large chamber E, there are several smaller ones, placed in succession. The gases, on leaving the room E, are not lost in the air. The temperature of this chamber is very high, and a considerable portion of sulphuric acid still remains there in the form of vapour. Moreover, the gases still contain nitrous products, of which they can be deprived, so as to economize the nitric acid. The gases, on leaving the chamber E, are passed through two leaden drums, F, G, which act as refrigerators, and in which are arranged shelves which interrupt the gaseous current, and thus assist the deposit of the vapours. The gases then reach a third refrigerator I, cooled externally with water; and, lastly, they reach a last leaden drum H, intended to absorb the nitrous gases, and thence escape into the atmosphere by the pipe T. The drum H is filled with large fragments of coke, supported by a diaphragm s, and on which a continuous current of con- centrated sulphuric acid descends from the vessel Q. This acid absorbs the nitrous vapours, and then passes, by the leaden tube mm'm", into a vessel L. This concentrated sulphuric acid, loaded with nitrous products, is then made to ascend into the vessel R, to fall again into the drum BB, where it is denitrified. A very simple arrangement facilitates this transfer: the top of the vessel R communicates with the bottom of the vessel L by the pipe zz'; and the top of the vessel L has a tube, furnished with a stopcock r, which joins the general steampipe aa'a". In order to cause the liquid of the vessel L to ascend into the vessel R, we merely open the stopcock r: the pressure of the steam in the boiler, always equal to several atmospheres, acting on the surface of the fluid L, causes it to rise to the level R. By means of the apparatus just described, the quantity of nitric acid necessary to convert 100 kilogrammes of sulphur into sul- phuric acid, has been reduced by one half of that formerly used. § 140. The solution of sulphuric acid, as it leaves the leaden chambers, has a density varying from 1.35 to 1.50. It is con- 194 SULPHUR. centrated in leaden pans until its density reaches 1.75. Its boil- ing point is then between 892° and 410°. Its concentration in leaden vessels cannot be carried farther, as it would attack the lead, and is completed in a large platinum retort, where it is brought to the state of monohydrated sulphuric acid, having a density of 1.85 and a boiling point of 617°. Hyposulphuric Acid, Sa05. § 141. If we digest a solution of sulphurous acid, with the peroxide of manganese, in the cold, the acid soon loses its charac- teristic odour, and the liquid contains the hyposulphate of the protoxide of manganese. Two equivalents of sulphurous acid combine with one equivalent of oxygen, given off by the peroxide of manganese, which passes into the state of protoxide. We have Mn03+2S0a=Mn0,S305. If, on the other hand, we pass the current of sulphurous acid through hot water, holding finely divided peroxide of manganese in suspension, the gas is likewise absorbed, but the reaction takes place between 1 equivalent of the peroxide and 1 equivalent of the acid, and the sulphate of the protoxide of manganese is formed, Mn03+2S03=Mn0.S305. Thus, the reaction differs according to the temperature. In order to prepare hyposulphuric acid in the laboratory, finely divided peroxide of manganese is suspended in water, and a cur- rent of sulphurous acid gas passed through the liquid. The two reactions first mentioned take place simultaneously, forming at the same time the sulphate and hyposulphate of manganese. The liquid is filtered and decomposed by a solution of caustic baryta, which precipitates the protoxide of manganese, and forms the sulphate and hyposulphate of baryta. The sulphate of baryta is completely insoluble in water, and precipitates along with the oxide of manganese, so that the hyposulphate of baryta alone remains in solution, from which it is crystallized by evaporation. The hyposulphate of baryta is again dissolved in water, and dilute sulphuric acid carefully added, until the addition of a single drop of this reagent no longer clouds the fluid. The baryta is entirely precipitated in the form of a sulphate, and the liquid only contains hyposulphuric acid. This solution is evaporated under the receiver of an air-pump, until sufficiently concentrated. The evaporation can only be conducted in the cold ; for, when the liquid is too highly concentrated, the hyposulphuric acid is decom- posed by heat into sulphurous and sulphuric acids. By double decomposition the various hyposulphates are obtained by means of the hyposulphate of baryta. It is sufficient to. pour HYPOSULPHURIC ACID. 195 carefully into the solution of hyposulphate of baryta, a diluted solution of the sulphate of the base we wish to combine with hypo- sulphuric acid, until no more precipitate is thrown down. The baryta is thus eliminated in the state of a sulphate, and the liquid contains the hyposulphate, which can be crystallized. § 142. The composition of hyposulphuric acid may be easily ascertained by the analysis of the hyposulphate of baryta. By calcining a given weight (5 grammes) of anhydrous hyposul- phate of baryta, the salt is decomposed, the sulphurous acid is disengaged, and neutral sulphate of baryta remains, which is exactly weighed. From the known composition of the latter, we infer that 100 parts of hyposulphate of baryta contain Baryta 51.51 Hyposulphuric acid 48.49 100.00* or, if we refer this composition to the weight 76.5 of baryta, which represents the equivalent of this base, Baryta 76.5 Hyposulphuric acid.... 72.0 148J> If the hyposulphate of baryta be a neutral salt, and if hyposul- phuric acid be a monobasic acid, the weight 72 should represent the equivalent of hyposulphuric acid, and should equal the sum of the equivalent of its constituent elements. Now, we obtain the number 72, by the addition of 2 equivalents of sulphur and 5 of oxygen : the composition of hyposulphuric acid is, therefore, 2 eq. sulphur 32.0 44.45 5 “ oxygen 40.0 55.55 1 “ hyposulphuric acid 72.0 100.0 This composition may be verified by direct analysis. In fact, if we take 5 grammes of dry hyposulphate of baryta, and treat it with concentrated and boiling nitric acid, the hyposulphuric will be converted into sulphuric acid, of which one half only will be saturated by the baryta. But, if we add chloride of barium to the liquid, all the sulphuric acid will be precipitated in the state of sulphate of baryta. We shall find that the weight of sulphate of baryta obtained is precisely double of that formed by the cal- cination of the hyposulphate. We hence conclude that 100 of hyposulphate of baryta contain Sulphur 21.55 Oxygen 26.94 Hyposulphuric acid 48.49 * The numbers given vary a little from the original, from the adoption of the equivalent 68.5 for barium instead of 68.64.—J. G. B. 196 SULPHUR. Hyposulphurous Acid, Sa02. § 143. This acid has not been hitherto obtained in an isolated state, and is only known combined with bases. Hyposulphites are obtained in several ways : By boiling a solution of sulphite of soda, or any other sul- phite, with flowers of sulphur in excess, a great quantity of sul- phur will be found to dissolve, and the sulphite of soda NaO,SOa is changed into hyposulphite Na0,Sa02. This salt crystallizes readily. If chlorohydric acid be poured into a very cold solution of hyposulphite of soda, the liquid is not clouded at first; but a precipitate of sulphur soon forms, and sulphurous acid is disen- gaged. Hyposulphites are also otherwise obtained: A piece of zinc disappears in a solution of sulphurous acid, without any disengagement of hydrogen gas. The oxidation takes place at the expense of a portion of the oxygen of the sulphurous acid, which passes into the state of hyposulphurous acid, and the fluid contains a mixture of sulphite and hyposulphite of zinc. Thus wo have 2Zn+3SOa=ZnO,SaOa+ZnO,SOa. The solutions of the alkaline sulphurets, exposed to the air, absorb oxygen rapidly, and are converted into hyposulphites. When solutions of potassa, baryta, or soda are boiled with an excess of sulphur, hyposulphites are obtained mixed wTith sulphurets saturated with sulphur. Thus, with potassa, we have the follow- ing reaction: 3KO+12S=2KSs+KO,SaOa. § 144. The composition of hyposulphurous acid is ascertained by the analysis of hyposulphite of baryta. Ten grammes of dry hyposulphite of baryta are treated with concentrated boiling nitric acid, which changes the salt into sul- phate of baryta, which is weighed. We thence deduce that 100 of hyposulphite of baryta contain Baryta 61.45 Hyposulphurous acid 38.55 100.00 or in equivalents, 1 eq. baryta 76.5 1 “ hyposulphurous acid 48.0 1 “ hyposulphite of baryta 124.5 The composition of sulphurous acid is therefore 2 eq. sulphur 32.0 66.67 2 “ oxygen 16.0 33.33 1 “ hyposulphurous acid 48.0 100.00 TRISULPHURETTED HYPOSULPHURIC ACID. 197 This composition may be verified by analysis, as described (§ 142). Monosulphuretted IIyposulpiiuric Acid, S305. § 145. Monosulphuretted hyposulphuric (trithionic) acid is ob- tained under the following circumstances :—A solution of baryta with sulphurous acid is supersaturated to obtain bisulphite of baryta, which is allowed to digest for several days with flowers of sulphur, at a temperature of about 112°. The liquid at first turns yellow, then loses its colour, and on cooling, it deposits a salt crystallized in long white needles, which is the monosulphuretted hyposulphate of baryta. By cautiously pouring sulphuric acid into the solution of this salt, the monosulphuretted hyposulphuric acid is obtained isolated. Its solution may be concentrated under the receiver of an air-pump, but heat readily decomposes it into sulphuric acid and sulphur. The analysis of monosulphuretted hyposulphuric acid is made in the same way as that of the preceding compounds: we know that its equivalent is 88, and that it contains, 3 eq. sulphur 48 54.54 5 “ oxygen 40 45.46 , 88 100.00 Bisulphuretted Hyposulphuric Acid, S405. § 146. This compound is obtained by dissolving iodine in a solution of hyposulphite of baryta, when the following reaction takes place: 2(Ba0,S302)-fI=IBa+Ba0,S405. The liquid contains iodide of barium and the salt of baryta formed by the new acid. This salt, being less soluble than the iodide of barium, separates by crystallization. In order to isolate the acid, the salt of baryta is decomposed by a proper quantity of sulphuric acid. The solution of bisulphuretted hyposulphuric acid (tetrathionic) may be concentrated in vacuo; ebullition decomposes it. The composition of this substance is ascertained by the analysis of the salt of baryta. We thus find that its equivalent is 104, and its composition as follows : 4 eq. sulphur 64 61.54 5 “ oxygen 40 38.46 104 lOChOO Trisulphuretted Hyposulphuric Acid, S505. § 147. Trisulphuretted hyposulphuric acid (pentathionie) is formed when the chlorides of sulphur are decomposed by a solu- tion of sulphurous acid, or even by pure water; but the reaction R 2 198 SULPHUR. from which it originates has not been well studied. This acid forms with baryta a crystallizable salt, from the analysis of which the composition of the acid has been deduced. Trisulphuretted hyposulphuric acid contains 5 eq. sulphur 80 66.67 5 “ oxygen 40 33.33 120" 10000 The composition of trisulphuretted hyposulphuric and hyposul- phurous acids are identical. These acids are isomeric compounds. But these salts are very differently compounded, for the quan- tities of the bases wdiich these acids saturate, are to each other as 5 : 2. RECAPITULATION OF THE COMBINATIONS OF SULPHUR WITH OXYGEN. EQUIVALENT OF SULPHUR DETERMINED. § 148. The seven compounds of sulphur and oxygen, just studied, present the following composition : Hyposulphurous acid Sulphur 66.67 Oxygen 33.33 100.00 Trisulphuretted hyposulphuric,acid Sulphur 66.67 Oxygen 33.33 ioooo Bisulphuretted hyposulphuric acid Sulphur 61.54 Oxygen 38.46 100.00 Monosulphuretted hyposulphuric acid Sulphur 54.54 Oxygen 45.46 100.00 Sulphurous acid Sulphur 50.00 Oxygen 50.00 100.00 Hyposulphuric acid Sulphur 44.45 Oxygen 55.55 100.00 Sulphuric acid Sulphur 40.00 Oxygen 60.00 100.00 If we refer the composition of these various substances to the same quantity, 100 of sulphur, we find Hyposulphurous acid Sulphur.... 100.00 Oxygen— 50.00 "15000 EQUIVALENTS. 199 Trisulphuretted hyposulphuric acid Sulphur.... 100.00 Oxygen.... 50.00 150.00 Bisulphuretted hyposulphuric acid Sulphur.... 100.00 Oxygen— 62.50 162.00 • Monosulphuretted hyposulphuric acid.... Sulphur.... 100.00 Oxygen.... 83.33 183.33 Sulphurous acid Sulphur.... 100.00 Oxygen.... 100.00 200.00 Hyposulphuric acid Sulphur.... 100.00 Oxygen 125.00 225.00 Sulphuric acid Sulphur.... 100.00 Oxygen.... 150.00 250.00 If we compare the quantities ” of oxygen which combines with the same weight of sulphur, we find that they are to each other as the numbers l:l:f:f:2:f:3. Let us suppose that the least oxygenated compound, hyposul- phurous acid, be formed of 1 equivalent of sulphur and 1 equiva- lent of oxygen = 8. It is evident that we shall obtain the equivalent of sulphur by making the proportion, 50.00 :100.00 : : 8 : x, whence x = 16. Hyposulphurous acid will therefore be SO Trisulphuretted hyposulphuric acid SO Bisulphuretted hyposulphuric acid S05 Monosulphuretted hyposulphuric acid S05 Sulphurous acid S02 Hyposulphuric acid SOg 2 Sulphuric acid S03 If the above formulae really represent the equivalents of these various acids, the numerical values of these equivalents, that is, the weights of these acids which combine with an equivalent of a base to form an anhydrous neutral salt, will be as follows: 200 SULPHUR. Hyposulphurous acid 24 Trisulphuretted hyposulphuric acid 24 Bisulphuretted hyposulphuric acid 26 Monosulphuretted hyposulphuric acid 29|- Sulphurous acid 32 Hyposulphuric acid 36 Sulphuric acid 40 Now, we have seen, by direct experiment, that the weights of these various acids which combine with 1 equivalent of a base, with the weight 76.5 of baryta, for example, to form the anhy- drous neutral salts, are, Hyposulphurous acid 48 Trisulphuretted hyposulphuric acid 120 Bisulphuretted hyposulphuric acid 104 Monosulphuretted hyposulphuric acid 88 Sulphurous acid 32 Hyposulphuric acid 72 Sulphuric acid t 40 Experiment thus shows us that the equivalents of sulphurous and sulphuric acid are those which we supposed by hypothesis; but that this is not the case with the other acids. The equivalents of hyposulphurous and hyposulphuric acids are twice as great, that of monosulphuretted hyposulphuric acid, thrice, and that of bisulphuretted hyposulphuric acid, four times; and, lastly, that of trisulphuretted hyposulphuric acid five times as great as those we supposed. The formula of these various combinations will therefore be, Hyposulphurous acid S30a Trisulphuretted hyposulphuric acid S50s Bisulphuretted hyposulphuric acid S405 Monosulphuretted hyposulphuric acid S305 Sulphurous acid S03 Hyposulphuric acid S205 Sulphuric acid S03 The number 16, which we will adopt as the equivalent of sul- phur, possesses, therefore, the property of representing the com- position of the numerous compounds of sulphur with oxygen, by entire formulae, the most simple possible. Again, the numerical values of the equivalents of these combinations, calculated from the formulae, are equal to those obtained by ascertaining experi- mentally the weights of those compounds necessary to form anhy- drous neutral salts with 1 equivalent of a base. We shall subsequently see that this weight 16 of sulphur, chosen as the equivalent, will give, for all the other compounds SULFHYDRIC ACID. 201 of sulphur, very simple formulae, and, when these combinations are acid, their formulae will also satisfy the second condition just indicated. In the atomic theory, we suppose 1 atom of sulphurous acid to be composed of 1 atom of sulphur and 2 atoms of oxygen: then 1 atom of sulphuric acid is formed of 1 atom of sulphur and 3 'atoms of oxygen. The atomic formulae of the compounds of sulphur with oxygen will therefore be the same as their formulae in equivalents, and the weight of the atom of sulphur will be 16. COMBINATIONS OF SULPHUR WITH HYDROGEN. SULFHYDRIC ACID, HS.* § 149. Sulphur and hydrogen do not combine directly, even wrhen passed through a porcelain tube heated to redness; but a gaseous combination of the two substances is obtained by decom- posing ce/tain metallic sulphides by dilute sulphuric acid. The protosulphuret of iron is the one generally used in the laboratory. The following is the reaction : FeS+S08+H0=Fe0,S08+HS. The same apparatus is used as for the preparation of hydro- gen. The sulphuret of iron, broken into pieces, is introduced into a two-mouthed bottle, a quantity of water poured there- in, and sulphuric acid gradually added by the funnel tube (fig. 215).f Chlorohydric acid may be sub- stituted for the sulphuric, when the reaction is as follows : Fig. 215. Sulphuret of iron... Sulphur.. Chlorohydric acid. Iron.. Hydrogen .. >Sulfhydric acid. >Chloride of iron. FeS + HCl=FeCl+HS. Chlorine. The sulphuret of iron usually employed in the laboratory to * This gas was formerly called sulphuretted hydrogen, more recently hydro- sulphuric acid, or, better, sulphohydric acid, but I prefer giving the French name, sulfliydric acid.—J. C. B. f A wide-mouthed bottle, with the two tubes passed through the single cork, is as convenient and less costly. Where the gas is to be used in analytic opera- tions, it should be washed by being passed through a small bottle, previous to its entrance into the liquid to be acted on.—J. C. B. 202 SULPHUR. obtain sulfhydric acid, is prepared expressly for that purpose, but as it often contains small quantities of metallic iron, which, in contact with dilute sulphuric or chlorohydric acid, gives off hydrogen, the sulfhydric acid gas is mixed wTith hydrogen. In many experiments, this is of no importance; but, where absolute purity is required, the sulfhydric acid must be prepared by treat- ing sulphuret of antimony with chlorohydric acid. The sulphuret of antimony is a natural product found abundantly in some veins. It can be acted on only by concentrated acids; but sulphuric cannot be used, for, if dilute, it does not attack the antimony, and if concentrated, it destroys the sulfhydric acid as fast as it is generated. In order to prepare the gas with sul- phuret of antimony, we place the latter, in fine powder, in a small flask (fig. 216), and add chloro- hydric acid gradually by the S-tube. Gentle heat is applied to accelerate the disengagement of the gas. § 150. Sulfhydric acid is a colourless gas, pos- sessing a most fetid odour, resembling that of rotten eggs. Its density is 1.1912. It liquefies under a pressure of 15 or 16 atmospheres at the ordinary temperature, and then forms a very mobile liquid of the density of 0.9. In order to obtain liquid sulfhydric acid, the apparatus in which it is generated is made to connect with the suction-pipe of a gas pump, which is at the same time a forcing-pump, the second pipe of which connects with a small bulb A (fig. 217) of thick glass, and kept in a refrigerating mixture. By raising the piston of the pump, the gas of the apparatus fills the body of the pump, and by depressing, it is forced into the bulb. The number of strokes of the piston is regulated by the quantity of sulfhydric acid gas the apparatus will furnish. The com- pressed gas liquefies in the bulb; and when it is three- fourths full, the neck must be sealed hermetically. But as the neck cannot be melted in the flame of a lamp, because the pressure is greater within the apparatus than without, the following plan is adopted. The tube attached to the bulb is composed of a narrow part ab and a larger one be : before fitting the tube to the pipe of the pump, a plug of mastic is placed in the latter, so as not to impede the passage of the gas; and, in order to seal the apparatus hermetically, it will suffice to melt it, Fig. 216. Fig. 217. SULFHYDRIC ACID. 203 then to give a stroke with the piston, which will drive the melted mastic into the narrow tube ab, where it becomes solid. Liquid sulfhydric acid may also be obtained by exposing to spontaneous decomposition, in a close vessel, the second combina- tion of sulphur with hydrogen, which we shall soon learn is the bisulphide of hydrogen. A certain quantity of this liquid bisul- phide is placed in the bottom of a curved tube, as in fig. 218, and the end b is sealed in a lamp. The bisulphide decomposes spontaneously into sulphur, which is depo- sited in the form of crystals, and into sulfhydric acid gas, which accumulates in the empty portion of the tube, where it liquefies by its own pressure. In order to separate the acid from the sulphur which is deposited, it is merely necessary to cool in a refrigerating mix- ture the curved part cd (fig. 219), when the sulfhydric acid passes over and collects at d. Sulfhydric acid is one of the most deleterious gases,* a bird perishing in an atmosphere containing and a dog 777 of this gas. Labourers who clean sinks are often exposed to asphyxia from this gas. It is remedied by chlo- rine, which decomposes the sulfhydric acid ; but this remedy must be carefully administered. The best plan is to use a napkin soaked in acetic acid, and enclosing some pieces of chloride of lime, through which the patient is made to breathe.f Heat partially decomposes sulfhydric acid into hydro- gen and sulphur; but, in order to obtain perfect decomposition, the gas should be repeatedly passed through a highly heated porcelain tube. Sulfhydric acid gas is combustible, burning in the air with a blue flame, and the product is water and sulphurous acid gas. If the gas be inflamed in a test-glass, the sulphur does not burn com- pletely, but is partly deposited on the sides of the glass. When a mixture of sulfhydric gas and air in a large bottle is in contact with a porous body, especially with lime, at a tem- perature of 104° to 122°, a considerable quantity of sulphuric acid is gradually formed. The reaction is interesting, because it explains the formation of sulphuric acid and sulphates in localities furnishing sulphuretted hydrogen. Oxygen, dissolved in water, slowly decomposes sulfhydric acid, water being formed, and finely divided sulphur deposited, render- ing the water milky. In order, therefore, to preserve a solution Fig. 218. Fig. 219. * This is certainly incorrect, although stated positively in almost all works on chemistry. I have breathed it, and witnessed its effects on others, in large quan- tity, and cannot say that it is a very deleterious gas. See Sulphur, in Encyclo- pedia of Chemistry.—J. C. B. f Spirits of hartshorn (ammonia) may be inhaled with good effect, and, still better, a mixture of ammonia and strong alcohol.—J. G. B. 204 SULPHUK. of sulfhydric acid, it should be kept in well-stoppered bottles entirely filled, and inverted. Sulfhydric acid therefore affords different products of combus- tion, according to the circumstances under which oxidation takes place ; in rapid combustion, it furnishes water and sulphurous acid; when in contact with a porous body, and at a temperature of 104° to 122°, water and sulphuric acid are formed; lastly, dissolved in water, and exposed to the air, it gives water and sulphur, which is precipitated. Chlorine, bromine, and iodine instantly decompose sulfhydric acid, affording sulphur, and chlorohydric, bromohydric, and iodohydric acids. If the chlorine, bromine, and iodine are in excess, they combine with the isolated sulphur, and form the chloride, bromide, and iodide of sulphur. Advantage is taken of this property to prepare a solution of iodohydric acid. Sulfhydric gas is a true acid, for it reddens litmus, but, like all feeble acids, it produces a purplish red ; whilst powerful acids, such as nitric and sulphuric, produce a light red. Its acid pro- perties are but feebly developed, and hence it is often called sul- phuretted hydrogen gas (§ 52). Water dissolves 2| to 3 times its volume of sulfhydric acid gas. The solution may be prepared in Woolf’s apparatus, by taking care to put into the bottles water recently boiled, and consequently deprived of air. The solution, when heated, parts wholly with the gas. Alcohol dissolves 5 or 6 times its volume of sulfhydric acid gas. The solution of sulfhydric acid is much used in the laboratory, being employed to precipitate many metals from their saline solu- tions, in the state of sulphides. These sulphurets, generally insoluble, have frequently characteristic colours, by which the metals contained in them are recognised. Thus, a solution of sulfhydric acid will detect the slightest traces of oxide of lead in a fluid, by the brown or black colour it produces. Reciprocally, the salts of lead discover the presence of the smallest portions of sulf- hydric acid. For this purpose, small strips of paper are used, imbued with a solution of sugar of lead. The strips are colour- less, but are instantly blackened when dipped into water containing the slightest traces of sulfhydric acid, or when, after having been moistened, they are exposed to an atmosphere containing the slightest traces of the gas. In nature, many mineral waters are found containing sulfhydric acid, and are used in medicine under the name of sulphurous waters. § 151. Sulfhydric acid may be analyzed by decomposing it in a curved tube, by means of potassium, as is done in the other combinations of hydrogen with the metalloids (§ 186). Potassium effectually decomposes sulfhydric gas, but the sulphuret which re- SULFHYDRIC ACID. 205 suits combines with the undecomposed sulfhydric acid, and forms a sulfhydrate of the sulphide of potassium ; so that a portion of the gas escapes decomposition. But the analysis may be exactly made (fig. 220) by substituting tin for the potassium. The glass being heated by an alcohol lamp, the tin combines with the sulphur, and the hydrogen is set free. The volume of the gas remaining is found to be exactly the same. We may assure ourselves that the sulfhydric acid has been entirely decomposed by introducing into the glass a fragment of moist potassa, for if sulfhydric acid remain, it is absorbed, and the volume diminished. We infer from the preceding experiment, that 1 volume of sulf- hydric acid gas contains 1 volume of hydrogen. Now if, from the density of sulfhydric acid gas 1.1912 we deduct the density of hydrogen 0.0692 there remains 1.1220 which nearly equals the l of the density of the vapour of sulphur 6.6546 = —g—=1.109. We therefore conclude that 1 volume of sulfhydric acid gas is composed of 1 volume of hydrogen and volume of vapour of sul- phur; or, if we refer the composition to two volumes of hydro- gen, its equivalent, we say that 2 volumes of sulfhydric acid gas contain 2 volumes of hydrogen and J volume of vapour of sul- phur. But as J volume of the vapour of sulphur represents the equivalent of gaseous sulphur, sulfhydric acid is therefore formed of 1 equivalent of sulphur and 1 equivalent of hydrogen, and its equivalent is 2 volumes. The volume J, which we have chosen as the equivalent of gaseous sulphur, has, therefore, the advantage of expressing the composition of sulfhydric acid in the simplest manner possible. § 152. We shall subsequently see that there is a remarkable analogy between the compounds of sulphur and those of oxygen: hence we ought to expect to find in the compounds of sulphur and hydrogen a constitution similar to that of the compounds of oxygen with this substance. Sulfhydric acid presents, however, in this respect, a very interesting anomaly. In the aggregate of its pro- perties, it ranks with water; and this position is so natural, that chemists, before the density of the vapour of sulphur was known, did not hesitate to attribute to it the same composition. But ex- periment has proved the supposed analogy to be false, since sulfhy- dric acid, for 2 volumes of hydrogen, contains only volume of vapour of sulphur, instead of 1 whole volume. This anomaly has Fig. 220. 206 SULPHUR. been attempted to be explained by saying that the molecule of the vapour of sulphur is a group formed by the union of three chemical molecules.* We have seen that the weight 1.1912 of sulfhydric acid contained 0.0692 of hydrogen and 1.1220 of sulphur; consequently, 100 parts in weight contain Hydrogen 5.81 Sulphur 94.19 100.00 If we refer this composition to the weight 1 of hydrogen, repre- senting its equivalent, we find Hydrogen 1 Sulphur 16 Sulfhydric acid 17 The formula of the acid in equivalents will therefore be HS; under the atomic theory, it would be II2S or IIS. Sulfhydric acid is therefore formed of 1 equivalent of sulphur and 1 of hydrogen, and the weight of its equivalent is 17. Bisulphide of Hydrogen, IIS2. § 153. Sulphur forms a second compound with hydrogen, an oleaginous, yellowish liquid, containing a greater quantity of sul- phur than sulfhydric acid; hut this quantity has not yet been determined with accuracy, because it is difficult to obtain the bisulphuret of hydrogen in a state of purity. It is prepared by pouring a solution of polysulphide of calcium or potassium into chlorohydric acid. The fluid becomes milky, and is poured into a large funnel, the aperture of which has been closed. In a short time, the bisulphide of hydrogen collects in the narrow part of the funnel, in the form of a yellow liquid, and is separated by carefully uncorking the funnel, until the liquid, which is heavy, runs off. Bisulphide of hydrogen is preserved only by contact with moderately concentrated hydrochloric acid, for it decomposes rapidly with pure water or the air; sulfhydric acid gas being disengaged, and sulphur separated. We have seen (§ 150) the use made of this spontaneous decomposition of the bi- sulphide of hydrogen, to obtain liquid sulfhydric acid. This substance is supposed to be composed of 1 equivalent of hydrogen and 2 of sulphur, and has been represented by the for- mula HS2. * It is probable that the density of the vapour of sulphur at its lowest point of evaporation, a little above the melting point of sulphur, is only 2.22. See Berzelius, Lehrbuch, I. 180, 5th edition.—J. C. B. SULPHIDE OF NITROGEN. 207 COMBINATION OF SULPHUR WITH NITROGEN. Sulphide of Nitrogen, NS3. §154. If dry ammoniacal gas be passed through the perchloride of sulphur, we obtain, at first, a brown flaky powder, the formula of which is , NH3,SC12. But, if the action of the ammonia be continued, the brown matter absorbs an additional quantity of it, and changes into a yellow substance, the formula of which is 2NH3,SC12. If the yellow body be treated with water, it decomposes into the chlorohydrate and hyposulphite of ammonia, which dissolve, and a yellow powder, consisting of free sulphur and sulphide of nitrogen. The powder is rapidly washed with a little water, dried under the receiver of an air-pump, and treated several times with ether, which dissolves the free sulphur, and leaves the sulphide of nitrogen. Sulphide of nitrogen is a yellow powder, which decomposes slowly into sulphur and nitrogen at a temperature a little above 212° ; but when suddenly heated, it decomposes with an explosion. Water decomposes it slowly at the ordinary temperature—much more rapidly at the boiling point. § 155. Sulphide of nitrogen can be exactly analyzed, by care- fully and gradually heating a mixture of a known weight of it and metallic copper in the apparatus described (§108). The sul- phur combines with the copper, and the nitrogen is disengaged. We have seen how the proportion of this last substance was as- certained, and the peculiar caution necessary in the experiment. Sulphur may also be directly determined by decomposing the sulphide of nitrogen by nitric acid, which converts the sulphur into sulphuric acid, which is precipitated by the chloride of barium. Sulphide of nitrogen has thus been found to contain 1 eq. nitrogen 14 22.58 3 “ sulphur 48 72.42 62 100.00 The formula NS3 corresponds to that of nitrous acid NOa, or that of ammonia NIL. 208 SELENIUM. Equivalent Se — 39.5 (493.75 0 = 100.) § 156. Selenium,* like sulphur, may be obtained in three states; solid at the ordinary temperature, it becomes liquid at 392°, and gaseous at about 300°. Solid selenium is of a deep brown colour, conchoidal and vitreous, with a fracture. The edges of fracture are often so thin as to be translucent, when it exhibits, by trans- mitted light, a beautiful red colour. This is also its colour when very finely divided, or when a drop of liquid selenium is pressed between two plates of glass. Selenium does not pass suddenly, like sulphur, from the liquid to the solid state, but becomes viscid before assuming the latter condition, and may then be drawn out in very fine threads ; whence, it has not yet been obtained in crystals by melting. The density of selenium varies according to its molecular condition, being 4.28 in vitreous selenium, and 4.80 in that which is granular and slowly cooled. Melted selenium is of a very deep brown colour; its vapour of an intense yellow. Selenium is combustible, burning with a bluish flame, and exhal- ing a fetid odour of horse-radish, which is characteristic. Seleni- ous acid and oxide of selenium are formed by the combustion, to the latter of which the fetid odour is owing. Selenious acid is soluble in water, and its solution is readily decomposed by substances which have a strong attraction for oxygen: thus sulphurous acid reduces it and passes into the state of sulphuric acid. The sele- nium, becoming free, is precipitated in the form of a red powder. The compounds of sulphur and selenium are very analogous, for which reason they are generally studied in connection with each other. Selenium is found in nature principally in the state of selenide of lead; and in studying this compound, we shall describe the mode of extracting it. § 157. Two combinations of selenium with oxygen are knowTn; selenious acid Se02 and selenic acid Se03, which correspond to sulphurous acid SOa and sulphuric acid S03. Chemists admit also COMBINATIONS OF SELENIUM WITH OXYGEN. * Selenium was discovered in 1817, by Berzelius. SELENIOUS ACID. 209 the existence of a third oxide, to which they attribute the fetid odour disengaged by selenium when burning in the air; but its properties are not known. §158. When selenium is burned in oxygen, it is converted into selenious acid. In order to obtain the acid by the com- bustion of selenium, a fragment of selenium is introduced into a bent tube abc (fig. 221), of which the end a connects with a small glass retort, containing chlorate of potassa. The chlorate is heated so as to give off oxygen ; and the portion b of the tube con- taining the selenium is heated. It burns with a blue flame, and the selenious acid con- denses in the upper part of the tube, in the form of white acicular crystals. Selenious acid may also be obtained by oxidizing selenium by concentrated nitric acid, or, better still, by a mixture of nitric and chlorohydric acids. It dissolves in the state of selenious acid, and if the solution be evaporated, the acid is obtained in the form of a white mass. Under the same circumstances, we have seen sulphur converted into sulphuric acid. Selenious acid is very soluble in water. It is not very retentive of oxygen; as many substances abstract the gas from it. Iron and zinc decompose dissolved selenious acid, and precipitate the sele- nium in the form of a red powder. Sulphurous acid effects a similar decomposition. § 159. The composition of selenious acid may be ascertained by finding the weight of this acid produced by 1 gramme of selenium treated with nitric acid. It may also be obtained by finding the quantity of selenium afforded by 1 gramme of selenious acid de- composed by sulphurous acid. Lastly, some selenite, that of lead or silver, for example, may be analyzed. Selenious acid has thus been found to contain Selenious Acid, SeOa. Fig. 221. Selenium 71.IT Oxygen 28.83 moo § 160. By heating together a mixture of nitrate of potassa and selenium or selenide of lead, we obtain the seleniate of potassa, which is purified by successive crystallizations. The seleniate of potassa, dissolved in water, is decomposed by a solution of nitrate of lead, insoluble seleniate of lead being precipitated, which is col- Selenic Acid, Se03. 210 SELENIUM. lected on a filter. The seleniate of lead, well washed, is suspended in water, and subjected to the action of a current of sulfliydric acid gas, whereby sulphuret of lead is precipitated in the form of a black powder, and the hydrated selenic acid is dissolved in the water. We have, in fact, Pb0,Se03+HS=PbS+Se03,H0. The solution of selenic acid may be concentrated by heat, the boiling point of the fluid rising to about 554°. But if we endeavour to concentrate it still further, the selenic acid is decomposed and oxygen disengaged. Selenic acid is decomposed by chlorohydric acid, disengaging chlorine and forming selenious acid. Sulphuric acid does not act on selenic, whilst it instantly decomposes selenious acid. When we wish to precipitate selenium from selenic acid, the latter must first be converted into selenious acid, by boiling its solution with chlorohydric acid. Sulphurous acid is then added, and the fluid again boiled. § 161. Selenic is a very powerful acid, closely resembling sul- phuric acid in its properties. Its composition has been determined by the analysis of some seleniate, the seleniate of lead, for ex- ample. A known weight p of seleniate of lead is suspended in water, and decomposed by sulphuretted hydrogen, which precipi- tates the lead in the state of sulphuret of lead. The sulphuret being collected on a small filter, is washed, dried, and calcined with the filter in a platinum crucible.* The sulphuret of lead, thus partly converted into a sulphate, is wholly converted by pour- ing on it nitric acid and some drops of sulphuric, and then calcin- ing it to redness. The sulphate of lead is weighed, and from its weight we deduce, by calculation, the quantity p' of oxide of lead which exists in the weight p of the seleniate of lead. The quantity of selenic acid is therefore {p—p')- The proportion of selenium in the weight {p—p') of selenic acid, is easily ascertained. It is sufficient to concentrate by evaporation the liquid obtained after the separation of the sulphuret of lead on the filter, and boil this concentrated liquid, first with hydro- chloric acid, which converts the selenic into selenious acid, then with sulphurous acid, which decomposes the selenious acid and precipitates the selenium. Let p" be the weight of selenium ob- tained, {p—p'—p") will be the weight of oxygen which constitutes selenic acid with the weight p" of selenium. We thus find that selenic acid contains Selenium 62.20 Oxygen 37.80 100.00 * This is a, dangerous operation except for an experienced chemist, and there- fore a porcelain crucible is to be preferred.—J. C. B. SELENIC ACID. 211 § 162. In short, the two known compounds of selenium contain Selenious acid Selenium Oxygen.. .... 71.17 .... 28.83 100.00 Selenic acid... ’ Selenium Oxygen.. .. 62.20 .. 37.80 100.00 If we refer the composition of the two acids to the same quan- tity 100 of selenium, it will be expressed as follows : Selenious acid Selenium.. Oxygen... . 100.00 . 40.51 140.51 Selenic acid.. Seleniuir Oxygen ... 100.00 ... 60.77 160.77* Therefore, for the same quantity of selenium, selenic acid con- tains 1| times as much oxygen as selenic acid. The most simple manner of expressing the formulae in equivalents of the composi- tion of these bodies, is to say, that selenious acid is composed of 1 equivalent of selenium and 2 of oxygen, and selenious acid of 1 equivalent of selenium and 3 of oxygen. If we grant this hy- pothesis, the weight of selenium will evidently be given by one of these proportions: 28.83 : 71.17 : : 16 : x \ , _qQ r 37.80 : 62.20 : : 24 : x /whence *-39*5‘ We shall soon see that this same hypothesis represents in the simplest manner possible the composition of selenohydric acid. § 163. We must, henceforward, remember that there is a physi- cal law which Ave have not hitherto applied in fixing the composi- tion of bodies by equivalents, and upon which we shall frequently insist hereafter. We have seen in the introduction (§40), that when two bodies are similarly composed, they generally affect nearly identical crystalline forms; and reciprocally, when two compounds have nearly identical crystalline forms, if they are iso- morphous, they are generally similarly composed. Now, the com- parative examination of the sulphates and seleniates has shown that the seleniates and sulphates of the same base are isomorphous: they ought, therefore, to have similar formulae. If, therefore, we Avrite the formula of sulphuric acid S03, we should write the for- mula of selenic acid Se03, and, consequently, that of selenious acid Se02. Experiment has shown that the density of the vapour of seleni- * 40.51 : 00.77 :: 1 : 1} :: 2 : 8. 212 SELENIUM. ous acid is 4.0 ; consequently, 1 volume of gaseous selenious acid contains 1 volume of oxygen. We have seen that gaseous sulphurous acid also contained its volume of oxygen. COMBINATION OF SELENIUM WITH HYDROGEN. Selenohydric Acid, HSe. §164. Selenium forms a gaseous compound with hydrogen— selenohydric acid, analogous to sulfhydric acid, and is obtained by decomposing the selenide of iron by chlorohydric acid. Seleno- hydric acid dissolves in water, but its solution decomposes in the air, in the same manner as sulfhydric acid, selenium being de- posited in the form of a red powder. Selenohydric acid is composed of Hydrogen 1 2.47 Selenium 39.5 95.53 40 lOOO The equivalent of selenohydric acid is therefore 40.5, and its formula HSe. 213 TELLURIUM. Equivalent Te = 64.5 (806.5 0 = 100). § 165. Tellurium* is very rare : it is found in nature, sometimes isolated, but more frequently combined with the metals, principally with gold, silver, bismuth, and lead. We shall subsequently see how it is separated from the bismuth. Tellurium has the physical properties of a metal, and in appearance greatly resembles anti- mony, but it approximates, on the other hand, in the properties of its compounds, to selenium and sulphur. Tellurium is of a silvery white colour, and has a bright metallic lustre. It melts at a dull red-heat, and, by careful cooling, crys- tallizes in large brilliant plates, which are very manifest in its fracture. From the various planes of cleavage, it will be readily seen that the primary form of crystallized tellurium is a rhombo- hedron. Tellurium may be rendered gaseous, but it requires a very high temperature. It may, indeed, be distilled, but the dis- tillation cannot be performed in the earthen or porcelain retorts used in our laboratories. The distillation of slightly volatile substances is greatly facili- tated by heating them in a current of gas which exerts on them no chemical action. Volatile substances sensibly give off vapours at a temperature far inferior to their boiling point. Thus, water, which boils at 212°, under the ordinary pressure of the atmo- sphere, disengages, at the ordinary temperature, considerable vapour; and the weight of this vapour cannot exceed, in a limited space, a certain maximum, which depends on the temperature ; but it can be conceived that, if these vapours are removed as soon as formed, the maximum will not take place, and new vapour will be constantly generated, until the substance is entirely volatilized. In order to distil tellurium, it is placed in a small platinum dish, introduced into a porcelain tube arranged in a reverberatory fur- nace. A current of dry hydrogen gas is passed into one end of the tube, and to the other, which should project somewhat from the furnace, a tube is fitted, to allow the escape of the gas. A current of hydrogen is first passed through the apparatus, so as to expel the atmospheric air completely; the porcelain tube is then * Tellurium was discovered in 1782, by Muller, of Reichenstein, in the gold-mines of Transylvania ; but to Klaproth we are indebted for a knowledge of its princi- pal properties. 214 TELLURIUM. heated as highly as possible, still keeping up the current of gas. The sublimed tellurium condenses in the anterior and colder portion of the tube. The density of tellurium is 6.26, which is very considerable, and in which respect it again resembles the metals properly so called. Tellurium, heated in the air, burns with a bluish flame, exhaling a peculiar odour difficult to describe. COMBINATIONS OF TELLURIUM WITH OXYGEN. § 166. Tellurium forms two compounds with oxygen, tellurous acid Te02, and telluric acid Te03, which are obtained in the manner directed for selenious and selenic acid. We shall not stop to de- scribe them. 215 CHLORINE. 'Equivalent Cl = 33.5 (443.75 0 = 100). § 167. Chlorine* is a gas readily distinguished from all those we have hitherto studied. In fact, all those gases are colourless, whilst chlorine is of a greenish yellow, to which property is due its name (from greenish yellow). If chlorine be com- pressed, so as to reduce it to a volume 5 times less than it oc- cupies at the ordinary pressure of the atmosphere, it becomes liquid. The density of the liquid, which is of a greenish-yellow hue, is 1.33. It has never yet been congealed by any degree of cold. The density of gase- ous chlorine is 2.44: that is, nearly 2J times that of air. § 168. Chlorine is obtained by treating the peroxide of manganese by chlorohydric acid. The pulverized per- oxide of manganese is placed in a glass flask (fig. 222), and chlorohydric acid poured thereon. A discharging-tube, fitted to the flask, conveys the gas into a bell-glass in the ? water-cistern. In this reac- tion, the peroxide of manga- nese gives off' its oxygen to the hydrogen of the chlorohydric acid, one-half of the chlorine disengaged combines with the man- ganese to form the protochloride of manganese, and the other half is set free. Fig. 222. f Manganese Peroxide of manganese... Chlorohydric acid.... Oxygen. Hydrogen.. .Water. Protochloride of man- ganese. t Chlorine. Chlorine. \ Chlorine Is set free. MnOs+2HCl=MnCl+2HO+Cl. The flask is slightly heated to facilitate the process. Chlorine may he more steadily generated by substituting a mix- * Chlorine was discovered in 1774, by Scheele. 216 CHLORINE. ture of sea-salt and sulphuric acid for the chlorohydric acid. We put into a flask 1 part of finely-powdered peroxide of manga- nese, 4 parts of sea-salt or chloride of sodium, and 2 of the con- centrated sulphuric acid of commerce, diluted with its weight of water. The chloride of sodium in contact with the sulphuric acid and water gives rise to sulphate of soda and chlorohydric acid: Chloride of sodium.. Sodium. Chlorine. Chlorohydric * acid Oxide of / Sodium. Water... Hydrogen (.Oxygen., i Sulphate of \ soda. Sulphuric acid. NaCl+HO+SO,=NaO,SO,+HCl. Chlorohydric acid, in contact with the peroxide of manganese, reacts as previously stated, chloride of manganese and sulphate of soda formed, and chlorine disengaged. The sulphuric acid, in excess, acts on the chloride of manganese as on the chloride of sodium, decomposing it with the assistance of the tvater, so that an additional quantity of chlorohydric acid and sulphate of man- ganese result; hence the ultimate products of the reaction are the sulphates of soda and manganese, which remain in the flask, and all the chlorine of the chloride of sodium, which is disengaged in the state of gas. The final reaction is represented by the following equation: NaCl+Mn0a+2S03=Na0,S0a+Mn0,S08+Cl. Chlorine is more soluble in water than the simple gases we have hitherto studied, 1 volume of water dissolving 2 of chlorine. This great solubility of chlorine in water, prevents us from pre- serving it over this fluid, and even embarrasses our collecting it there. However, it may be done by working rapidly and causing the discharging-tube to ascend to the upper part of the bell-glass, so that the gas is not obliged to pass through the water in the form of hubbies, and is less exposed to its dissolving power. Chlorine cannot be collected over mercury, because it combines immediately with this metal, even at the ordinary temperature. Dry chlorine may be obtained in the following manner:—After having conveyed the gas first into a washing-bottle B (fig. 223), containing a little water, which retains the chlorohydric acid which might have come over, it is made to pass through a tube ah, filled with chloride of calcium, or a tube curved like the letter U, con- taining pumice-stone soaked in concentrated sulphuric acid. These substances rapidly absorb the water, and dry the gas, which is then conveyed by another tube to the bottom of a small-mouthed bottle C. The chlorine, from its great density, occupies the inferior portion, and rises successively in the bottle, driving out the atmo- CHLORINE. 217 Fig. 223. spheric air. After some time, the bottle may he supposed to be filled with chlorine, the tube is slowly withdrawn, and the bottle closed with a ground-glass stopper. The aqueous solution of chlorine is often used in the laboratory and in the arts. For this purpose it is best prepared in a Woolfs apparatus. The gas which does not dissolve in the first bottle, traverses the fluid contained in the second, third, and so on. The aqueous solution of chlorine and its gas have the same colour. When one of the bottles of the preceding apparatus is surrounded with ice, a flaky crystalline substance is soon formed, of a more intense greenish yellow than the fluid surrounding it. This substance is a combination of chlorine with water, a hydrate of chlo- rine, containing 28 of chlorine and 72 of water. The crystals may be easily isolated, if the external temperature be sufficiently low. They may be collected in a funnel, allowed to drain com- pletely, and then rapidly compressed between sheets of bibulous paper, previously cooled. They are then introduced into a curved tube abc (fig. 224) closed at a. The portion ab of the tube con- taining the hydrate is kept cold with ice, while the opposite end be is sealed in a lamp. The hydrate of chlorine decomposes at a few degrees above 32°. If we heat the portion of the tube contain- ing the hydrate of chlorine, by plunging it into water at 95°, the crystalline matter will be seen to separate into two liquid strata, one of which, of a deep-yellow, Mis to the bottom of the tube, and is liquid chlorine; the other, of a much lighter hue, is a saturated solution of chlorine in water. If the leg be of the tube be cooled with ice, the liquid chlorine boils in the leg ab, and condenses in the coldest part of be: it is thus separated from the aqueous solution. Chlorine has powerful affinities. It combines directly with hy- drogen, and an explosion always takes place when a lighted taper is plunged into a mixture of the two gases. It combines directly with a majority of the metals. Many substances, amongst them arsenic and antimony, take fire when thrown in a finely-powdered Fig. 224. 218 CHLORINE. state into a bottle filled with chlorine. If the vapour of water and chlorine are passed through a porcelain tube, the water is decomposed, oxygen being set free, and chlorohydric acid formed. The aqueous solution of chlorine often acts as a powerful oxidiz- ing agent: thus, it instantly converts sulphurous into sulphuric acid. The water in this case is decomposed, chlorohydric acid being formed, and oxygen in the nascent state added to the sul- phurous acid. S03+Cl+HO==HCl-f S03. The solution of chlorine may be preserved unchanged in the dark in a well-stoppered bottle; but, when exposed to solar light, it decomposes the water, chlorohydric and hypochlorous acids being formed. 2C1+H0=C10+HC1. Chlorine is used in the arts for bleaching linen and cotton fabrics, and, in general, to destroy vegetable colours. Vegetable colouring matters, like all substances of organic origin, are composed of carbon, hydrogen, oxygen, and sometimes of nitro- gen. Chlorine acts powerfully on many of them, and decom- poses them, by seizing their hydrogen to form chlorohydric acid; the colouring matter bleaches by decomposition. In a similar way, chlorine discolours ordinary writing-ink, the colour- ing principle of which is a combination of the sesquioxide of iron with an organic substance called tannin. If we wish to efface the writing completely, after having removed the cha- racters by chlorine water, we must wash it several times with weak chlorohydric acid, which completely dissolves the sesquioxide of iron. Without this precaution, the characters would reappear on w’ashing the place with a solution of prussiate of potash, which gives, with the sesquioxide of iron, a blue compound. But chlorine will not act on India ink, nor on printing-ink, because the colouring matter of these inks is very finely-divided carbon, which does not combine directly with chlorine. Chlorine is also used to destroy the putrid miasmata arising from the decomposition of organic matter. These miasmata are owing to the presence of organic substances in the air, so minute, how- ever, that chemical analysis has hitherto been unsuccessful in de- tecting them. Chlorine destroys these substances by taking away their hydrogen. Chlorine acts as a poison on the animal economy. Respired in small quantities, it excites coughing, and long exposure to its influence may produce more serious symptoms, bloody expecto- ration, etc. etc. CHLORIC ACID. 219 COMBINATIONS OF CHLORINE WITH OXYGEN. § 169. These combinations are very numerous, five of them being well ascertained, and others more complex exist, which may be regarded as resulting from the union of the former with each other. The five most important compounds are— 1. Hypochlorous acid CIO 2. Chlorous acid C103 3. Hypochloric acid C104 4. Chloric acid C10s 5. Perchloric acid C107. We shall begin with chloric acid, which may be considered as the starting point of all the rest. Chloric Acid, C10s. §170. When a concentrated solution of potassa is saturated with chlorine, there separate, after some time, white crystalline spangles of chlorate of potassa, while the fluid contains a large quantity of chloride of potassium, and a small proportion of chlo- rate of potassa retained in solution. Reaction takes place between 6 equivalents of chlorine and 6 equivalents of potassa; 5 equiva- lents of chloride of potassium KC1 are formed, and 1 equivalent of chlorate of potassa K0,C105; that is, we have the equation 6Cl+6K0=5KCl-fK0,C10s. The chlorate of potassa is purified by solution in boiling water, the greater portion being deposited in the form of crystals during the cooling of the liquid. Chloric acid, and all the other compounds of chlorine with oxygen, are prepared by means of chlorate of potassa. In order to obtain chloric acid, we pour into a solution of chlo- rate of potassa, an excess of hydrofluosilicic acid, whereby a gelati- nous precipitate of insoluble silicofluoride of potassium is formed, and the chloric acid remains in the liquid. If only the quantity of hydrofluosilicic acid necessary to exactly precipitate the potassa were poured in, the chloric acid would remain alone in the solution. But, as the silicofluoride of potassum is a transparent gelatinous precipitate, scarcely to be distinguished in the liquid, it is impos- sible to know when the potassa is completely precipitated, and we are obliged to use an excess of hydrofluosilicic acid. The filtered liquid, containing therefore a mixture of chloric and hydrofluosilicic acids, is saturated with a solution of baryta, until it becomes slightly alkaline. The baryta forms with the hydrofluosilicic acid an insoluble salt, and a soluble chlorate with the chloric acid. The 220 CHLORINE. liquid is evaporated after having been again filtered, and crystallized chlorate of baryta is obtained. The chloric acid is isolated, by dissolving the chlorate of baryta in water, and carefully adding sulphuric acid until no precipitate is formed. The sulphate of baryta is separated on a filter, and the liquid, which contains only chloric acid, is placed under the receiver of an air-pump, where it is brought to the consistence of syrup. We cannot use heat to concentrate the solution of chloric acid, because it rapidly decomposes at a temperature exceeding 104°. Two acids are formed, one of which, perchloric acid C107, re- mains in the liquid; and the other, chlorous acid C103, is disen- gaged in the form of a yellow gas, or decomposes immediately into chlorine and oxygen, according to the temperature. Litmus paper, dipped into a solution of chloric acid, at first red- dens, but is soon as completely bleached as though it were dipped into a solution of chlorine. If a few drops of a concentrated solution of chloric acid be poured on a piece of linen or paper, and then gently dried, the parts which were moistened take fire and deflagrate. Chloric acid, mixed with a solution of hydrochloric, gives off chlorine copiously, as may be represented by the following equa- tion : C105-f 5HC1=6C1+5H0. Substances easily oxidizable decompose chloric acid by taking its oxygen: thus, by contact with chloric acid, sulphurous is changed into sulphuric acid; phosphorous into phosphoric acid. § 171. The composition of chloric acid may be easily deduced from that of chlorate of potassa, which is an anhydrous salt that can be accurately analyzed. By calcining a weight p of chlorate of potassa in a platinum crucible, oxygen will be disengaged, and there will remain a weight p' of chloride of potassium : (p—p') therefore represents the oxy- gen which existed in the chloric acid and in the potassa. If we knew the composition of the chloride of potassium, we could immediately ascertain the quantity c of chlorine and the quan- tity k of potassium which existed in the wreight p' of chloride of potassium. We would then find that a weight p of chlorate of potassa contained a weight k of potassium. c of chlorine. p—p' of oxygen. But, supposing that the composition of the chloride of po- tassium is unknown, we can readily ascertain it in the following manner: CHLORIC ACID. 221 By dissolving in water a known weight p' of chloride of potas- sium, and pouring into the solution an excess of nitrate of silver, the chloride of silver will be precipitated, which is easily collected, and its weight ascertained, after having been previously dried. We thus find that a weight p' of chloride of potassium gives a weight p" of chloride of silver. Let us admit, for a moment, that the composition of the chloride of silver is known; we then know that, in a weight p" of this substance, there is a weight c of chlorine. But, if the composition of chloride of silver were itself unknown, it would suffice, in order to ascertain it, to take 10 grammes of this substance, introduce it into a glass tube, and heat it in a cur- rent of hydrogen, when it would be brought to the metallic state, the chlorine separating in the state of chlorohydric acid. By weighing accurately the metallic silver remaining in the tube, we should have the composition of the chloride of silver. We have first determined the composition of chloride of silver by analysis; but it may also be done by synthesis. In fact, by dissolving 10 grammes of perfectly pure metallic silver in nitric acid, diluting the fluid with water, and then adding carefully chlorohydric acid, until it ceases to precipitate, the silver will be deposited in the state of a chloride, which can be easily washed by decantation, and then dried. The weight of the chloride ob- tained, diminished by 10 grammes, will give the weight of chlorine combined with 10 grammes of silver. Let us admit that potassa is formed by the combination of 1 equivalent of potassium with 1 equivalent of oxygen. Now, the chloride of potassium is obtained treating 1 equivalent of po- tassa KO, with 1 equivalent of chlorohydric acid HC1: K0+HC1=KC1+H0. Thus, the composition of chloride of potassium being known, we can thence deduce immediately that of potassa, by making this proportion : the quantity of chlorine combined with a certain quan- tity of potassium, is to the quantity of oxygen which would form potassa with this same quantity of potassium, as the equivalent of chlorine = 35.5 is to the equivalent of oxygen = 8. We therefore conclude from all these determinations that 100 parts of chlorate of potassa contain Potassium 31.95 Chlorine 28.93 Oxygen 39.12 Ioo7oo Now, since 31.95 of potassium require 6.52 of oxygen to form potassa, there remains, for chloric acid, 222 CHLORINE. Chlorine 28.93 Oxygen 32.60 6L53 100 parts of chloric acid are therefore composed of Chlorine 47.02 Oxygen 52.98 10000 The number we shall assume as the equivalent of chlorine, as will soon be seen, is 35.5. We shall thence find that the composi- tion of chloric acid corresponds to 1 eq. chlorine 35.5 5 “ oxygen 40 1“ chloric acid 75.5 § 172. We have just seen, that by boiling a solution of chloric acid, chlorous or hypochloric acid is disengaged, and perchloric acid, which remains in the fluid, is formed at the same time. If sulphuric acid be poured upon chlorate of potassa, the mix- ture assumes a brownish-yellow tinge, a yellow gas, hypochloric acid, is disengaged, and perchlorate and bisulphate of potassa are formed, which remain in the vessel. The reaction must be as- sisted by gentle heat, by placing the vessel in a water-bath. This experiment requires great caution, for hypochloric acid is a very explosive gas. The perchlorate of potassa is easily separated from the bisulphate by crystallization, being much less soluble than the latter salt. The perchlorate of potassa is more readily obtained in another way. When chlorate of potassa is heated in a glass retort, in the preparation of oxygen, the substance first melts, and then gives oxygen off for some time; but if the temperature is not con- tinually increased, the fluidity of the substance decreases, and a point arrives at which it assumes a doughy consistence; the evolution of oxygen then ceases, and commences only when the temperature is again raised. The saline mixture remaining in the retort is composed of perchlorate of potassa and chloride of potassium; it is pulverized, and treated with a small quantity of cold water, which dissolves nearly the wdiole of the chloride of potassium, while it scarcely affects the perchlorate of potassa, which is not very soluble. The residuum is then treated with boiling water, so as to dissolve it wholly. A larg6 portion of the perchlorate of potassa crystallizes on cooling. Perchloric acid is obtained from the perchlorate of potassa, by exactly following the process indicated for obtaining chloric acid Perchloric Acid, C107. HYPOCHLOROUS ACID. 223 from the chlorate. But, perchloric acid being much more fixed than chloric, the same precautions in concentrating the dilute solution are unnecessary; for the latter may be evaporated by heat, and the concentrated liquid may be distilled in a glass retort. The first portions which pass over are more watery; the tem- perature in the retort rises to 392°, and an acid is distilled, having a density of 1.65: it is perchloric acid, at its maximum of con- centration. This acid is liquid and colourless, and reddens litmus without bleaching it. It is much more fixed than chloric acid, not only when heated, but even in the presence of oxidizable sub- stances. Thus, when cold, it does not act on sulphurous acid. §173. The composition of perchloric acid is deduced from that of perchlorate of potassa, the salt being analyzed in a mode pre- cisely similar to the chlorate of potassa (§ 171). We thus find that perchloric acid contains, 1 eq. chlorine 35.5 38.78 7 “ oxygen 56.0 61.22 1 u perchloric acid 91.5 100.00 Hypochlorous Acid, CIO. § 174. If we pass a current of chlorine through a dilute solu- tion of potassa, without heat, no chlorate of potassa is formed, as is the case with a concentrated solution and an elevated tempera- ture ; but a liquid is obtained, which possesses, in the highest degree, the property of bleaching organic colouring matter, and which contains chloride of potassium and hypochlorite of potassa. The reaction takes place between 2 equivalents of chlorine and 2 of potassa, and is represented by the following equation: 2K0+2C1=K0,C10+KC1. If whiting be substituted for the potassa, we obtain a corre- sponding hypochlorite of lime. These products are of vast im- portance in the arts, as they are used for bleaching, and are often called bleaching salts. A solution of hypochlorous acid is obtained as follows: After pouring into a large bottle filled with chlorine gas, red oxide of mercury, ground and mixed with water, the bottle is corked and shaken. The chlorine is rapidly absorbed, forming an insoluble oxychloride of mercury and hypochlorous acid, which dissolves in the water. The filtered liquid contains hypochlorous acid only. Hypochlorous acid may also be procured free from water, by passing a current of dry chlorine slowly through a glass tube ab (fig. 225) containing oxide of mercury, and preventing the temperature from rising during the reaction. For this purpose, the tube ab is 224 CHLORINE. Fig. 225. surrounded with ice or cold water. Chloride of mercury is again formed, and an orange-yellow gas disengaged, which may be lique- fied by conveying it into a tube cooled by a mixture of ice and sea-salt. The temperature should not rise during the reaction; otherwise the hypochlorous acid would entirely decompose, and oxygen alone would be disengaged. The best oxide of mercury is obtained by decomposing, by an excess of potassa, the nitrate or the chloride of mercury, ■washing the precipitate, and heating it to a temperature of about 572°. Hypochlorous acid forms a deep-red liquid, which boils at about 68°, producing an orange-yellow vapour. Water dissolves at least 200 times its volume of the acid, and becomes of a beautiful yellow-colour. The vapour of hypochlorous acid detonates at a very slightly elevated temperature. The solution of hypochlorous acid exerts powerful oxidizing qualities, decomposing the solutions of protochloride of lead and of manganese, from which it precipitates the binoxide of lead Pb03 or the sesquioxide of manganese Mn303. A solution of chlorine does not produce this effect except under the influence of the solar rays. § 175. Hypochlorous acid is analyzed as follows: The gaseous acid is obtained by conveying the chlorine slowly into a tube well cooled and containing the oxide of mercury: to the other end of this tube is fitted a capillary tube, on which seve- ral bulbs A, B, C (fig. 226), of a capacity of 20 or 30 cubic centi- Fig. 226. metres (1 to 2 cub. in.) have been blown. Heat being applied to the portion ab of the tube, the gas decomposes as it reaches this part, and the bulbs are successively filled with a mixture of chlo- rine and oxygen, in the proportions in which the two gases exist in CHLOROUS ACID. 225 hypochlorous acid. When we have thus decomposed a sufficient quantity of gas to completely expel the air which originally filled the apparatus, each of the bulbs is closed, by projecting the blowpipe flame upon the points a, b, c, d of the capillary tube. Each bulb is then filled with a mixture of oxygen and chlorine in the proportions which form hypochlorous acid, this mixture balancing the pressure of the external atmosphere, at the surrounding temperature. If we open one end of one of the bulbs in a weak solution of potassa, the chlorine is absorbed and the alkaline fluid ascends into the bulb. We sink the bulb so that the level of the fluid shall be the same within and without; then apply the finger to the open end, and remove the tube. Let be the weight of the bulb with the fluid it contains. We then fill the bulb completely with the same alkaline fluid, and find its weight p". Lastly, after having washed and dried the bulb, it is weighed, and found to weigh p. It is evident that the ratio of the weights is equal to that of the volumes of chlorine and oxygen which enter into the composition of hypochlorous acid. Experi- ment shows that this ratio is as 2 to 1. We may hence conclude that hypochlorous acid is formed of 2 volumes of chlorine and 1 of oxygen, or 1 equivalent of chlorine, and 1 equivalent of oxygen. We shall therefore have for its composition in weight, 1 eq. chlorine 35.5 81.61 1 “ oxygen 8.0 18.39 1 “ hypochlorous acid 43.5 100.00 Direct experiment has given, for the density of hypochlorous acid gas, the number 2.977, which shows that it is composed of 2 volumes of chlorine and 1 volume of oxygen condensed into 2 volumes. In fact, 2 vol. chlorine weighing 4.880 1 “ oxygen “ 1.105 5M5 of which the half is 2.9925. If chlorohydric acid be poured into a concentrated solution of hypochlorous acid, we obtain a copious evolution of chlorine. But, if the two liquids, greatly cooled, are mixed, chlorine is not dis- engaged ; it combines with the water and forms a hydrate of chlo- rine, which causes the liquid to assume the solid form. Chlorous Acid, C103. § 176. If chlorate of potassa be treated with nitric acid, the chlorate dissolves in the liquid without discoloration, provided the temperature does not exceed 120° to 140° ; but, if nitrous acid be poured into the solution, or if the deutoxide of nitrogen be 226 CHLORINE. passed through it, reaction instantly ensues, and a yellow gas, which is chlorous acid, is evolved. The easiest way of obtaining this acid consists in heating a mixture of chlorate of potassa, nitric acid, and arsenious acid. The arsenious acid converts the nitric into nitrous acid, which, in its turn, reacts on the chloric acid, depriving it of its oxygen, and reducing it to the state of chlorous acid. The experiment is made in the following manner: We take 3 parts of arsenious acid, 4 “ of chlorate of potassa, and pulverize them together, rub them into a thin paste with water, and add a mixture of 12 parts of ordinary nitric acid, and 4 “ of water; introduce the whole into a flask which is filled to the neck, and heat it carefully in a water-bath. Chlorous acid is a greenish-yellow gas, which does not liquefy in a refrigerating mixture of ice and sea-salt. Water dissolves about 5 or 6 times its volume of it, and assumes a golden yellowT- colour. § 177. Chlorous acid cannot be analyzed in the manner pointed out for hypoclilorous acid, because, in the decomposition of chlorous acid by heat, a small quantity of perchloric acid is constantly formed, disturbing the results of the analysis. Chlorous acid combines with bases and forms well-defined com- pounds, but the combination requires some time to effect it. By pouring into a solution of chlorite of potassa, a solution of nitrate of lead, we obtain a yellowish-white precipitate of chlorite of lead Pb0,C103, which may be analyzed by changing it into a sulphate by sulphuric acid. We thus find that 100 parts of chlorite of lead give 88.62 of sulphate of lead, which contain 65.23 of oxide of lead: 100 parts of chlorite of lead are therefore com- posed of Oxide of lead 65.25 Chlorous acid 34.75 100.00 Now, the equivalent of the oxide of lead is 1394.5; the chlorite of lead is therefore composed of 1 eq. oxide of lead : 111.7 1 “ chlorous acid 59.5 1 “ chlorite of lead 171.2 Again, the equivalent 59.5 of chlorous acid corresponds to the following composition of the acid: 100.00 HYPOCHLORIC ACID. 227 1 eq. chlorine 35.5 59.66 3 “ oxygen 24.0 40.34 1 “ chlorous acid 59.5. 100.00 The composition of chlorous acid may also be directly ascer- tained, by finding the quantity of chlorine contained in 100 parts of chlorite of lead. To do this, we melt in a platinum crucible a known weight of chlorite of lead,* intimately mixed with twice its weight of carbonate of potassa or soda. The oxide of lead is converted into a carbonate, and the chlorous acid affords chloride of potassium. The melted mass is treated with hot water, which dissolves the chloride of potassium and the carbonate of potassa in excess, whilst the carbonate of lead remains in the state of an insoluble residue. The filtered liquid is supersaturated with nitric acid, and an excess of nitrate of silver poured in. The quantity of chlorine contained in the 100 parts of chlorite of lead is inferred from the weight of chloride of silver obtained. Hypochloric Acid, C104. § 178. This compound is obtained by causing concentrated sul- phuric acid to act on chlorate of potassa, but the experiment requires great caution, for hypochloric acid detonates with great violence, endangering the apparatus. Fused chlorate of potassa is preferred. The salt is coarsely broken, placed in a tube closed at one end (fig. 227), sul- phuric acid is poured into the tube, and to its open end is fitted a curved tube which is carried to the bottom of a well-dried bottle. The tube is slowly and carefully heated in a water- bath, and it is important not to plunge the tube into the bath as far as the i level of the mixture, as the gas might explode. A yellow gas is evolved, which cannot be collected over mer- cury, because that metal instantly decomposes it, nor over water, which dissolves it in large quantities. If the bottle containing the hypochloric acid be cooled in a refrigerating mixture, it liquefies and forms a red liquid, which boils at 68°. Hypo- chloric acid detonates with great violence, even in the liquid state. Water dissolves 20 times its volume of it. Hypochloric acid may be analyzed in the mode described for Fig. 227. * Great care should be used not to employ too high a heat, which would cer- tainly ruin the crucible.—J. C. B. 228 CHLORINE. hypochlorous acid, and we thus find it to be composed of 1 volume of chlorine and 2 volumes of oxygen, or of 1 eq. chlorine 35.5 52.59 4 “ oxygen 32.0 47.41 1 “ hypochloric acid 67.5 lOO.OO § 179. Hypochloric is not an acid per se; for with bases, it forms a chlorate and chlorite. It is therefore proper to regard it as analogous to liyponitric acid, that is, to suppose it composed of 1 equivalent of chloric and 1 equivalent of chlorous acid: we have, in fact, 2C104=C105+C103. RECAPITULATION OF THE COMPOUNDS OF CHLORINE AND OXYGEN. EQUIVALENT OF CHLORINE. § 180. The five compounds of chlorine and oxygen, which we have studied, present the following composition: Hypochlorous acid Chlorine 81.61 Oxygen 18.39 100.00 Chlorous acid Chlorine 59.66 Oxygen 40.34 moo Hypochloric acid Chlorine 52.59 Oxygen 47.41 100.00 Chloric acid Chlorine 47.02 Oxygen 52.98 100.00 Perchloric acid Chlorine 38.78 Oxygen 62.22 100.00 If we refer these compounds to the same quantity, 100 of chlo- rine, we shall find: Hypochlorous acid Chlorine 100.00 Oxygen 22.53 122.53 Chlorous acid Chlorine 100.00 Oxygen 67.61 167.61 Hypochloric acid Chlorine 100.00 Oxygen 90.14 190H4 CHLORINE AND OXYGEN. 229 Chloric acid Chlorine 100.00 Oxygen. 112.68 212938 Perchloric acid Chlorine 100.00 Oxygen 157.77 257777 The quantities of oxygen which combine with the same quan- tities of chlorine are to each other as 1: 3 : 4 : 5 : 7. These num- bers are among the most simple of those which can represent similar proportions. Let us therefore suppose that the first com- pound, hypochlorous acid, be formed of 1 equivalent of chlorine and 1 of oxygen: the numerical value of the equivalent of chlo- rine will be given by the proportion 18.39 : 81.61:: 8 : x, whence x = 35.5. The compounds of chlorine and oxygen will then take the fol- lowing formulae and numerical values : Hypochlorous acid CIO 43.5 Chlorous acid C103 59.5 Hypochloric acid C104 67.5 Chloric acid C10s 75.5 Perchloric acid C107 91.5 Now, we have found, whilst ascertaining the weight of these various acids which form a neutral anhydrous salt with 1 equiva- lent of a base, that The equivalent of chlorous acid is 59.5 “ chloric acid 75.5 “ perchloric acid 91.5 The formulae of chlorous, chloric, and perchloric acids, as we have just defined them, are therefore verified by the composition of the neutral salts. It is possible that the formulae of the hypo- chlorous acid should be C1303, and that of hypochloric acid C1308 =C103,C105. It will be seen, by this splitting of the formula that hypochloric acid may be considered as composed of definite proportions of chlorous and chloric acids. We shall therefore adopt the number 35.5 as the equivalent of chlorine. We shall subsequently see, that this equivalent possesses the property of giving the simplest possible formulae to the nume- rous compounds formed by chlorine. We have seen that, in chloric acid, 1 volume of chlorine was combined with 2J of oxygen, or 2 volumes of chlorine with 5 of oxygen : that, in perchloric acid, there were 1 volume of chlorine and 3J of oxygen, or 2 volumes of chlorine and 7 of oxygen. Now, 230 CHLORINE. the equivalent in volume of gaseous oxygen being represented by 1 volume, it is evident that the equivalent in volume of chlorine becomes 2 volumes. § 181. If we admit the hypothesis (§ 88) that all simple gases contain, for equal volumes, the same number of atoms, we may say that, in chloric acid, 2 atoms of chlorine have combined with 5 atoms of oxygen, and in perchloric acid, 2 atoms of chlorine have combined with 2 atoms of oxygen. The equivalent of chlorine = 35.5, corresponds therefore to 2 atoms, and the atomic weight of chlorine is 17.75, that is, one-half of the equivalent. The atomic formulae of the compounds of chlorine and oxygen would he written as follows : Hypochlorous acid C120 or 610 Chlorous acid C1203 6103 Hypochloric acid C1204 6104 Chloric acid C1205 610s Perchloric acid C1207 C107 COMBINATION OF CHLORINE WITH HYDROGEN. Chloroiiydric Acid, IICl. § 182. Chlorine and hydrogen combine directly: if a lighted taper be brought near the mouth of a small bottle containing a mixture of these two gases, they combine with an explosion. Ex- plosion also takes place if a bottle containing the mixture be exposed to the direct rays of the sun. If the bottle be exposed to diffuse light, the two gases still combine, but slowly, and the time required is in proportion to the degree of light. Lastly, in absolute darkness, the two gases appear to have no action on each other. We here see that light produces the same effect as heat. These gases may be combined so as to ascertain the proportions in which they unite. We select a balloon and a bottle of exactly the same capacity, and grind the neck of each, so that the former exactly fits the latter. The two vessels being perfectly dried, we fill the bottle (% 228) with chlorine, and the balloon (fig. 229) with hydrogen, both carefully dried, and then fit the balloon to the bottle: Ave thus have equal volumes of chlorine and hydrogen. To facilitate the admixture, the balloon is in- verted for a few moments (fig. 230): the chlorine, by virtue of its greater density, has a tendency to descend Fig. 228. Fig. 229. Fig. 230. CHLOROHYDRIC ACID. 231 into the balloon, and the hydrogen, on the contrary, rises into the bottle. The apparatus is left in a well-lighted room, but not ex- posed to the direct rays of the sun. The green colour of the chlo- rine fades rapidly, and when no longer perceptible, the apparatus is exposed, for a few moments, to the solar rays, which complete the combination without danger of explosion. The apparatus being taken apart under mercury, we observe that no gas escapes, and that the mercury does not rise in the vessels. Thus, the hydrogen and chlorine, by combining, have produced a gaseous compound which has preserved the same volume under the same pressure. The want of colour of the gas and the non-alteration of the mercury prove that no more free chlorine remains; but an excess of hydrogen may exist. We can prove that there is no more hydrogen, by introducing a small quantity of water into the vessel, when the gas is then entirely absorbed, and the mercury fills the vessel. The water introduced has become strongly acid. This experiment proves that 1 volume of hydrogen combines with 1 volume of chlorine, and produces 2 volumes of a very acid gas, highly soluble in water. This gas is chlorohydric or hydro- chloric acid. §183. Chlorohydric acid gas is obtained by treating common salt or chloride of sodium by concentrated sulphuric acid; the water contained in the sulphuric acid taking part in the reaction. f Chlorine Chloride of sodium... Sodium Oxygen. Sodav acid. Concentrated sulphuric acid Water.. Sulphuric acid Hydrogen. Sulphate of soda. The reaction is represented by the following equation: NaCl+S03,H0=Na0,S03+HCl. The chlorohydric acid is collected in a well-dried bell-glass, over mercury. § 184. Chlorohydric acid gas is colourless, gives off copious fumes in the air, which do not form if the atmosphere be perfectly dry. Atmospheric air always contains a certain quantity of va- pour of water, with which the chlorohydric acid gas combines, and the compound which results, possessing less tension than that of pure water, is, consequently, precipitated in a form of mist. Chloro- hydric acid is very soluble in water, which, at the temperature of 82°, dissolves more than 500 times its volume of it. The solu- bility diminishes as the temperature rises, so that at 68° water dissolves only 460 times its volume of the gas. The absorption of the gas by water is instantaneous, and is demonstrated as was done in the case of ammonia (§ 128). The density of the liquid acid, concentrated in the cold, is 232 CHLORINE. 1.21. When heated, it first gives off a considerable quantity of acid gas, which soon ceases, and an acid liquid passes over, pre- senting an unvarying composition, the boiling point of which is 230°. A concentrated solution of the acid gives off copious fumes in the air. If a very dilute solution be distilled, more water than acid passes over at first, and the liquid concentrates in the retort, until it attains the composition of the normal liquid, which boils at 230°. The solution of chlorohydric acid is one of the most common re- agents used in a laboratory. In order to prepare it, we put into a large balloon (fig. 231) equal parts of common salt and oil of vitriol, Fig. 231. to which is added one-third of its weight of water. The balloon connects with a tubulated bottle, which serves as a washing-bottle, and retains the small portion of sulphuric acid brought over by the gas. Succeeding the first bottle, are two others of larger size and three-fourths filled with water. The tubes conveying the gas do not dip deep into the liquid. As the solution becomes more and more dense as it concentrates, the upper strata of the fluid are always less charged, and, consequently, more fitted to dis- solve the gas rapidly. The liquid acid is rarely prepared in the laboratory, being manufactured on a large scale in chemical works, and found at a very cheap rate in commerce. It is obtained by decomposing common salt by sulphuric acid; but, instead of earthen vessels, large cast-iron cylinders, arranged horizontally in a furnace, are used, which send the acid gas into stoneware receivers, having two mouths, and half filled with water, resembling exactly those repre- sented in figs. 181 and 182. CHLOROHYDRIC ACID. 233 § 185. The chlorohydric acid* of commerce is rarely pure, almost always showing a yellow tinge, owing to the presence of chloride of iron, and it also contains a small quantity of sulphuric and sometimes of sulphurous acid. It is readily purified by distil- lation ; but it is advisable first to pour into the fluid a small quan- tity of chloride of barium and shake it, so that the sulphuric acid may be precipitated in the form of sulphate of baryta. If it con- tains sulphurous acid, some bubbles of chlorine, passed through the liquid, will convert the sulphurous into sulphuric acid. Fig. 232 represents the apparatus used for the distillation of chlorohydric acid. The retort is heated in a sand-bath. Succeed- Fig. 232. ing the balloon-receiver, we arrange a bottle containing a small quantity of water, which retains the greater part of the acid gas driven from the solution by heat.f A solution of the pure acid is perfectly colourless. § 186. We have ascertained the composition of pure chlorohydric acid gas by synthesis; but it is more readily done by analysis. In order to do this, a known volume of gas is introduced into a bent tube (fig. 233), over the mercurial trough, and a globule of potassium,passed into this glass by means of a small iron wire, and deposited on the horizontal portion of the tube. An alcohol lamp being applied, the potassium decomposes the chlorohydric acid gas, Fig. 233. * Chlorohydric acid is often called muriatic acid in commerce: this is the name given to it by the older chemists. They regarded chlorine as a combination of muriatic acid and oxygen, and called it oxygenated muriatic acid. f A very simple and effectual method is to pour oil of vitriol through an S-tube into a flask or retort containing strong and common muriatic acid, whereby the gas is driven over into Woolf’s bottles charged with water, without the aid of heat until towards the close of the process. See fig. 195, § 114, for the arrangement, except that a small lamp may be used instead of the furnace.—J. C. B. 234 CHLORINE. by seizing upon its chlorine and setting free the hydrogen. The hydrogen is made to pass into the graduated tube in which the chlorohydric gas was measured, and we find its volume to occupy one-half of that previously occupied by the acid. Now, if from the density of hydrochloric gas... 1.2474 we deduct one-half of the density of hydrogen. 0.0345 there remain 1.2129 nearly one-half the density of chlorine: therefore, 1 volume of chlorohydric acid gas contains §• a volume of chlorine, and \ a vo- lume of hydrogen, without condensation. If we wish to know the composition by weight of 100 parts of chlorohydric acid, we make the proportions: 1.247 : 0.0345 : : 100 : x= 2.74 1 1.247 : 1.2129 : : 100 : y=97.26. Therefore, 100 parts of chlorohydric acid contain Hydrogen 2.74 Chlorine 97.26 This composition is expressed in another manner, by referring it to the equivalent 1 of hydrogen: we thus find Hydrogen 1 Chlorine 35.5 “30 Now, 35.5 is precisely an equivalent of chlorine. The acid therefore contains 1 equivalent of hydrogen and 1 of chlorine, and its equivalent weighs 36.5. 1 volume of chlorohydric acid gas contains J a volume of hydro- gen and \ a volume of chlorine. If we refer this composition to 2 volumes of hydrogen, we shall say that 4 volumes of the acid gas contain 2 volumes of hydrogen and 2 volumes of chlorine. The equivalent of chlorine in volume will be therefore represented by 2 volumes like that of oxygen, and that of chlorohydric acid by 4 volumes. We have seen (§ 124) that 1 volume of ammoniacal gas combines with | of chlorohydric acid gas, to form the chlorohydrate of ammo- nia : consequently, 4 volumes or 1 equivalent of ammonia combine with 4 Volumes or 1 equivalent of chlorohydric acid. The formula of chlorohydrate of ammonia is therefore NH3,HC1. We do not know any other compounds of chlorine and hy- drogen. COMBINATIONS OF CHLORINE WITH SULPHUR. § 187. Chlorine and sulphur combine in several proportions, but some of these compounds have only been obtained combined with CHLORIDE OF SULPHUR. 235 other chlorides. We shall here treat only of the two which have been isolated. The formula of the first is C1S2, which corresponds to no known compound of chlorine with oxygen in which the chlo- rine acts as the electropositive element, nor to any compound of sulphur, as the electropositive element, with oxygen. The formula of the second compound is CIS, which corresponds to hypochlorous acid CIO, or to hyposulphurous acid S202. In order to obtain the first compound, chlorine is combined with sulphur, so that the sulphur is in excess: the second is obtained when the chlorine predominates. The apparatus used is represented in fig. 234. Fig. 234. Chlorine is evolved in the balloon A by causing chlorohydric acid to react on the peroxide of manganese ; the gas is washed in the bottle B containing water, and then dried by being passed through a tube containing chloride of calcium. The tubulated retort D containing a certain quantity of sulphur, is connected with a tubulated receiver E, which is kept at a low temperature by a current of cold water from the vessel F. The retort containing the sulphur is heated to a temperature above 212°, while the chlorine is slowly disengaged and brought nearly to the surface of the liquid sulphur; and as it then comes into contact with an excess of vapour of sulphur, the first compound C1S2 only is formed, which distils over as fast as it is produced. The process is continued until the sulphur in the retort has nearly disappeared. The chloride of sulphur collected in the receiver, contains an excess of sulphur brought over by volatilization; but it is easily removed by redistillation, as the sulphur is much less volatile than the chloride, and remains in the retort. This chloride of sulphur is a reddish yellow liquid, having a peculiar, disagreeable odour: it boils at 138°. Its density, when 236 CHLORINE. liquid, is 1.687. The density of the vapour has been found by experiment to be 4.668. It decomposes by contact with water ; sulphur separating, and chlorohydric, sulphuric, and sulphurous acids being formed. It is composed of 2 eq. sulphur 32 1 “ chlorine 35.5 WL5 1 volume of it, in the gaseous state, is composed of 1 vol. chlorine 2.440 J “ 2.218 theoretical density 4.658 This theoretical density approaches, in fact, very nearly to the density 4.668, which has been found by experiment. If the solution of the preceding chloride be saturated with chlorine, it absorbs a large quantity of it, and furnishes a deep red fluid, which, for the same quantity of sulphur, contains a double quantity of chlorine. If this substance be subjected to the action of heat, the excess of chlorine which was in solution is at first evolved ; but the ebullition soon becomes regular at the tem- perature of 147°. The density of this chloride is 1.620. The density of its vapour .is 3.549. Its composition is 1 eq. sulphur 16 31.07 1 “ chlorine 35.5 68.93 5L5 100.00 1 volume of vapour contains vol. of vapour of sulphur 1.109 1 “ chlorine 2.440 ■049 § 188. The chlorides of sulphur are easily analyzed, as follows: We weigh a certain quantity p of it in a closed tube, then insert this tube, previously opened, into a bottle containing 1 litre, half filled with water: the bottle is corked and shaken. The chloride of sulphur is decomposed, forming chlorohydric, sul- phuric, and sulphurous acids, and a deposit of sulphur. The latter being separated by filtration, taking care not to lose a drop of the liquid, nitrate of silver is poured into the solution until no pre- cipitate any longer forms, and the precipitated chloride of silver is weighed after desiccation. Let P be its weight: if its composition be known, we know that it contains a weight p' of chlorine, and conclude from the experiment that a weight p of chloride of sul- phur contains p' of chlorine, and consequently {p—p') of sulphur. AQUA REGIA. 237 COMBINATION OF CHLORINE WITH NITROGEN. Chloride of Nitrogen, NC13. § 189. This compound is obtained by passing chlorine through a solution of chlorohydrate of ammonia, or any ammoniacal salt. The solution turns yellow, and yellow oleaginous drops soon form, which fall to the bottom of the vessel. Its formation is assisted by a temperature of 77° to 86°. The reaction takes place accord- ing to the equation NH3,HC1+6C1==4HC1+NC13. These oily drops are extremely dangerous to handle, for they often explode spontaneously, and may cause severe accidents. Hence, it is important to understand the circumstances under which this dangerous substance is formed, less wThen preparing it than to avoid its accidental generation. Chloride of nitrogen is an orange-yellow fluid, of a density of 1.653. It may be distilled unaltered under a less pressure than that of the atmosphere ; but its vapour detonates with extreme violence when it attains the temperature of 212°. Chloride of nitrogen detonates immediately, at the ordinary temperature, in contact with certain substances, principally with phosphorus, the fixed oils, the essence of turpentine. Its formula is NC13, corresponding to ammonia NII3. § 190. A mixture of chlorohydric and nitric acids is called aqua regia, a name conferred on it by the alchemists, because it dis- solves gold, which they regarded as the king of the metals. When a mixture of chlorohydric and nitric acids is heated, the liquid turns yellow, and if it be boiled, a yellow gas is disengaged, the odour of which recalls at once that of chlorine and of nitrous vapour. This gas is composed of a mixture of chlorine and two peculiar compounds, which we shall call hgpochloronitric and chloronitrous acids. The two compounds are evolved in different proportions, according to the composition of the aqua regia, and as the reaction has more or less progressed. Hypochloronitric acid is obtained by heating, in a bottle A (fig. 235), in a water-bath, an aqua regia made of 1 volume of nitric and 3 of chlorohydric acid. The gase- ous product is conveyed into the first bottle B, where some drops of liquid are deposited, then into a tube D filled with pieces of chlo- ride of calcium, which absorbs the moisture; lastly, through a bulb E placed in a refrigerating mixture. In order to judge of the colour of the gas, we generally place an empty bottle C in front of the bulb, and a similar one Gr after it; and the apparatus is terminated by the tube IT, containing a small quantity of water, Aqua Regia. 238 CHLORINE. which allows us to judge of the rapidity of the generation of the Fig. 235. The bottle C becomes of a slightly brownish citron-yellow colour, which is the peculiar colour of the gaseous mixture. The greater part of the hypochloronitric and chloronitrous gases con- denses in the bulb, in the form of a reddish-brown liquid, and the gas which arrives in the bottle G presents the ordinary colour of chlorine. When a sufficient quantity of liquid has condensed in the bulb, we seal the points a and b in a lamp, if we wish to preserve the product. With the proportions of nitric and chlorohydric acids we have supposed, the substance which at first condenses in the bulb is nearly pure hypochloronitric acid; it is a very volatile fluid, which boils at about 44°. Its composition is represented by the formula NOsCl#; and may be regarded as hyponitric acid in which 2 equivalents of oxygen have been replaced by 2 equivalents of chlorine. The reaction from which it arises is represented by the following equation: By prolonging the experiment, the condensed product contains proportionally larger quantities of chloronitrous acid. This last compound is slightly less volatile than hypochloronitric acid; its formula is N02C1: it represents nitrous acid of which 1 equivalent of oxygen is replaced by 1 equivalent of chlorine. Chloronitrous and hypochloronitric acids may be produced by direct combination of chlorine and the deutoxide of nitrogen, by conducting the gaseous products into a bulb cooled by a mixture of ice and crystallized chloride of calcium. When aqua regia acts on any substance, we may generally suppose that the following reaction takes place between the nitric and chlorohydric acids: NO*+3HC1=N02, Cl3+3H0 + Cl. If a metal be plunged in this liquid, it dissolves rapidly in the N05+2HC1=N08+2H0+2C1. AQUA KEGIA. 239 state of a chloride, as in a concentrated solution of chlorine. The metal meets, in fact, in the aqua regia, chlorine in a nascent state, that is, under circumstances in which the combination takes place most easily. Aqua regia thus acts as a very powerful oxidizing agent; for it converts sulphur into sulphuric acid, much more rapidly than nitric acid algne. This circumstance is owing, on the one hand, to the fact that nitric mixed with nitrous acid oxidizes more power- fully than nitric acid; and, on the other, that chlorine, in contact with water, acts itself as a powerful oxidizing agent, by forming chlorohydric acid and presenting the oxygen in the nascent state. 240 BROMINE. Equivalent Br=80 (100 0 = 100). § 191. Bromine* is liquid at the ordinary temperature, of a very deep brownish red-colour; almost black when the layer is thick, and of a reddish yellow by transmitted light when the layer is thin. It congeals at —4°, into a crystalline, laminated mass with a grayish tinge. It boils at 116.6°, and at ordinary temperatures the tension of its vapour is considerable. A drop of bromine, in a bottle, volatilizes immediately, filling the vessel with a brownish-red vapour. The density of liquid bromine is 2.97: that of its vapour is 5.39. Bromine has a peculiar, very disagreeable odour, whence its name (from a stench). Like chlorine, it acts as a poison on the animal economy, and affects severely the organs of respira- tion. In all its compounds, it bears a strong analogy to chlorine; its affinities, however, are less active, for chlorine drives it from its combinations. Like chlorine, it destroys organic colouring matter. Bromine in contact with water at the temperature of 32°, com- bines with a portion of the water, forming a crystallized hydrate of a brownish red colour, which is more fixed than that of chlorine, and is destroyed only at about 60° or 70°. Bromine may be procured from bromide of sodium by the Fig. 236. * Bromine was discovered in 1826, by Mr. Balard, in the mother waters of the salines of the Mediterranean. BROMIC ACID. 241 process adopted to procure chlorine from the chloride of sodium, it being only necessary to heat a mixture of bromide of sodium, peroxide of manganese, and sulphuric acid diluted with its weight of water. The mixture is introduced into a tubulated retort (fig. 236) by a funnel in the tubulure t. The neck of the retort con- nects by a cork with the adapter B, communicating with a receiver C, which is ceoled by a current of cold water, or by enveloping it in ice. The retort is heated in the water-bath, by placing it in a small kettle filled with water heated over a furnace. The reac- tion is exactly the same as with chlorine, sulphates of soda and manganese being formed, which remain in the retort, while bromine distils over and condenses in the receiver. We shall hereafter see how bromine is prepared in manufacto- ries. Its price has hitherto been too great to allow its extensive use in the arts. COMBINATIONS OF BROMINE WITH OXYGEN. Bromic Acid, BrOs. § 192. Bromic acid is obtained from the bromate of potassa, which is prepared by dropping bromine into a concentrated solu- tion of potassa, until no more bromine will dissolve in the liquid. The solution is boiled for some time, and, on cooling, deposits small crystals of bromate of potassa. Bromic acid is then extracted from the bromate of potassa, exactly as chloric acid from the chlorate of potassa. The dilute solution of bromic acid may be evaporated by gentle heat, to the consistence of a syrup; but, if the evaporation be carried further, the bromic acid decomposes. The composition of the acid is deduced from the analysis of bromate of potassa, in the same way as we deduced from the analysis of the chlorate the composition of chloric acid. We thus find bromic acid to be composed of Bromine 66.67 Oxygen 33.33 5M00 Admitting that the formula of bromic acid is BrOs, similar to that of chloric acid, we find for its composition, 1 eq. bromine 80 5 “ oxygen 40 120 Bromine probably forms several other compounds with oxygen, but they have not been hitherto studied. 242 BROMINE. COMBINATION OF BROMINE WITH HYDROGEN. Bromohydric Acid, IIBr. § 193. Bromine combines with hydrogen, with much more diffi- culty than chlorine. Thus a mixture of hydrogen and vapour of bromine does not inflame in contact with a lighted taper, and may be exposed to the direct rays of the sun without any combi- nation ensuing; but the combination takes place if the mixture be passed through a porcelain tube heated to redness. When bromide of sodium is treated with concentrated sulphuric acid, a fuming acid gas is disengaged, which is bromohydric acid; but it is impure, as it contains sulphurous acid and vapour of bro- mine, owing to the fact that it is decomposed by concen- trated sulphuric acid. Water, sulphurous acid and bromine are formed, Bromine. Bromohydric acid.. Sulphuric acid. Hydrogen. Oxygen.... Water. HBr+SOa=SOa+HO+Br. Sulphurous acid. Pure bromohydric acid gas may be obtained by decomposing bromide of phosphorus, by a small quantity of water. Bromine . Bromide of phosphorus. [ Phosphorus., fOxygen Phosphorous acid Bromohydric acid. Water The reaction is represented by the following equation : Hydrogen... PBr8+3H0=P03+3HBr. This experiment is made in the apparatus fig. 237, con- sisting of a thrice-bent tube abode open at both ends. We place at d some bits of phosphorus, and fill the leg de with small fragments of moistened glass. By the aperture a we pour in the bromine, which falls to b. The curve b being heated with a live coal, the bromine is volatilized, and meets the phosphorus, with which it combines; but the bromide of phos- phorus formed is instantly destroyed by contact with the water, and produces phosphorous acid, which remains in the tube, whilst the bromohydric acid is disengaged, and may be collected over mercury. Very little water should be in the tube, otherwise the bromohydric acid would be entirely dissolved. Ilydrobromic is a colourless acid gas, fuming in the air: its density is 2.731. It is decomposed by chlorine, which seizes its Fig. 237. BROMOHYDRIC ACID. 243 hydrogen to form chlorohydric acid, and frees the bromine, which appears in the form of a brown vapour. If the chlorine be in excess, chloride of bromine is formed. Bromohydric acid is ex- tremely soluble in water, and a concentrated solution gives olf co- pious fumes in the air. § 194. It is analyzed in the same way as chlorohydric acid, by decomposing in a bent tube a known volume of bromohydric acid by potassium: we thus find that 1 volume of it contains J volume of hydrogen. Now, if from the density of bromohydric acid ... 2.7310 we subtract half the density of hydrogen 0.0344 there remains 2.6966 that is, half the density of the vapour of bromine. Bromohydric gas is thus composed similarly to chlorohydric gas, containing a J vol. of hydrogen and a \ vol. of vapour of bromine. Its composi- tion by weight is 1 eq. hydrogen 1 1.24 1 “ bromine 80 98.76 1 “ hydrobromic acid 81 100.00 We shall take as the equivalent of bromine 80, and the equiva- lent of bromohydric acid will then be 81. The equivalent in volume of the gaseous acid will be represented by 4 volumes. 244 IODINE. Equivalent 1 = 127 (1587.5 0 = 100.) § 195. Iodine* is solid at the ordinary temperature, presenting the appearance of dark-gray spangles, possessing a high degree of metallic lustre. It melts at 224.6°, forming a brown or nearly black liquid; it boils at about 356°, and gives off a very deep violet-coloured vapour. Iodine gives off very appreciable vapours at the ordinary temperature, which are much more copious toward 120° or 140°, when they are of a beautiful purple-violet hue. From the colour of these vapours it has received its name (from violet). The vapour of iodine has a peculiar odour, analogous to that of chlorine. Iodine crystallizes readily. We often find, in the upper part of the bottles which contain it, perfectly regular crystals, deposited there by sublimation. It also crystallizes very readily from solution, which we shall see when treating of iodohydric acid. Water dissolves but a small proportion of iodine, about be- coming yellow, and it probably exists in this solution in the state of a hydrate. Water dissolves much larger quantities of iodine when it contains certain substances in solution, principally iodides or iodohydric acid, when it assumes a very deep brown colour. The density of solid iodine is 4.95; that of its vapour 8.716. It greatly resembles chlorine and bromine in its combinations, but its affinities are weaker. It does not destroy the majority of organic substances, and vegetable colours generally resist its action. It combines with several organic substances, imparting to them peculiar colours. It colours the skin brown, but the stain soon disappears. The most remarkable phenomenon of colouring is that presented by iodine with starch, for an extremely small quantity of it will colour a considerable mass of starch very deeply blue. Advantage is taken of this fact, in the laboratory, to detect the presence of iodine in liquids which are supposed to contain very small quan- tities of it; and by its means we can ascertain the existence of a millionth part of iodine in solution. The starch is used either in the state of paste, or dissolved in boiling water, and allowed to cool. * Iodine was discovered in 1812, by Courtois: its properties were studied by M. Gay-Lussac. IODIC ACID. 245 Iodine is one of the most active poisons, but is used medicinally in goitre and scrofulous diseases. Iodine is obtained from iodide of sodium, by treating the salt with the peroxide of manganese and sulphuric acid diluted with its weight of water, the same apparatus being used as for bromine (fig. 238). The iodine condenses in the form of crystalline scales Fig. 238. in the adapter and receiver. It can be more easily obtained by decomposing a solution of iodide of potassium by a current of chlorine, the iodine being precipitated in the form of a gray pow- der, which is washed with a little water, and purified by subli- mation. COMBINATIONS OF IODINE WITH OXYGEN. § 196. Three of these compounds are known, the first of which will not be described : 1. Hypiodic acid I04 2. Iodic acid I05 3. Hyperiodic acid I07. Iodic Acid, I05. § 197. Iodic acid is obtained by heating iodine with highly con- centrated nitric acid. When the iodine has entirely disappeared, the liquid is allowed to cool, and the greater portion of the iodic acid deposits in the form of crystals. Iodic acid may also be obtained from the iodate of potassa. This salt is prepared by adding iodine gradually to a boiling solu- tion of potassa until it is saturated. The liquid, on cooling, de- posits iodate of potassa, and iodide of potassium remains in solu- tion. The reaction resembles the production of chlorate of potassa in a similar manner. Iodate of potassa is dissolved in hot water, and a concentrated and boiling solution of chloride of barium poured into the still hot liquid, which precipitates iodate of baryta. It is washed, heated, and decomposed by sulphuric acid, forming insoluble sulphate of baryta, while the evaporated liquid deposits crystals of iodic acid. 246 IODINE. The best method of preparing iodic acid, in larger quantities, consists in putting equal parts of iodine and chlorate of potassa into a flask with 5 parts of water, to which a few drops of nitric acid have been added. By heating it, chlorine is given off copiously, and the iodine remains in solution in the state of iodic acid. The theory of this operation is very simple : the small quantity of nitric acid added, assisted by heat, decomposes a corresponding quantity of chlorate of potassa, forming a small quantity of nitrate of potassa and chloric acid, which parts with all its oxygen to a corresponding quantity of iodine, while chlorine is disengaged. The iodic acid formed reacts, in its turn, on the chlorate, decom- posing an additional quantity of it; wdiereby chloric acid is set free, and is decomposed in the same manner as before. The iodic acid combines, as fast as it forms, with the potassa of the chlorate, so that, at last, all the chlorate is converted into iodate, the small quantity of nitric acid originally added only serving to commence the reaction. A solution of chloride of barium in hot water, being poured into that of the iodate of potassa, a copious precipitate of iodate of baryta ensues, which is washed several times, and the iodic acid separated by sulphuric acid. Crystallized iodic acid contains 1 equivalent of water. If these crystals be heated, they lose at first a small quantity of water, but soon decompose into iodine and oxygen. The composition of iodic acid is deduced from the analysis of iodate of potassa, the analysis being like that of the chlorate. Iodic acid contains 1 eq. iodine 127 76.05 5 “ oxygen 40 23.95 167 100.00 The formula of the crystallized acid is IOs+HO. Periodic Acid, I07. § 198. A current of chlorine is passed through a solution of iodate of soda, to which carbonate of soda is added, and kept con- stantly boiling. If the liquid be then allowed to cool, periodate of soda is deposited in silky tufts, which are dissolved in nitric acid, and nitrate of silver is added, which precipitates periodate of silver but slightly soluble. By resolution in boiling nitric acid and cooling, periodate of silver is again deposited. Treated with water, periodate of silver decomposes into basic periodate of silver, which remains, and acid periodate, which dissolves. The solution, when evaporated, yields crystals of periodic acid. The crystals fuse at about 266° ; at a higher temperature, first lose their water of crystallization, and then decompose. They IODOHYDRIC ACID. 247 first change into iodic acid, by giving off oxygen, and then, iodic acid itself is decomposed into iodine and oxygen. The composition of periodic acid corresponds to that of perchlo- ric, and is represented by 1 eq. iodine 127 69.40 1 “ oxygen 56 30.60 1 “ periodic acid 183 100.00 COMBINATIONS OF IODINE WITH HYDROGEN. Iodohydric Acid, HI. § 199. The affinity of iodine for hydrogen being much more feeble than that of bromine, they do not directly combine, even when a mixture of hydrogen gas and vapour of iodine are passed through a porcelain tube heated to redness. If iodide of sodium be heated with concentrated sulphuric acid, iodohydric acid is not obtained, but only a mixture of sulphurous acid gas and vapour of iodine. There is a mutual decomposition of the sulphuric and iodohydric acids. Iodohydric acid.. Iodine. Sulphuric acid.. Hydrogen. Oxygen.. Sulphurous acid. Water. hi+so3=so3+ho+i. Iodohydric acid is obtained by decomposing iodide of phosphorus by a small quantity of water. Alternate layers of phosphorus, iodine, and broken glass, moistened with water, are introduced into a tube closed at one end (fig. 239), and gently heated. The iodide of phosphorus decomposes as fast as it forms, by contact with water, phosphorous acid being formed, which remains in the tube, and the gaseous iodohydric acid given off. The gas cannot be collected over mercury, which decom- poses it by seizing upon the iodine, and liberating the hydrogen, but it must be collected in a dry tincture-bottle, like chlorine (§ 167). The density of iodohydric acid gas is 4.448. It is colour- less, fumes copiously in the air, is extremely soluble in water, and generates a strongly acid solution, which fumes when concentrated. Iodohydric acid is not a very stable compound; for bromine and chlorine readily decompose it by seizing upon its hydrogen and liberating its iodine. It is also decomposed, when in solution, by the oxygen of the air, at ordinary temperatures. In fact, its solution soon becomes coloured in the air, a portion of the acid being decomposed by the oxygen of the air, water being formed, Fig. 239. 248 IODINE. and the liberated iodine dissolving in the unchanged iodohydric acid; for a solution of the acid dissolves a large quantity of iodine. As the decomposition of the acid progresses, the colour of the liquid becomes more brown, and there soon remains too little un- altered acid to hold all the iodine in solution, so that it begins to be slowly deposited in very regular and often large crystals. Iodohydric acid cannot be analyzed in the same manner as chlo- rohydric and bromohydric acids, that is, by decomposition with potassium in a bent tube over mercury, as this metal itself decom- poses it. But it can be readily shown that the acid is composed of 1 volume of hydrogen and 1 volume of vapour of iodine united without condensation. In fact, if we add, to the density of hydrogen 0.0692 the density of iodine vapour 8.7160 we find 8.7852 nearly the double of 4.443, which has been found by experiment to be the density of iodohydric acid gas. Its composition in weight is, therefore, Hydrogen 00.78 Iodine 99.22 100.00 Or, Hydrogen 1 Iodine 127 128 By taking 127 as the equivalent of iodine, that of iodohydric acid gas becomes 128; and the equivalent of the gaseous acid is 4 volumes, like that of chlorohydric and bromohydric acids. COMBINATION OF IODINE WITH NITROGEN. Iodide of Nitrogen, NI3. § 200. Iodide of nitrogen is a fulminating compound, like the chloride, but is solid at ordinary temperatures. It is obtained by pouring concentrated ammonia upon small quantities of powdered iodine in watch-glasses. In a quarter of an hour, the compound being formed, is poured upon small filters, and appears as a gray- ish-black powder, which is rapidly washed. It does not generally detonate while moist, although at times explosion takes place, even in the watch-glasses, when it is touched with a glass rod; but, as soon as it is dry, it detonates on the slightest friction, even that of a feather, and often explodes spontaneously. Its formula, analogous to that of the chloride, is NI3. SULPHURETS AND CHLORIDES OF IODINE. 249 COMBINATIONS OF IODINE WITH SULPHUR. Sulphurets of Iodine. §201. No definite sulphurets have yet been obtained; for when heated together, they combine, but if the temperature be raised, the combination is destroyed and the iodine volatilized. COMBINATIONS OF IODINE WITH CHLORINE. Chlorides op Iodine. § 202. If a current of chlorine he passed over iodine in a glass tube, the two substances combine, forming at first a brown liquid, but, by continuing the action of chlorine, it forms a yellowish- white crystalline body. These combinations have been hitherto but little studied. 250 FLUORINE. Equivalent F = 19 (237.5 0 = 100). § 203. The properties of isolated fluorine are, as yet, unknown, owing less to the difficulty of separating this substance from its combinations, than to its great affinity for the materials of which our chemical apparatus is generally made ; for it instantly attacks glass and all the metals, even platinum. It has only been obtained in vessels cut out of fluor-spar, by decomposing fluoride of silver by chlorine, the fluorine being evolved in the form of a colourless* gas. The compounds of fluorine with oxygen are unknown, but we can readily prepare its compound with hydrogen, fluohydric acid, an acid of great practical importance. COMBINATION OF FLUORINE WITH HYDROGEN. Fluohydric Acid, HF. § 204. This acid is obtained by acting with sulphuric acid on fluoride of calcium, or fluor-spar, a common mineral. As fluohy- dric acid attacks glass, porcelain, and the majority of metals, it is prepared in vessels of lead or platinum. The apparatus generally used in the laboratory consists of a leaden retort (fig. 240), made of two pieces vdiich fit into each other, the lower piece, shaped like a cup, con- taining the mixture, and the upper forming the head and neck of the re- tort, which conveys the acid vapours into a receiver. The latter is a bent leaden tube, fitted on the neck of the retort, and with a small hole at its end, to give vent to the expanded air, or the excess of vapour : the receiver is surrounded with ice during the operation. The jiuor-spar, finely powdered, is placed in the cup, and twice its weight of concentrated sulphuric acid poured upon it, and stirred with a platinum or leaden spatula. The apparatus is then fitted together, and the joints covered with an earthy luting, kept in its place by a paper band. The retort is then heated, taking care not to elevate the temperature to the point of fusion of the Fig. 240. * Some say a yellowish gas.—J. C. B. FLUOHYDRIC ACID. 251 lead. When the operation is terminated, the fluohydric acid which has condensed in the receiver is poured into a silver or leaden ves- sel, closed with a well-ground stopper of the same. The fluohydric acid thus obtained is anhydrous, and in order to procure it diluted with water, a certain quantity of water is put into the receiver, and greatly facilitates the condensation of the acid vapours. The theory of the process is simple, being the same as that for preparing chlorohydric acid (§ 184): CaF+S03,H0=Ca0,S03+HF. Fluohydric acid is very dangerous to handle, a drop of anhy- drous acid on the skin producing very acute inflammation—often accompanied with fever. A burn over a large surface might prove fatal. When diluted with water, it is much less corrosive, but, even then, must be handled with caution. Anhydrous fluohydric acid is a colourless liquid, of the density 1.06, does not congeal at any temperature, and boils at about 86°. It gives off thick, white fumes in the air, from its combination with aqueous vapour, showing a great affinity for water, with which it combines in every proportion; but when sufficiently di- luted, it ceases to fume in the air. When the anhydrous acid is poured into water, each drop produces a hissing sound like that of red-hot iron. Fluohydric acid attacks glass, by a chemical action which will subsequently be explained. It is hence used to engrave on glass, and mark the divisions on thermometer-scales and graduated tubes (§ 83). Engraving can also be executed by gaseous fluohydric acid, whereby still finer divisions are obtained, and more visible, because opaque; while those made by the liquid acid are trans- parent, and must be deep to be readily seen. To engrave with gas, the body to be marked is exposed to its fumes, arising from a mix- ture of fluor-spar and concentrated sulphuric acid in a leaden box. Fluohydric being very analogous to chlorohydric, bromohy- dric, and iodohydric acids, its composition is probably similar; that is, it is composed of a J volume of fluorine and a \ volume of hydrogen, without condensation. But the composition has not yet been verified by direct experiment, because fluorine has not been isolated so as to determine the proportion, nor has the density of the gaseous acid been determined. §205. The composition by weight of fluohydric acid and the equivalent of fluorine may be ascertained as follows. A certain weight of fluor-spar, reduced to an impalpable powder, is treated with concentrated sulphuric acid, in a platinum crucible, until it is completely converted into a sulphate, to effect which it must be moistened several times with sulphuric acid, and the ex- cess of acid driven off by heat. The sulphate of lime is at last heated to redness. 252 FLUORINE, It is thus shown that 10 grammes of fluor-spar or fluoride of calcium, CaF, give 17.436®“ of sulphate of lime, CaO,S03. Now, the composition of sulphate of lime, or its proportion of lime and sulphuric acid, is easily determined, synthetically, by moistening 10 grammes of pure quicklime, CaO, with sulphuric acid, in a platinum crucible, evaporating off the excess of acid and calcining the sulphate of lime produced. It is thus found that 10 grammes of lime give 24.286®“ of sul- phate of lime; and hence we infer that the sulphate of lime con- tains, Lime 41.18 Sulphuric acid 58.82 100.00 Now 58.82 of sulphuric acid contains 35.292 of oxygen, and we have seen (§ 135) that, in neutral sulphates, the quantity of oxygen of the base is of that in the acid; hence 41.18 of lime contains, Oxygen 11.766 Calcium 29,414 41.180 Consequently, 100 of lime contain, Oxygen 28.57 or 1 eq. oxygen 8 Calcium 71.43 or 1 “ calcium 20 100.00 or 1 “ lime 28 We may hence calculate, by a simple proportion, that the quan- tity of calcium in our 17.436®“ of sulphate of lime is 5.129. In the 10 grammes of fluoride of calcium, there are, therefore, 5.129®“ of calcium, but as we regard it as formed exclusively of calcium and fluorine, the 10 grammes contain 4.871®“ of fluorine, and the composition of fluoride of calcium is Fluorine 48.72 Calcium 51.28 100.00 If we admit that the formula of fluoride of calcium is CaF, that is, composed of 1 equivalent of fluorine and 1 of calcium, we can determine the equivalent of fluorine from the following propor- tion : 51.28 ; 48.72 : : 20 ; x ; whence x =19. Moreover, the reaction which produces fluohydric acid, and which is represented by the equation CaF+S 08+HO=CaO,SOa+HF, FLUOHYDRIC ACID. 253 shows that fluohydric acid is composed of 1 equivalent of fluorine and 1 of hydrogen, and that it therefore contains 1 eq. fluorine 19 95.00 1 “ hydrogen 1 5.00 1 “ fluohydric acid 20 100.00 This example shows how the composition of bodies which cannot he directly analyzed can he ascertained. But, it is important to ob- serve that our reasoning is based upon this hypothesis, that fluoride of calcium contains only calcium and an element, fluorine, which has not yet been certainly isolated, and, consequently, the fore- going formulae are inaccurate, if our hypothesis is unfounded. 254 PHOSPHORUS. Equivalent P=32 (400 0 = 100). § 206. Phosphorus* may be procured in three states ; solid, liquid, and gaseous. At the ordinary temperature of summer, it is as soft and yielding as wax; but at the temperature of melting ice, it is hard and friable. Crystallized phosphorus cannot he obtained by fusion, because it passes gradually from the liquid to the solid state, a circumstance always opposed to crystallization; but it can be made to crystallize from its solution. If 2 parts of phosphorus and 1 of sulphur be melted together under water, a compound is ob- tained containing an excess of phosphorus in solution, from which a portion of it is deposited on cooling, and frequently assumes the form of regular rhombic dodecahedra. (See fig. 22.) Sulphuret of carbon may also be used as a solvent of phosphorus, and when the solution is slowly evaporated, in a current of carbonic acid gas, at the ordinary temperature, it affords beautiful crystals. The density of phosphorus is about 1.77. It is nearly colourless and translucent when perfectly pure, but it generally has a slightly yellowish tint. It changes colour and becomes red, even in vacuo, when exposed to solar light, which proves that the change is due to molecular modifications, and not to chemical action. It melts at above 111.5°, and boils at 554°: its vapour is colourless, and has a density of 4.326. Phosphorus has a powerful affinity for oxygen, and when heated in the air to about 140°, inflames, an effect which may often be produced by simple friction. Exposed to the air, it undergoes a slow combustion, even at ordinary temperatures, so that a stick of phosphorus, in the air, is always surrounded by a light vapour, which is constantly renewed, and is luminous in the dark. From this property it has received its name (from light, and $°P°q hearing). It diminishes considerably by exposure to the air, and at last, if continued sufficiently long, disappears entirely. It is easy to prove that this phenomenon is accompanied by a true com- bustion of the phosphorus; for if the experiment be made in a bell- glass containing a certain volume of air and placed over the pneu- matic cistern, the volume of the gas will be observed to diminish in consequence of the absorption of the oxygen of the air. After * Phosphorus was discovered in 1669, by Brandt, an alchemist in Hamburg, who obtained it by calcining the residue after the evaporation of urine. Brandt kept his process a secret. Kunckel discovered it some years subsequently. But it was only in 1769 that Gahn and Scheele discovered that phosphorus existed in large quantities in the bones of animals, and made known the process for ex- tracting it from them. PHOSPHORUS. 255 some time the light ceases, and with it the diminution of volume; but the phenomenon is repeated, if an additional quantity of pure air be introduced. Air which has been for some time in contact with phosphorus has been deprived of all its oxygen, and can no longer support combustion. If pure oxygen he substituted for air in the bell-glass, the phosphorus is observed to shine only when the temperature is above 68°, while the light would be apparent in atmospheric air at a much lower temperature. We might hence infer that phosphorus burns more readily in atmospheric air than in pure oxygen; and yet we know that its combustion is much more active in oxygen. It has been ascertained that it only com- bines directly with oxygen, at a low temperature, when this gas is highly expanded, as when it has only the density it possesses in atmospheric air, where | of oxygen is mixed with f of nitrogen. If a fragment of it be placed in a balloon filled with oxygen, com- municating with an air-pump, at a low temperature, it will be ob- served that the phosphorus is not luminous when the elasticity of the gas is equal to that of the atmosphere; but, upon rarefying the gas by the pump, the phenomenon of light immediately appears. If marks be made on a wall with a stick of phosphorus, in the dark, they continue luminous for some time, and cease to be so only when the phosphorus which adhered to the wall has disap- peared by evaporation and combustion. Phosphorus, inflamed in oxygen or in the air, produces a white, pulverulent, very deliquescent substance, the phosphoric acid. But when it undergoes slow combustion in the air, at ordinary tem- peratures, it does not form phosphoric acid, but an inferior degree of oxidation, the phosphorous acid. We thus find the same sub- stance produce, by its direct combination with oxygen, two differ- ent compounds, according to the temperature at which the com- bination takes place. Phosphorus is a very dangerous substance to handle, as it so rea- dily inflames; and a burn from it is painful and difficult to heal. It is kept in the laboratory in bottles filled with water. When we wish to use a piece of phosphorus, one of the sticks is taken from the bottle, and a fragment cut off with scissors, while still wet; it is dried with filtering-paper, and handled as little as possible. It is much more combustible when impure than when perfectly pure. We frequently find use in the laboratory for the phospho- rus remaining from divers processes, and in which it is mixed with a small quantity of red oxide of phosphorus. These fragments are more combustible than pure phosphorus, and require to be still more carefully handled, as they often take fire, when dry, in the higher temperatures of the air. It changes even under water, in corked bottles, when exposed to light, losing its superficial transparency. In this case, it seems to experience only a change in its molecular condition. The change 256 PHOSPHORUS. being more slow when protected from light, it should be kept in an opaque vessel, or the bottle containing it should be in a tin or pasteboard case. By rapid cooling, phosphorus undergoes a modification analo- gous to that of sulphur under the same circumstances; but it is more difficult to effect it. If melted phosphorus, heated nearly to its boiling point, be poured into very cold water, a dark-brown mass is obtained, the consistence of which is very different from that of ordinary phosphorus. The experiment only succeeds with very pure phosphorus, which has been several times distilled. The presence of a small quantity of foreign matter sensibly changes the physical properties of phosphorus: thus, a thousandth part of sulphur renders it brittle, even at a temperature above 68°. As phosphorus boils at a low temperature, it may be readily dis- tilled in glass vessels, but the operation demands great caution, on account of its inflammability. In order to distil a small quantity, it is put into a glass retort (fig. 241), the neck of which fits a moderately large tube, abc, bent in the form of the letter U, at the bottom of which a layer of water intercepts the com- munication with the external air and preserves the distilled phosphorus. The retort being heated, the dilated air depresses the water and causes it to rise in the second leg of the tube U, until it can traverse the fluid column in the shape of bubbles. The phosphorus soon distils over, condenses, and falls to the bottom of the bent tube, where it remains fluid, if the temperature of the water be above 104°. If the distillation stops, or even slackens, absorption may take place ; but it is not dangerous if the apparatus be properly arranged. The vapour of phosphorus condensing in the retort causes a vacuum, so that the water rises in the leg a by the pressure of the atmosphere, and if this leg be not sufficiently large, the water may be driven into the retort, which would burst, and the operator run the risk of a severe burn by phosphorus. But if the leg a be large enough to contain all the water, the air enters the retort in the form of bubbles, and no explosion need be feared. The tube ab serves, at the same time, as a receiver and a safety-tube. We have stated that phosphorus became red when exposed to solar light. It is then converted into a very remarkable isomeric modification, in which it presents properties entirely different from those of ordinary phosphorus. The red modification is obtained in large quantities by keeping phosphorus for several hours at a temperature between 446° and 482°, in a gas on which it exerts no chemical action. The experiment may be made in a retort pre- Fig. 241. PHOSPHORUS. 257 viously filled with hydrogen or carbonic acid gas. A considerable portion of it distils and condenses as ordinary phosphorus, while another portion is converted into the red modification, the quantity of which increases as the operation is continued. The retort is allowed to cool, and the substance treated several times with sul- phuret of carbon, which dissolves the ordinary and leaves the modi- fied phosphorus, in the form of an amorphous powder of a more or less deep red colour. Red phosphorus differs from the ordinary modification, not less in its chemical than in its physical properties; for while ordinary phosphorus melts at 111°, the red may be heated to 482°, without becoming liquid; but at 500° it passes into the ordinary modi- fication. Red phosphorus has no sensible odour at ordinary tempera- tures, but remains unchanged in the air, and is not luminous unless heated to 392°. It does not combine with sulphur, even at the point of fusion of the latter, while ordinary phosphorus, slightly heated with sulphur, combines with it explosively. These two modifications afford the most remarkable example of isomerism, presenting greater differences in their physical properties and behaviour to reagents than many different simple bodies. The chemical identity of the particles composing the two modifications is only demonstrated by the absolute identity of the compounds which they form. § 207. Phosphorus plays an important part in the animal eco- nomy, forming a constituent of bones. When bones are burned in the air, their organic matter is completely destroyed, and given off in the form of gaseous products, and the ashes which remain are only a mixture of carbonate and basic phosphate of lime. From these bone-ashes the phosphorus of commerce is extracted. To 3 parts by weight of ashes are added 2 pts. of sulphuric acid, and 15 or 20 pts. of water; the mixture is stirred, and allowed to stand for 24 hours. The sulphuric acid decomposes the car- bonate of lime, forming with the lime sulphate of lime, and driv- ing off the carbonic acid. Another portion of the acid acts on the basic phosphate of lime, without entirely decomposing it; for it merely removes a portion of the lime, by forming an additional quantity of sulphate of lime, and leaves the phosphate in the state of an acid phosphate of lime. The latter salt is very soluble in water, while sulphate of lime is but sparingly soluble. The two salts are separated by pouring the whole into a bag of close mus- lin, which retains the sulphate of lime, and allows the solution of acid phosphate to pass through. The solution being evapo- rated in a copper vessel to the consistency of syrup, powdered charcoal is gradually added, and the mass completely dried. The mixture, dried at a dull red-heat, is put into an earthenware retort (fig. 242), coated externally with an argillaceous luting, with 258 PHOSPHORUS. Fig. 242. its neck fitted into the tube of a copper receiver B, half filled with water, and supplied with a discharging tube t. A range of several of these retorts is placed in a reverberatory furnace, communi- cating with one or two fires, the flame of which passes through the furnace by the horizontal flue u, and escapes from the chimney T. The receivers B are placed in the same trough, filled with wa- ter, which is kept at a temperature of about 104°, in order that the phosphorus may not become solid and obstruct the tube. A gentle heat being applied at the commencement of the operation, inflammable gases, consisting of hydrogen and oxide of carbon, are disengaged, arising from the water of the acid phosphate of lime, with which it is chemically combined, and retains with force until subjected to a high temperature. As soon as the water becomes free, it is decomposed by the incandescent carbon, producing hydrogen and oxide of carbon, HO+C=CO+H. The acid phosphate of lime is decomposed into a basic phos- phate, which is not altered, and phosphoric acid, which, by con- tact with ignited carbon, gives off phosphorus and oxide of carbon: Phosphoric acid., Carbon. Phosphorus. Oxygen Oxide of carbon. POs+5C=P+5CO. The phosphorus distilling over condenses in a liquid state in the tube and receiver, "while basic phosphate of lime remains in the re- tort, mixed with the excess of charcoal. The phosphorus is filtered by being pressed through a chamois-skin, under hot water, and thus cleansed of its impurities. Lastly, to give it the usual form of sticks, a slightly conical glass tube is plunged into the phosphorus melted under water, sucked at the other end, and when the column of liquid phosphorus ascends the tube, the opening is suddenly closed with the finger, PHOSPHORUS. 259 and the tube plunged into a bucket of cold water to solidify the the phosphorus. It may then be pushed out by thrusting a rod into the narrower end of the moulding tube. § 208. The ready combustibility of phosphorus has led to its application to friction-matches and apparatus for producing in- stantaneous light, and hence its manufacture has greatly increased within the last .few years. Phosphoric lights consist of small leaden vials, at the bottom of which is a small stick of phosphorus. They must be kept tightly closed, and, in order to use them, an ordinary sulphur match is plunged in, to which some particles of phosphorus adhere. The match does not inflame at once, but must be rubbed on a piece of cork or wood. Such apparatus is dangerous; and, moreover, soon becomes useless, when not kept well corked, for the phosphorus, absorbing oxygen from the air, is converted into phosphorous and phosphoric acids, which attract moisture and destroy the efficiency of the apparatus. Phosphoric matches, also called chemical matches, are ordinary sulphur matches, on the end of which is a small quantity of a har- dened combustible paste, which inflames by friction on a hard body. The combustible principle of such pastes is always phos- phorus, but other substances yielding oxygen are added, to faci- litate the combustion, such as nitrate and chlorate of potassa, and certain metallic oxides, as binoxide of manganese and sesquioxide of lead, or red lead, which readily part with a portion of their oxygen. Chlorate of potassa renders the paste detonating by friction, so that a portion of the burning substance may sometimes be projected to some distance. That made with nitrate of potassa burns tranquilly, but a small quantity of the chlorate seems neces- sary to render them sufficiently inflammable. To prepare the paste, phosphorus is melted in a due proportion of water, at 122°, a given quantity of chlorate and nitrate of po- tassa added, which dissolve in the water; then the metallic oxides, if any be used, and, lastly, mucilage of gum. The whole is stirred until a homogeneous paste is obtained, in which no globule of phosphorus can be seen. The paste is usually coloured with Prus- sian blue, or red lead. The ends of sulphur matches are dipped into the paste, and allowed to dry. By rubbing them on a rough hard body, the phosphuretted substance inflames, communicates the same to the sulphur and thence to the wood. To render the friction more effectual, a small quantity of pounded glass is sometimes added to to the paste.* * There are two classes of phosphoric matches in use ; those containing little or no admixture of a body yielding oxygen, which inflame quietly, and those con- taining such body, and inflaming more or less vigorously, in proportion to its quan- tity. Those containing Prussian blue as colouring matter are also mixed with 260 PHOSPHORUS. COMBINATIONS OF PHOSPHORUS WITH OXYGEN. § 209. Phosphorus affords four compounds with oxygen, three of which are acids, viz.: 1. Phosphoric acid P05 2. Phosphorous acid P03 3. Hypophosphorous acid PO The fourth is a neutral compound, an oxide of phosphorus, con- taining less oxygen than the acids. Phosphoric Acid, POs. § 210. Phosphorus, when burned in oxygen or in the air, gives off dense Avhite fumes, which is deposited in the form of a white powder, and rapidly attracts moisture from the air: it is phospho- ric acid. In order to obtain any considerable quantity of it, a large dry bell-glass is placed upon an equally well-dried plate (fig. 243), on which a saucer is put con- taining some pieces of quicklime, and allowed to remain for several hours, in order to dry the enclosed air. The saucer being removed, is replaced by a smaller one, contain- ing a piece of previously ignited phosphorus. Combustion goes on under the bell-glass, as long as it contains sufficient oxygen; phos- phoric acid is deposited, in the form of a white powder, on the sides of the glass and on the plate, and after the complete combus- tion of the phosphorus, there remains in the saucer a reddish sub- stan'ce, the oxide of phosphorus. The pulverulent phosphoric acid is rapidly collected by means of a platinum spatula, and sealed up in a dry bottle. The process may be rendered continuous by means of the appa- ratus represented in fig. 244, in which the phosphorus is burned in a large three-necked balloon, previously dried. The cork which closes the upper tubulure is traversed by a large tube ah of 12 or Fig. 248, clay, chalk, and the like neutral absorbents, whereby the deliquescent acids of phosphorus are either prevented from forming or absorbed. The best matches of rapid ignition contain a mixture of nitrate and binoxide of lead, made by treat- ing red lead with nitric acid, and evaporating to dryness. The pastes with the above ingredients are put on the end of sulphur matches, but some are now made for domestic use without being previously dipped into sulphur. They are made of a very resinous wood, or of soft pine imbued with a little terpentine, and the paste put on their end usually contains, beside nitrate and binoxide of lead, a little chlorate of potassa and sulphur.—J. C. B. PHOSPHORIC ACID. 261 Fig. 244. 14mm diameter, open at both ends, and descending to about the middle of the balloon, where a small porcelain saucer v is attached, by a platinum wire. To the second neck d a tube C is fitted, filled with some desiccating substance, such as pumice-stone im- bued with oil of vitriol. Lastly, to the third neck g a large bent tube gh is fitted, the other end of which dips into a dry bottle B. This latter is connected by the tube Id with a suction apparatus, which may be either a suction bellows or an aspirator, or, lastly, a simple metal tube, of some length, placed either obliquely or vertically, and heated so as to produce a strong draught. A con- tinuous current of air is thus established, which is dried in the tube C, passes through the apparatus, and reaches the aspirator. A piece of phosphorus is dropped through the tube kindled by a hot wire, and the upper end a then closed with a cork. The phosphorus burns into phosphoric acid, a portion of which is deposited in the balloon A, and the remainder in the bottle B. When the first piece of phosphorus has nearly disappeared, a second may be dropped in, and so on, as long as desirable. It need hardly be said, that the phosphorus should be carefully dried by filtering- paper before being dropped into the saucer. The phosphoric acid thus obtained is anhydrous, has a great affinity for water, rapidly attracting moisture from the air, and deliquescing; when thrown into water, it produces a sound like that of red-hot iron plunged into this liquid, showing that there is a great deal of heat disengaged in the combination of the anhy- drous acid with waiter. When the aqueous solution of the acid is evaporated, it yields a syrupy liquid, which deposits crystals of hydrated phosphoric acid, when sufficiently concentrated, but if the solution be still further 262 PHOSPHORUS. heated in a platinum capsule, it loses the last portions of water which can be expelled by heat, and fuses, at a red-heat, into a transparent viscid fluid, which solidifies in the form of a vitreous mass. The fused acid gives off sensible vapours, at a red-heat, but even then is still very far from its boiling point at the ordinary pressure of the atmosphere. Vitreous phosphoric is not anhydrous phosphoric acid, for it still retains 11.2 per cent., or an equivalent of water, which heat alone cannot expel; so that phosphoric acid, once combined with wTater, can never be restored to the anhydrous state by heat alone. § 211. The hydrated acid may be obtained directly by dissolving phosphorus in nitric acid. One part of phosphorus and 13 pts. of nitric acid, diluted to the density of 1.20, are heated in a glass retort (fig. 245), the neck of which connects with a cooled receiver. Reddish fumes are copiously given off, and the phosphorus rapidly disap- pears. When the nitric acid is more concentrated, the ac- tion may become so violent that the vapours and gases, unable to escape by the neck of the retort, may produce a dangerous explosion. If the acid be too dilute, the action is too feeble, and a portion of it distils over without acting on the phos- phorus. When the greater part of the liquid has passed into the receiver, the process is arrested, the distillate poured back into the retort, and redistilled. This operation is called cohobation. When the phosphorus is completely dissolved, the distillation is continued until the liquid in the retort has assumed a syrupy con- sistence, when it is poured into a platinum capsule, and the con- centration completed, for, in order to drive off the last portions of water and nitric acid, a degree of heat is required at which the phosphoric acid would attack the glass of the retort, and, conse- quently, become impure. Since fused phosphoric acid contains 11.2 per cent, of water, the quantity of oxygen in this water is to that in the anhydrous acid as 1:5; so that the formula of the hydrate is P05-f-II0. If the vitreous acid be left under a bell-glass, with twice as much water as it already contains, it is converted into a crystalline mass, which is also a definite hydrate, having the formula of POs-f-3HO. The same crystals frequently form in a solution of phosphoric acid sufficiently concentrated. Lastly, if the vitreous acid come in contact with as much more water only as it already contains, we obtain crystals different from the preceding, and represented by POs+2IIO. Fig. 245. PIIOSPIIORIC ACID. 263 We are thus acquainted with three well-defined hydrates of phosphoric acid: 1. Monohydrated phosphoric acid POs+HO 2. Bihydrated “ “ P05+2HO 3. Trihydrated “ “ POs+3HO Each of these acids generates a series of peculiar salts, present- ing distinct properties, which will be noticed more in detail, when treating of the phosphates: 1. Monobasic phosphates POs+RO 2. Bibasic “ P05+2R0 3. Tribasic “ POs+3RO.* Phosphoric acid is sometimes obtained by calcining phosphate of ammonia, which is procured by decomposing by ammonia the acid phosphate of lime, obtained by treating bone-ashes with sul- phuric acid, as in the preparation of phosphorus. The process is economical; but the acid obtained always contains some am- monia. Phosphoric is a very powerful acid, less energetic, however, at common temperatures, than sulphuric; but, as it is much more fixed, it always expels the latter from its combinations, when the temperature is sufficiently elevated. § 212. The composition of phosphoric acid is thus determined: 10 grammes of it are converted into phosphoric acid, by nitric acid in a glass matrass, and the excess of nitric acid, with the greater part of the water, driven off by boiling. 100 grammes of pure oxide of lead being then weighed in a large platinum crucible, the acid contained in the matrass is poured on it, and the matrass several times carefully washed with distilled water, which is added to the liquid in the crucible. After evaporation to dryness, it consists of the oxide of lead and the phosphoric and nitric acids which combined with this oxide; but if the crucible be heated to redness,f the nitric acid is expelled, and the 100 grammes of oxide of lead has increased in weight by the phosphoric acid produced from 10 grammes of phosphorus. We hence conclude, that 10 grammes of phosphorus produce 22.50sm of phosphoric acid; which gives the following composition of phosphoric acid : Phosphorus 44.44 Oxygen 55.56 100.00 * RO or MO is a general expression for protoxide bases, R signifying radical, and M, metal.—J. C. B. -j- This would unquestionably endanger the crucible.—J. C. B. 264 PHOSPHORUS. The quantity of water contained in hydrated phosphoric acid is ascertained by the process described for sulphuric acid (§ 136). Phosphorous Acid, P03. § 213. We have seen that, when phosphorus is burned freely in oxygen or atmospheric air, it is converted into phosphoric acid. But its combustion may be regulated so as to produce an inferior degree of oxidation, (phosphorous acid,) by allowing a slow current of air to pass over the phosphorus gently warmed. In order to perform this experiment, a piece of phosphorus is put into a tube ab (fig. 246), drawn out to a very fine opening at one end, the other end being connected with an aspirator filled with water. The phosphorus being warmed, and the water of the aspirator made to flow very slowly, almost drop by drop, air enters by the opening at a, and its oxygen burns the phosphorus only into phosphorous acid which condenses, in the form of a pulverulent sublimate, in the anterior portion of the tube ab. The sublimate may be volatilized, from one spot to another, in the atmosphere of nitrogen which fills the tube, but it takes fire when heated in contact with the air, and is converted into phosphoric acid. Phosphorus, exposed to the air at common temperatures, is always surrounded by a white vapour, which is luminous in the dark, and condenses by contact with water into an acid liquid, which is chiefly composed of phosphorous acid, and in order to obtain any quantity of the acid by this process, a stick of phos- phorus is put into each one of a number of glass tubes, as ab (fig. 247), terminated at one end by an opening of 1 or 2 millimetres in diameter, and entirely open at the other. Some twenty of them thus charged are placed in a funnel (fig. 248,) in a bottle containing water. The bottle is placed on a plate, and the whole covered by a bell-glass open at top. Phosphorus burning slowly in the air, at common temperatures, and the phosphorous acid which results from the combustion, being heavier than the air, descends into the bottle, and dissolves in the water, so that a pretty con- centrated solution of this acid may be obtained in a few days. If the sticks of phosphorus were placed unprotected in the funnel, the heat evolved by their slow combustion would raise the temperature sufficiently to cause their rapid combustion, when Fig. 246. Fig. 247. Fig. 248. PHOSPHOROUS ACID. 265 phosphoric acid would be the principal product. The glass tubes surrounding the sticks prevent this effect, and the combustion is still less active because the air has not free access to the surface of the combustible. Nevertheless, the solution thus obtained always contains a por- tion of phosphoric acid, from the fact that phosphorous acid, by contact with the air, rapidly absorbs oxygen and changes into phosphoric acid. It will therefore be observed that, in the experi- ment just described, it is difficult to prevent the transformation of a portion of the phosphorous into phosphoric acid. Phosphorous acid may also be obtained very pure by decompos- ing protochloride of phosphorus PC13 by water; 3 equivalents of chlorohydric and 1 of phosphorous acid being formed. The reac- tion may be thus represented: PC13+3H0=3HC1+P03. The phosphorous and chlorohydric acids remain in the liquid, but, by evaporating it to the consistence of a syrup, the former acid is disengaged; and if the concentrated liquid be placed under the receiver of the air-pump, it often becomes a mass of crystals, which are hydrated phosphorous acid, with the formula PO3+BHO. If the evaporation of hydrated phosphorous acid, by heat, be pushed still farther, the acid will soon begin to decompose, evolv- ing a mixture of hydrogen and phosphuretted hydrogen, which takes fire in the air, and phosphoric acid remains in the liquid. The water and phosphorous acid are simultaneously decomposed; a portion of the hydrogen arising from the decomposition of water being disengaged, while another portion combines with phosphorus of the decomposed phosphorous acid, and the oxygen of this acid, as well as that arising from the decomposition of water, unites with the phosphorous acid remaining, and changes it into phosphoric. § 214. This acid is further ob- tained, by causing chlorine to react upon phosphorus in contact with water. A quantity of phosphorus is put under a stratum of water at the bottom of a test-glass (fig. 249), which is kept in a water-bath at 104°-120°, in order that the phosphorus may remain liquid. Chlorine, being conveyed by a tube to the bottom of the glass, combines with the phosphorus; but the chloride of phosphorus is Fig. 249. 266 PHOSPHORUS. immediately decomposed, by contact with the water, into phos- phorous and chlorohydric acids. It is difficult, however, to obtain very pure phosphorous acid by this method, because an excess of chlorine rapidly converts the phosphorous into phosphoric acid in contact with water. § 215. The composition of phosphorous acid is easily deduced from that of the protochloride of phosphorus; for it appears that, when the protochloride is decomposed by contact with water, its 3 equivalents of chlorine are replaced by 3 equivalents of oxygen. If, therefore, we knew the composition of the protochloride, we could readily calculate the composition of phosphorous acid from the numerical values of the equivalents of chlorine and oxygen. Now, the elementary composition of protochloride of phosphorus may be exactly determined by decomposing 10 grammes of it by shaking it with water in a ground-stoppered bottle, and then ascertaining the weight of chloride of silver which the solution thus obtained precipitates in a solution of nitrate of silver in excess. It would be found that this weight is 31.085sm, containing 7.686s111 of chlorine. The 10 grammes of protochloride consequently contain 7.68of chlorine; whence, 100 grammes of protochloride of phosphorus are composed of Chlorine 76.86 Phosphorus 23.14 100.00 or, 1 eq. phosphorus 32 3 “ chlorine 106.5 1 “ protochloride of phosphorus 138.5 Phosphorous acid being formed by means of the protochloride, by replacing the chlorine with an equivalent quantity of oxygen, it must evidently contain, 1 eq. phosphorus 32 57.14 3 “ oxygen 24 42.86 1 “ phosphorous acid 56 100.00 Hypophosphorous Acid, PO. § 216. When phosphorus is boiled with a solution of potassa, soda, baryta, or wTith whiting, the water is decomposed, phos- phuretted hydrogen disengaged, and a hypophosphite of the base is formed, which remains in solution in the liquid. A similar reaction takes place when the phosphuret of lime or baryta is decomposed by water. Free hypophosphorous acid is easily obtained from the hypo- phosphite of baryta, by precipitating the baryta with sulphuric OXIDE OF PHOSPHORUS. 267 acid. The liquid may then be evaporated to the consistence of syrup without decomposition, but it never crystallizes; and when still further heated, the hypophosphorous acid is decomposed; spontaneously inflammable phosphuretted hydrogen gas being evolved, and phosphoric acid remaining. Hypophosphorous acid exhibits a great affinity for oxygen, by reducing a great number of metallic oxides, and converting those of mercury and copper into their metallic state. Aided by a gentle heat, it decomposes concentrated sulphuric acid, disengaging sulphurous acid, and depositing sulphur. Hypophosphorous acid forms definite salts with bases, several of which are susceptible of crystallization, and are easily obtained by decomposing the hypophosphite of baryta by soluble sulphates. The composition of hypophosphorous acid deduced from the analysis of the hypophosphites, has been found to be 1 eq. phosphorus 32 80.00 1 “ oxygen 8 20.00 40~ 100.00 It is important to remark here, that all the hypophosphites hitherto analyzed contain water, which cannot be removed without decomposing them; and as it is possible that the elements of this water enter into the composition of the acid, its formula would be less simple than that we have assigned to it. Oxide of Phosphorus. § 217. When a piece of phosphorus contained in a small capsule is burned in the air or in oxygen, there always remains after the combustion a red residuum, which is an oxide of phosphorus con- taining less oxygen than hypophosphorous acid. But the product is impure from admixture with phosphoric acid. It is obtained in a pure form by putting phosphorus in a test-glass (fig. 250), filled with hot water to keep the phosphorus melted, and passing a current of oxygen to the bottom of it. The phosphorus then burns under water, producing phosphoric acid which dissolves, and oxide of phosphorus which floats on the liquid in the form of red flocculi. The flocculi are col- lected on a filter, rapidly dried on filtering-paper, after being well washed, and then treated with sulphuret of carbon, which dis- solves the free phosphorus mixed with the oxide. Fig. 250. 268 PHOSPHORUS. Oxide of phosphorus rapidly absorbs the oxygen of the air, and is finally converted into phosphoric acid; but if heated without contact of the air, it is decomposed into phosphorous and phospho- ric acids. When phosphorus is mechanically mixed with a small quantity of oxide of phosphorus, it is much more combustible than when pure. Such mixtures are frequently found in laboratories when old residues of phosphorus kept in badly-stoppered bottles are melted; such impure phosphorus, being more combustible, requires more careful handling than transparent phosphorus. In order to ascertain the composition of the oxide of phosphorus, a given weight of it (say 1 gramme) is converted into phosphoric acid, by means of nitric acid, and a known weight p of oxide of lead is added to the liquid, more than sufficient to saturate the phosphoric acid formed. It is evaporated to dryness, and the residuum calcined until reddish vapours cease to be given off. If p' be the weight of the residue, it is evident that {p'—p) is the weight of the phosphoric acid formed. As the composition of phosphoric acid is known, we know the weight q of phosphorus contained in {p'—p) of phosphoric acid, and conclude from the experiment that 1 gramme of oxide of phosphorus contains q of phosphorus, and therefore (1—q) of oxygen. The various analyses thus made of the oxide of phosphorus have given different results, and its true formula is yet uncertain. RECAPITULATION OF THE COMBINATIONS OF PHOSPHORUS WITH OXYGEN. Equivalent of Phosphorus. § 218. The three well-ascertained compounds of phosphorus with oxygen are composed as follows: Ilypophosphorous acid Phosphorus 80.00 Oxygen 20.00 100.00 Phosphorous acid Phosphorus 57.14 Oxygen 42.86 100.00 Phosphoric acid Phosphorus 44.44 Oxygen 55.56 100.00 The composition of these substances, referred to the same quan- tity 100 of phosphorus, is EQUIVALENTS. 269 Hypophosphorous acid Phosphorus 100.00 Oxygen 25.00 125.00 Phosphorous acid Phosphorus 100.00 Oxygen 75.00 175.00 Phosphoric acid Phosphorus 100.00 Oxygen 125.00 225.00 The quantities of oxygen which have combined with the same quantity of phosphorus in these three compounds, are to each other as the numbers 1:3:5. The most simple formulae which can be assigned to them are, therefore, Hypophosphorus acid PO Phosphorous acid P03 Phosphoric acid P05 The equivalent of phosphorus is therefore deduced from one of the following proportions : 20.00:80.00:: 8 : z 42.86 : 57.14 :: 24 : z 55.56 : 44.44 :: 40 : a; ■whence x = 32. The numerical value of the equivalents of the three compounds will therefore be, Hypophosphorous acid PO = 40 Phosphorous acid P03 = 56 Phosphoric acid P05 =72 Let us compare these theoretical equivalents with the equivalents immediately deduced from the analysis of the salts. The analysis of the hypophosphite of lead has shown that the equivalent of hypophosphorous acid was equal to 40; so that the formula PO is the one proper to that acid. The examination of the phosphites has led to a similar conclu- sion for phosphorous acid, and confirmed the formula P03. We shall hereafter find that phosphoric acid forms several series of salts with the same base, and that we must admit of its forming three classes of salts. 1. Salts in which 1 eq. of acid saturates 3 eq. of base. 2. Salts “ 1 eq. of acid “ 2 eq. “ 3. Salts “ 1 eq. of acid “ 1 eq. “ 270 PHOSPHORUS. The numerical value for the equivalent of phosphoric acid, deduced from the analysis of these various salts, always remember- ing their various modes of saturation, is constantly 72. Hence, a detailed examination of the phosphates confirms the formula P05 for phosphoric acid. The density of the vapour of phosphorus has been found to be 4.326, by direct experiment. It is easy to calculate the volume of gaseous phosphorus which represents its equivalent in volume. In fact, starting from the composition of phosphorous acid, which we regard as composed of 1 equivalent of phosphorus and 3 equiva- lents of oxygen, the 3 equivalents of oxygen are represented by 3 volumes, which weigh 3 (1.1056)=3.3168; and from which the proportion is made 24: 32:: 3.3168: x. This gives for the weight of the vapour of phosphorus which has combined with 3 volumes of oxygen, #=4.4224, which differs but little from the density 4.326, determined by direct experi- ment. Phosphorous acid therefore contains 3 volumes of oxygen and 1 volume of vapour of phosphorus, and the equivalent of gaseous phosphorus is represented by 1 volume. In the atomic theory, the compounds of phosphorus and oxygen are written as follows: Ilypophosphorous acid P30 or PO Phosphorous acid P203 or P03 Phosphoric acid P20s or P05 Two atoms of phosphorus therefore correspond to our equiva- lent, so that the atomic weight of phosphorus is 16. This mode of composition has been adopted because it gives formulae to the compounds of phosphorus with oxygen and hydrogen similar to those of the compounds of nitrogen with the same elements. Had we started with the hypothesis (§ 88) that all simple gases contain the same number of atoms for equal volumes, we would never arrive at different atomic formulae. In fact, the compo- sition of phosphoric is different from that of nitric acid, for in the latter, 5 volumes of oxygen are combined with 2 of nitrogen, while, in phosphoric acid, the 5 volumes of oxygen have combined with only 1 volume of vapour of phosphorus. If, therefore, the atomic formula of nitric acid be N205, conformably to the hypo- thesis alluded to, that of phosphoric acid must be P05. The formulae of the two acids, and, consequently, of the other combi- nations of nitrogen and phosphorus, would no longer be similar. COMBINATIONS OF PHOSPHORUS WITH HYDROGEN. § 219. Phosphorus and hydrogen combine in three proportions: 1. A gaseous compound called phosphuretted hydrogen; 2. A liquid PHOSPHURETTED HYDROGEN GAS. 271 compound with excess of hydrogen; 3. A solid compound contain- ing the greatest proportion of phosphorus. Phosphuretted hydrogen gas is obtained by several processes: 1. A small flask (fig. 251) is two-thirds filled with a concentrated solution of caustic potassa, to which a few pieces of phosphorus are added: heat being applied, small bubbles of gas are disengaged, which inflame as soon as they reach the air. A small quantity of gas is allowed to escape before adapting the discharging-tube, in order to expel the air from the flask: an indispensable precaution, for, if the flask were closed immediately, the inflammable gas, coming in contact with the confined air of the flask, might produce an explosion. The gas is evolved under water, and each bubble, as it reaches the air, inflames, producing a curling ring of white vapour, which enlarges as it rises. The circles are very regular when the air is calm. If the bubbles be passed into a bell-glass containing oxygen, the flame is much more brilliant, but the experiment requires great caution, and the bubbles of gas must be very small, or otherwise an explosion would ensue. The following is the theory of this reaction:—Phosphorus alone does not decompose water, but in contact with potassa, the affi- nity of the base for hypophosphorous acid, which is one of the pro- ducts of the reaction, causes the reaction in the same manner as, in the preparation of hydrogen gas, the presence of sulphuric acid causes the decomposition of water by zinc, at ordinary tempera- tures (§ 69). A portion of phosphorus combines with oxygen to form hypophosphorous acid, which, with the potassa, produces the hypophosphite of potassa, wrhile the hydrogen combines with another portion of phosphorus, and is disengaged in the state of phosphuretted hydrogen. The gas thus obtained is often mixed with free hydrogen, as may be ascertained by introducing into the bell-glass a solution of sulphate of copper, which absorbs the phosphuretted hydrogen and sets the hydrogen free. The presence of the latter gas is thus explained: if a solution of hypophosphite of potassa be heated with an excess of potassa, water is decomposed, its oxygen chang- ing the hypophosphite into a phosphate of potassa, and its hydro- gen being disengaged. This reaction may occur simultaneously with the first, in the process just described. Hydrated lime may be substituted for the solution of potassa, by making a paste of slacked lime and water, and rolling it into small balls, each of which contains a small piece of phosphorus. Fig. 251. 272 PHOSPHORUS. A number of such balls being put into a matrass, and heated, the phosphorus melts and produces a reaction similar to that just described. But the best process, and that which affords the purest gas, consists in decomposing the phosphuret of calcium by water. The phosphuret is prepared by heating lime in a current of vapour of phosphorus. A strong glass tube, closed at one end, is filled with balls made of the hydrated lime and calcined, and some pieces of phosphorus are placed at the bottom of it. The tube being heated to redness, some hot coals are brought near the end containing the phosphorus, which is volatilized, and its vapour, passing through the tube, combines with the lime. To procure a large quantity of the phosphuret, a large earthen crucible (fig. 252), having a hole in the bottom, to which is fitted the neck of a small flask con- taining phosphorus, is filled with balls of lime. The crucible is placed over the grate of a fur- nace, so that the flask containing the phosphorus shall be below the grate. The crucible being heated to redness, and some coals brought near the flask to vaporize the phosphorus slowly, its vapour passes into the crucible and combines with the lime. If the phosphuret of calcium be thrown into water (fig. 253), reaction immediately takes place, and spontaneously inflammable phosphu- retted hydrogen is disengaged. § 220. Phosphuretted hydrogen is a colourless gas, of an extremely fetid and characteristic odour: its density is 1.185 : water dissolves but a small quantity of it. If it be kept for some time over mercury, it undergoes a remarkable change, a brownish deposit takes place on the sides of the glass, and the gas has lost the property of spontaneously inflaming by con- tact with the air. The volume of the gas has scarcely changed, and, on analysis, its composition is found to be nearly the same. The gas may be procured immediately, but not spontaneously inflammable, by decomposing the phosphuret of calcium by chloro- hydric acid instead of water. It may also be obtained by heat- ing phosphorous and hypophosphorous acids, which are hydrated, and under the influence of heat, the water and acid are both de- composed at once, a portion of the acid gives off its phosphorus, which combines with the hydrogen to form phosphuretttd hydrogen, while its oxygen combines with another portion of the acid and converts it into phosphoric acid. Fig. 252. Fig. 253. PHOSPHURETTED HYDROGEN GAS. 273 The difference* in the behaviour of phosphuretted hydrogen prepared by one or other of these processes is due to the presence in the spontaneously inflammable gas of a small quantity of another phosphuretted hydrogen, richer in phosphorus, and which takes fire on contact with the air. In order to separate this liquid, the spon- taneously inflammable gas is passed through a tube bent in the form of the letter U and surrounded by a refrigerating mixture. There condense in the tube, at the same time, wrater which solidifies, and a colourless liquid which may be separated by allowing it to run to that part of the tube unoccupied by water, and then closing it by a flame. The gas which escapes from the U-tube has lost the property of self-inflammability in the air. Liquid phosphuretted hydrogen is not very fixed, and can be preserved only in the dark, for it decomposes rapidly by light into phosphuretted hydrogen gas and a solid body of an orange- yellow colour, which is a third phosphuret of hydrogen containing still more phosphorus than the liquid phosphuret. It is the same substance that is deposited on the sides of the bell-glass in which spontaneously inflammable hydrogen gas is kept, and which thus loses its inflammability. Liquid phosphuretted hydrogen is much more easily decomposed by certain acids, as the chlorohydric, etc.; and hence, a non- spontaneously inflammable gas is always obtained when the phos- phuret of calcium is decomposed by chlorohydric acid. Pure phosphuretted hydrogen gas, entirely deprived of liquid phosphuret, is not spontaneously inflammable at ordinary tempera- tures, but a slight elevation of temperature, as 212°, restores this property. Many substances deprive phosphuretted hydrogen gas of its property of being spontaneously inflammable, such as those which readily decompose the liquid phosphuret. Others, chiefly oxidizing substances, like deutoxide of nitrogen, etc., restore its spontaneous inflammability, by decomposing a small quantity of phosphuretted hydrogen gas, depriving it of a portion of its hydro- gen, and thus converting it into the liquid hydruret of phosphorus, which remains in a state of vapour in the undecomposed gas. A very simple experiment proves that it is the presence of the liquid phosphuret in vapour in the phosphuretted hydrogen gas which communicates to this gas the property of spontaneously in- flaming in the air at ordinary temperatures ; for the same property may be communicated to all combustible gases, by adding to them a small quantity of the vapour of liquid phosphuret. Thus, if into a bell-glass filled with hydrogen gas a drop of liquid phosphuret of hydrogen be introduced, a gaseous mixture is obtained, which immediately inflames on contact with the air. The vapour of the * M. Paul Thenard first isolated liquid phosphuretted hydrogen, the vapours of which afford the spontaneously inflammable hydrogen gas, and thus explained the anomalies which had been found in the properties of the gas. 274 PHOSPHORUS. liquid phosphuret takes fire and communicates inflammation to the hydrogen. § 221. Phosphuretted hydrogen gas is analyzed by passing it through a tube A (fig. 254), filled with copper heated to redness; Fig. 254. the gas is decomposed, the copper seizing on the phosphorus, and hydrogen being set free. The latter gas then traverses a second tube B, heated to redness and filled with oxide of copper, in which it burns, forming water, which condenses in a tube C filled with pumice-stone imbued with sulphuric acid. The first tube A, having been weighed before the experiment, is weighed afterward, and its increase in weight gives the quantity of phosphorus. In order that the tube A may remain unaltered during the experiment, it is heated by alcohol lamps, as represented in the figure. The tubes must also be filled with nitrogen before commencing the ex- periment, and again washed with it at the close, by connecting a gasometer filled with nitrogen with the end a of the tube A. It has thus been found that 100 parts of phosphuretted hydrogen gas contain Hydrogen 8.57 Phosphorus 91.43 100.00 This composition corresponds to the following in volumes: 1J vol. of hydrogen 0.1032 vol. of vapour of phosphorus 1.0815 1.1847 nearly agreeing with the density 1.185, as found by experiment. It has been shown that 1 volume of ammoniacal gas also contains 1J vol. of hydrogen, but it contains a | vol. of nitrogen, Avhile phosphuretted hydrogen only contains \ vol. of vapour of phos- phorus. We have asserted that the compounds of nitrogen and phosphorus corresponded exactly;. we have, therefore, between ammonia and phosphuretted hydrogen, an anomaly precisely similar to that already found between sulfhydric acid gas and the vapour of water (§ 152). The anomaly disappears by supposing vapour PHOSPHURET OF NITROGEN. 275 of phosphorus to be formed by a grouping of two chemical molecules. Having adopted the number 32 as the equivalent of phosphorus, let us now calculate the composition of phosphuretted hydrogen gas, with reference to this weight of phosphorus, by making the proportion 91.43 : 8.57 : : 32 : a?, whence x=3, which represents 3 equivalents of hydrogen; and phosphuretted hydrogen gas therefore contains 1 eq. phosphorus 32 3 “ hydrogen 3 1 “ phosphuretted hydrogen 35 The composition of liquid phosphuretted hydrogen has been de- termined from the quantity of solid and gaseous phosphuretted hydrogen which it gives by decomposition, from which its composi- tion in equivalents is represented by PHa. Lastly, the composition of the solid phosphuret is ascertained by finding the volume of hydrogen afforded by a known weight of it, when it is decomposed by metallic copper in a tube heated to red- ness. The formula of the solid phosphuret is PaII. COMBINATION OF PHOSPHORUS WITH NITROGEN. Phosphuret of Nitrogen, NaP. § 222. If dry ammoniacal gas be passed through liquid proto- chloride of phosphorus, it is absorbed in large quantities, and a white crystallized body is obtained, the formula of which is PC13,4NH3. By contact with water, it is converted into phosphite and chlo- rohydrate of ammonia, according to the following reaction : PC13,4NH3-KH0=3(HC1,NH3)+P03(NH3,H0). If the product be heated in a small retort, different gases are disengaged, and a large quantity of sal ammoniac sublimed, and by continuing the heat until the disengagement ceases, phosphuret of nitrogen remains, as a white residuum, at the bottom of the retort. Phosphuret of nitrogen bears a red-heat without decomposition, volatilization, or fusion, is insoluble in water and nearly all acids, but is easily analyzed, by heating a known weight of it, mixed with oxide of copper, in the apparatus described (§108). It is thus found to be composed of 276 PHOSPHORUS. 1 eq. phosphorus 32 53.33 2 “ nitrogen 28 46.67 60 100.00 Its formula is, therefore, NSP. COMBINATIONS OF PHOSPHORUS WITH SULPHUR. § 223. Sulphur and phosphorus combine in several proportions. When pieces of sulphur and phosphorus are brought into contact, and gently heated to fuse them, they combine with evolution of heat, and sometimes with explosion. The experiment is dangerous, and requires great care; but to perform it with safety, the phos- phorus is put under water, in a glass flask, and heated until it fuses, when sulphur is gradually introduced in small pieces. We can thus combine a considerable quantity of sulphur with phos- phorus without disturbing the fluidity of the mixture, but, if allowed to cool, a considerable portion of the sulphur separates by crys- tallization. If, on the other hand, but little sulphur be added, and the phosphorus be in excess, the latter crystallizes during cooling. By combining 1 equivalent of phosphorus with 1 of sulphur, that is, 1 part by weight of the former with 2 pts. of sulphur, a product is obtained which is still fluid at 41°, but solidifies below that point, without regularly crystallizing. Phosphorus forms with sulphur a great number of definite com- pounds which generally correspond to those with oxygen; but as these compounds are often more combustible than isolated phos- phorus, manipulation with them requires great caution. COMBINATIONS OF PHOSPHORUS WITH CHLORINE. § 224. Chlorine and phosphorus combine in two proportions, the formulae of which are PC13 and PC15, corresponding to phosphorous acid P03 and phosphoric acid POs. The apparatus used in their preparation resembles that de- scribed (§ 187) for the preparation of the chlorides of sulphur. Phosphorus is put into a tubulated retort D (fig. 255). The com- bination of phosphorus and chlorine takes place with a great eleva- tion of temperature, and often with flame, so that a piece of phos- phorus inflamed in a capsule continues to burn with a greenish flame when plunged into a bottle filled with chlorine. The high temperature developed during the combination fre- quently breaks the tubulated retort, but the danger may be obvi- ated by putting at the bottom of the retort a layer of sand, on which the phosphorus rests. In order to prevent the formation of perchloride, the retort must be heated nearly to the boiling point PROTOCHLORIDE OF PHOSPHORUS. 277 Fig. 255. of phosphorus, so that the chlorine is constantly in an atmosphere of phosphorus in excess, and the protochloride distils over as fast as it forms. The operation is arrested before all the phosphorus has disappeared. The distilled liquid contains phosphorus in solu- tion, which is separated by redistillation. Protochloride of phosphorus is a colourless, very limpid liquid, of the density 1.45; it boils at 172.4°, and the density of its vapour is 4.742. In contact with water, protochloride produces chlorohydric and phosphorous acids, and we have used this property in its analysis, when it was found (§ 214) to he composed of 1 eq. phosphorus 32.0 23.13 3 “ chlorine 106.5 76.87 138.5 .100.00 1 volume of the protochloride is composed of £ vol. vapour of phosphorus = 1.0845 “ chlorine 3.6600 4.7445 The theoretical density of its vapour is therefore 4.744, which is identical with that given by direct experiment. § 225. Subjected to the action of chlorine, it absorbs a large quantity of it, and is converted into a white crystalline substance, which is the perchloride of phosphorus. This body boils at about 298°, which is also near its point of fusion, so that at the ordinary pressure of the atmosphere, it passes immediately from the solid to the gaseous state. In contact with water, the perchloride is thus changed into chlorohydric and phosphoric acids. PCl5+5HO=POs-f 5HC1. 278 PHOSPHORUS. It may be analyzed in the same way as the protochloride; but its composition may also be deduced from that of phosphoric acid, which we have directly ascertained, it being only necessary to substitute 5 equivalents of chlorine for the 5 equivalents of phos- phoric acid. We thus have 1 eq. phosphorus 32.0 15.29 5 “ chlorine 177.5 84.71 209.5 100.00 The density of its vapour has been found to be 3.66: 1 volume of the vapour is therefore composed of \ vol. of vapour of phosphorus 1.085 1 “ chlorine 2.440 3^525 It may be regarded as formed by the combination of 1 vol. protochloride of phosphorus 4.744 1 “ chlorine 2.440 >.184 of which one-half is equal to 3.59 without condensation.* COMBINATIONS OF PHOSPHORUS WITH IODINE. § 226. Iodine and phosphorus heated together, combine with the evolution of heat, but no definite compounds have been hitherto isolated. The combinations are destroyed by water, producing iodohydric, phosphorous, and phosphoric acids ; and it was such a reaction we made use of to obtain iodohydric acid gas (§ 199.) * Mitscherlich found the specific gravity of the vapour of perchloride of phos- phorus to be 4.85, from which it appears to be composed of 1 vol. P 10 vols. Cl, condensed to 6 vols., i. e. 4.4224 P-f- (2.453x10) = 28.952, and hence one vol. = = 4.825, nearly the same as that found, and hence the formula PC15. But Regnault gives the result of experiment as 3.66, from which he makes it consist of 1 vol. P-|- 4 vols. Cl, or 4.422 P -f- (2.453 x 4) = 14.234, con- densed to 4 vols.; thus = 3.558. Again, regarding it as composed of J vol. protochloride of phosphorus and £ vol. chlorine uncondensed, its formula becomes PC15, and its volume weighs 3.60. Upon the former view, its formula is PC12, which is certainly incorrect. The true volume is i vol. of vapour of phosphorus 0.5528 & “ chlorine 8.0662 4 3.619.—J. C. B. 279 ARSENIC. Equivalent As = 75 (937.5 0 = 100). § 227. Arsenic closely resembles the metals in its physical pro- perties, but its compounds are so analogous to the corresponding compounds of phosphorus, that it is advisable to study them in conjunction. Arsenic is of an iron-gray colour, very brittle, possessing a metallic lustre, and a density of about 5.8. Heated to dull red- ness, it sublimes at once without fusion; so that, at first sight, it would seem capable of assuming only the solid and gaseous states, but the apparent anomaly arises from the fact that its point of fusion very nearly approaches that at which it boils under the pressure of the atmosphere. Volatile substances give off vapour much below their boiling points, a property belonging alike to solid as well as liquid bodies. Arsenic, therefore, gives off vapour copiously at a temperature much below its boiling point, and may wholly sublime without attaining that of fusion. The distance between the point of fusion and that of ebullition of any body may, however, be increased at pleasure. For the 'point of ebullition of a body is the temperature at which the tension of its vapour is equal to the pressure exerted upon it, and hence, by increasing the pressure, the boiling point is raised without sensibly affecting the point of fusion. We can thus obtain melted arsenic, if, instead of heating it in an open tube, it is heated in a thick glass tube hermetically sealed, so that the increased pressure in the tube opposes the ebullition of the arsenic, which may be fused long before it boils. Reciprocally, it is evident that a volatile solid body may be always subjected to so slight a pressure that it will boil at a tem- perature inferior to that at which it melts. Thus, ice at the tem- perature of 30.2° possesses an elastic force represented by 4.27mra (0.168 inches); in other words, it boils at a temperature of 30.20°, under the pressure of 4.27mra. Ice may therefore be entirely volatilized by ebullition under this feeble pressure, without reach- ing its point of fusion, which is 32°. The vapour of arsenic is colourless, and has a very well-marked odour, similar to that of garlic, as may be shown by throwing some powdered arsenic on an ignited coal. The density of its vapour is 10.37. The vapour of arsenic is always deposited in the form of crystals, so that crystallized arsenic can be readily obtained by sublimation. For this purpose, a quantity of arsenic is put into an earthen retort, so as to fill about one-third of it, and the retort placed over a furnace, the coals only touching its lower part. To 280 ARSENIC. prevent the external air from entering too freely into the retort, the beak is partly closed by inserting a pierced cork into it. The sublimed arsenic condenses in the upper part and neck of the retort, and when the operation is terminated, the retort is allowed to cool completely, and, upon being broken, the dome is found filled with very brilliant crystals. They are rhombohedrons, but, as they are generally grouped in masses, it is often difficult to recognise their forms. Arsenic oxidizes in the air? even at common temperatures, its surface becoming tarnished and covered with a blackish powder; but the metallic lustre is easily restored by leaving it for a few hours in a solution of chlorine. It is combustible, burning with a livid flame, and producing arsenious acid. This acid is commonly called arsenic, or white arsenic, and is obtained by roasting metallic arseniurets. The acid is easily decomposed by carbon, which deprives it of its oxygen and restores it to the metallic state. Metallic arsenic is prepared for the arts by decomposing a natural compound of arsenic, sulphur, and iron, known to mineralo- gists as mispickel. This ore is charged into earthenware pipes of about 1 metre (3| feet) in length and 0.3 (1 foot) in diameter, together with some pieces of sheet or cast-iron, in order to retain more effectually the sulphur, and the first pipe is covered by a second shorter and larger one, which serves as a receiver. A certain number of these pipes being placed in the same furnace and heated to redness, the arsenio-sulphuret of iron is converted into sulphuret, and arsenic sublimes in the receiver. It is puri- fied by redistillation with carbon. COMBINATIONS OF ARSENIC WITH OXYGEN. § 228. Two combinations of arsenic with oxygen are known, corresponding to phosphorous and phosphoric acids. Arsenious Acid, As03. § 229. When arsenic is heated in a current of atmospheric air or oxygen, it is converted into a white substance which sublimes: it is arsenious acid. It is found in commerce, and is largely used in painting, principally in the form of arsenite of copper, which furnishes a beautiful green-colour. Arsenious acid is obtained by -roasting metallic arsenio-sul- phurets, such as those of iron, nickel, and cobalt. The principal object of the process usually being the extraction or concentration of the metal combined with the arsenic. The mineral being gene- rally spread on the hearth of a reverberatory furnace, is traversed by a current of hot air which has passed over the grate, and con- verts sulphur into sulphurous, and arsenic into arsenious acid. ARSENIOUS ACID. 281 The sulphurous acid escapes by the chimney, whilst the arsenious acid condenses in the recipients arranged between the furnace and the chimney. In order to obtain pure arsenious acid, the crude acid produced by this process is redistilled in sheet-iron tubes. The freshly prepared acid presents the appearance of perfectly colourless vitreous masses, which, after some time, become opaque and resemble porcelain. The change gradually takes place from the surface to the centre, so that when a piece is broken, which looks externally like porcelain, it is frequently found vitreous inside. The vitreous and porcellainous acids are two isomeric states of the same body, no change of weight having been observed during the transformation; but the acid, in its two modifications, presents remarkably different properties. The vitreous is three times as soluble in water as the opaque acid, and dissolves more rapidly. The opaque is converted into vitreous acid by prolonged ebulli- tion in water, 1 litre (If pints) of which dissolves about 110 grammes (1700 grs.) of the vitreous acid. Under the influence of water and a low temperature, the vitreous is transformed into the opaque acid, so that a solution of the vitreous acid, after a certain time, falls to the point of saturation proper to the opaque acid. Mechanical division transforms the vitreous into the opaque acid; so that, if the vitreous acid be very finely pulverized, it possesses only the solubility of the opaque acid. A solution of arsenious acid reddens the tincture of litmus, but only like a feeble acid. It dissolves more easily and largely in dilute chlorohydric acid than in pure water. Arsenious acid has no sensible odour at ordinary tempera- tures; when put on a heated brick, it volatilizes with a white vapour, exhaling a faint odour; but when thrown on an ignited coal, it gives oft' a very strong odour of garlic. This odour is pro- duced by the vapour of metallic arsenic, to which the carbon has reduced a portion of the acid. The composition of arsenious acid might be ascertained by find- ing the increase in weight of a given weight of arsenic, which is converted into arsenious acid by heating it in a current of oxygen; but it is better to deduce its composition from the analysis of the protochloride of arsenic, as the composition of phosphorous acid was deduced from the analysis of protochloride of phosphorus (§ 2i4). . The chloride of arsenic is decomposed by contact with water into arsenious and chlorohydric acids, which gives for the compo- sition of arsenious acid, 282 ARSENIC. Arsenic 75 75.75 Oxygen 24 24.25 99 100.00 Arsenic Acid, As05. § 230. Arsenic acid is obtained by boiling arsenious acid with aqua regia in excess, and evaporating to dryness to drive off the chlorohydric and nitric acids. The dried residuum dissolves but slowly in water, although arsenic acid is very soluble; but if the solution be evaporated slowly, it deposits large crystals of hydrated arsenic acid, which dissolve readily in water. The solution of the anhydrous acid, i. e. deprived of its water of crystallization, is more slow. When arsenic acid is heated to a dull red, it decomposes into arsenious acid which sublimes, and oxygen which is evolved. Its composition is readily ascertained by finding the weight of arsenic acid afforded by 1 gramme of arsenious acid. For this purpose, the arsenious acid is heated with concentrated nitric acid, evapo- rated nearly to dryness, and 10 grammes of oxide of lead added to it. It is perfectly dried, and the residue calcined. The residue is composed of 10 grammes of oxide of lead, increased by the weight p of arsenic acid produced from 1 gramme of the arsenious. 1 gramme of the latter, therefore, absorbs (p—1) gramme of oxygen when converted into arsenic acid. We thus find that arsenic acid is composed of 1 eq. arsenic 75 65.22 5 “ oxygen 40 34.78 1 u arsenic acid 115 100.00 When finely powdered metallic arsenic is exposed to a damp at- mosphere, it changes into a black substance, considered by some chemists as a peculiar oxide containing less oxygen than arsenious acid. When heated in a closed tube, it is converted into metallic arsenic and arsenious acid. COMBINATIONS OF ARSENIC WITH HYDROGEN. § 231. Two compounds of arsenic and hydrogen are known, the first of which is gaseous, and known as arsenuretted hydrogen gas ; the second is solid. Arsenuretted hydrogen gas is obtained by treating arseniuret of tin with concentrated chlorohydric acid. The arseniuret is pro- cured by melting 3 parts of tin with 1 of arsenic in a crucible. The pulverized arseniuret being put into a small flask, and chloro- hydric acid poured on it through an S-tube, the evolution of gas commences in the cold, and may be accelerated by a few coals. ARSENURETTED HYDROGEN. 283 The resulting changes are chloride of tin, which remains in the flask, and arsenuretted hydrogen gas, which is disengaged. The gas thus obtained is always mixed with free hydrogen, because all the tin does not combine with the arsenic, and the free metal dis- engages hydrogen with chlorohydric acid. The presence of hy- drogen may be easily ascertained, by introducing into the bell- glass a solution of sulphate of copper, wdiich absorbs the arsenuret- ted hydrogen. Arsenuretted hydrogen is a colourless gas, having a peculiar nauseating odour. Its density is 2.69 ; it liquefies at about —22°, under ordinary pressure. In contact with any burning substance, it inflames in the air, burning with a livid flame, and forming water and arsenious acid, but there is always deposited, on the sides of the glass a brown powder, due to incomplete combustion, which is solid arsenuretted hydrogen. Heat decomposes arsenuretted hydrogen, for if passed through a tube heated to redness, hydrogen becomes free, and a brilliant ring of metallic arsenic is deposited beyond the heated part of the tube. This behaviour serves to detect the smallest quantities of arsenuretted hydrogen mixed with hydrogen. Chlorine instantly decomposes arsenuretted hydrogen gas, each bubble of the latter which enters a test-glass filled with chlorine taking fire, and producing chlorohydric acid and chloride of arsenic. Arsenuretted hydrogen is very poisonous, and great care must be taken not to respire the smallest quantity of it. The composition of this gas is ascertained exactly in the same manner as that of phosphuretted hydrogen gas (§ 221); by which it is found that 1 vol. of it contains 1J vol. hydrogen 0.1032 \ “ vapour of arsenic 2.5910 £6942 Its composition, in equivalents, is AsH3. Water dissolves a small quantity of it, but also decomposes it, for a bottle filled with it, and left over water for several weeks, is entirely decomposed, forming a brown deposit of solid arsenuretted hydrogen on its sides. The exact composition of the latter is unknown. COMBINATION OF ARSENIC WITH CHLORINE. § 232. Only one compound of arsenic and chlorine is known, and is obtained by passing chlorine over metallic arsenic, in the apparatus represented in fig. 234, the arsenic being put into the tubulated retort D, which is gently heated, to distil the chloride of arsenic as it forms. The affinity of arsenic for chlorine is very strong, for when the powdered metal is thrown into a bottle filled with chlorine, it inflames, and produces dense white fumes of 284 ARSENIC. chloride of arsenic. It may also be obtained by distilling in a re- tort a mixture of 1 part of metallic arsenic and 6 parts of chloride of mercury. When prepared by the action of chlorine gas on arsenic, it has a yellowish tinge, from dissolved chlorine, which is removed by shaking it with a small quantity of finely-powdered arsenic, and redistilling. Chloride of arsenic is a colourless liquid, wdiich boils at 269.6°; the density of its vapour has been found to be 6.3. In contact with water, it instantly decomposes into arsenious and chlorohy- dric acids. AsC18+3H0=As08+3HC1. It consequently corresponds to arsenious acid, and is composed of 1 eq. arsenic 75.0 41.35 3 “ chlorine 106.5 58.65 1 “ chloride of arsenic 181.5 100.00 1 vol. of its vapour contains % vol. vapour of arsenic 2.591 1J “ chlorine 3.660 6.251 COMBINATIONS OF ARSENIC WITH SULPHUR. § 233. Arsenic and sulphur form many compounds, of which we shall mention only the three more important. A crystallized sulphide is found in nature, the formula of which is AsS3, corresponding to no known compound of arsenic with oxy- gen, and is called by mineralogists realgar. It can be artificially prepared by fusing together suitable proportions of arsenic and sulphur. It is a vitreous body, of a beautiful orange-red colour, and is used in painting. It fuses and sublimes unaltered. The second compound, AsS3, corresponding to arsenious acid, is likewise found crystallized in nature, and is known by the name of orpiment. Orpiment, or sulpharsenious acid, may be prepared by fusing together proper proportions of arsenic and sulphur, or by passing a current of sulfhydric acid through a solution of ar- senious acid, when it forms a bright-yellow, flocculent precipitate. The third compound, corresponding to arsenic acid, has the for- mula AsS5, and has been called sulpharsenic acid. It is obtained by pouring a solution of sulfhydric acid into a solution of arsenic acid, when it slowly precipitates, often requiring the lapse of several days. Sulpharsenic acid is more conveniently prepared by passing a current of sulfhydric gas to saturation, through a solution of ar- seniate of potassa, 2KO,AsOs, converting it into a sulphosalt, 2KS,AsS3, in which the monosulphide of potassium acts the part ARSENICAL TESTS. 285 of a base, and the pentasulphide of arsenic that of an acid, the sul- pharsenic. Sulpharseniate of the sulphide of potassium remains in solution in the liquid, and is decomposed by chlorohydric acid, which disengages sulfhydric acid, and precipitates sulpharsenic acid in the form of a yellow powder. The reaction is expressed by the following equation: 2KS,AsS5+2HC1=2KC1+2HS+AsS5. ON POISONING BY ARSENIOUS ACID. § 234. Poisoning by arsenious acid is almost always fatal when the poison has had sufficient time to pass into the circulation, but it may be relieved when recent. The patient should first be made to vomit, in order to force the stomach to reject the greater part of the poisonous matter it retains. We should then administer hydrated peroxide of iron, or, better still, caustic magnesia, sus- pended in water. These oxides, combining with the arsenious acid, form insoluble arsenites, and destroy the effects of the poison. Hydrated peroxide of iron is prepared by pouring carbonate of soda into a hot solution of a salt of peroxide of iron, and washing the precipitate. Caustic magnesia is obtained by calcining, at a moderate heat, the white magnesia of the shops, which is a hydrocarbonate of magnesia. It is sufficiently calcined when it effervesces but feebly with acids; nor should it be too highly heated, for it then combines less readily with arsenious acid. § 235. Arsenious acid, by itself, is readily recognised by the characters which distinguish it, and which we now proceed to give more in detail than in § 229. A pinch of it, thrown on a burning coal, exhales its characteristic garlicky odour. If a small quantity of the suspected substance in powder be mixed with charcoal, the mixture introduced into a small tube ad, closed at one end, (fig. 256), with some splinters of charcoal above it, and then heated by an alcohol lamp, first at that part of the tube containing the charcoal, and progressively from b to a, that part containing the suspected substance, the arsenious acid will be decomposed by the charcoal, and the vola- tilized arsenic will condense at c, in the form of a brilliant metallic ring, above the heated portion of the tube. All the distinctive characteristics of arsenic may be observed in this ring; thus, it may be sublimed by heat from one part of the tube to another, and may be changed into arsenious acid by com- bustion in the air. For this purpose, a scratch is made on the tube ad (fig. 256), with a file or a diamond, below the deposit of arsenic, Fig. 256. 286 ARSENIC. and the lower part od of the tube detached. Being placed in an inclined position, as in fig. 257, and the ring being heated with an alcohol lamp, it burns in the current of air, and is deposited as arsenious acid, in the form of a white powder, on the up- permost part of the tube. This small quantity may display all the properties which distinguish it. For example, if dissolved in a drop of hy- drochloric acid diluted with water, put into a tube closed atone end, and treated with a solution of sulfhydric acid, a clear, yellow, flocculent precipitate of sulpharsenious acid, or orpi- ment, is formed. The precipitate is insoluble in chlorohydric acid, but dissolves readily in ammonia, producing a colourless solution. The brilliant ring ‘of arsenic, or the deposit of arsenious acid produced by roasting it, may be dissolved in a small quantity of concentrated nitric acid, the solution poured into a porcelain cap- sule, carefully evaporated to dryness, and then treated with a small quantity of a solution of perfectly neutral nitrate of silver. A brick-red precipitate of arseniate of silver is thrown down. It is essential that the solutions be perfectly neutral, for arseniate of silver dissolves in an excess of acid. The arseniate of silver, heated with charcoal in a small tube (fig. 256), affords the brilliant ring of arsenic. § 236. Lastly, arsenious acid may be converted into arsenuret- ted hydrogen, and the properties of the gas ascertained. The operation is extremely important, and requires suitable apparatus ; for, it not only furnishes valuable marks for the detection of ar- senic, but also allows of the easy separation of a minute quantity of arsenious acid diffused through a large quantity of liquid. Let us suppose an apparatus (fig. 258) arranged as for the evolution of hydrogen. In the central tubulure of the bottle A is fitted a straight tube mn, of 8 or 10 millimetres (0.3—0.4 inches) internal diameter, acting as a safety-tube, and allowing the gradual introduction of liquids into the bottle. A smaller bent tube ab, drawn out at its end b, is fitted into the second tubulure. Scraps of very pure zinc are put into the bottle, some water added, and, lastly, small quan- tities of pure sulphuric acid are poured in, so as to produce hy- drogen gas. When the air has been entirely driven out of the apparatus, the jet of gas at the end b is ignited; and the flame presents the ordinary characters of pure hydrogen when burning, not being brilliant, and, if a cold body, such as a porcelain plate or saucer, be brought near it, small drops of water only are de- posited. If a solution of arsenious acid be now introduced through Fig. 257. Fig. 258. ARSENICAL TESTS. 287 the tube, the appearance of the flame is soon changed, assuming a livid tinge, and white fumes of arsenious acid are disengaged. The arsenious acid has been decomposed by contact with the zinc, water, and sulphuric acid, its oxygen having gone to the zinc, and the arsenic, combined with a portion of nascent hydrogen, forming arsenuretted hydrogen. The hydrogen which burns at the end of the tube, therefore, contains arsenuretted hydrogen, which repro- duces fumes of arsenious acid by combustion. When the proportion of arsenious acid introduced into the bottle is not very small, the change in the flame is so evident that the presence of arsenic can be instantly recognised. If the end b of the discharging tube ab be passed into a larger tube, open at both ends, and inclined, a portion of the arsenious acid resulting from combustion will be deposited on the sides of this tube, and the tests before mentioned may be applied. But if the quantity of arsenious or arsenic acid be very small, the change in the flame is no longer sufficiently evident, and the arsenious acid produced by combustion may be completely carried off by the current of gas. We then have recourse to another character, which enables us to detect and even isolate the smallest quantities of arsenic. Arsenuretted hydrogen is formed of two elements of very differ- ent combustibility, its hydrogen having more affinity for oxygen than arsenic. It therefore follows, that if the gas burns in an insuffi- cient quantity of oxygen, the arsenic will oxidize only when all the hydrogen is consumed, and, since arsenuretted hydrogen is easily decomposed by heat, arsenic, arising both from the decomposition of the gas by heat and from its partial combustion, will be deposited. These circumstances may be observed in certain parts of the flame at the end of the tube ab (fig. 258). If this flame be care- fully examined, it will be found closely to resemble fig. 259, being composed of an interior dark por- tion a'cand a luminous envelope oabc, in which the temperature is very elevated. In the pointed part of the flame, toward the extremity of the interior dark portion, the maximum of temperature exists. These two portions of flame and their respective dimensions can be easily seen, by cutting the flame at different points by a plate of glass, and looking behind it. On the external surface of the luminous envelope, the combus- tion is perfect, on account of the excess of atmospheric air ; in the strata of the envelope adjacent to the interior dark part, the com- bustion is imperfect, on account of the want of oxygen; and, lastly, in the dark part there is no combustion, although, in certain parts, toward the plane xz, the temperature is sufficiently elevated to decompose the gas into hydrogen and arsenic. If the flame be Fig. 259. 288 ARSENIC. left free, the arsenic burns toward the end, and is finally disen- gaged in the form of arsenious acid. But if it be cut at xz by a cold body, such as a porcelain saucer, the metallic arsenic is de- posited on the saucer, forming a brilliant spot, possessing metallic lustre when the layer is thick enough. By causing the plane to impinge on different points of the saucer, it may be covered with spots of arsenic, and a sufficient amount collected to recognise its characteristics. The apparatus just described is called Marsh's apparatus, from the English chemist who first devised it, to detect the presence of arsenic in medico-legal researches. It is evident that but a small portion of arsenic is condensed by this process, and, when it is present in very small quantities, the spots are not thick enough to present a metallic lustre, but remain brown ; and although a skilful chemist might not be mistaken, particularly if he carefully test the spots, it is to be feared that errors might arise in less experienced hands. In fact, spots may be produced on porcelain, even when the gas does not contain the least traces of arsenic; but it can always be ascertained whether the spots are arsenical, by subjecting them to the proper chemical tests. Spots are produced on the saucer when the liquid in the bottle is viscous, either because it contains too much sulphate of zinc, or holds organic matter in solution. The disen- gaged bubbles of gas throw out an infinite number of minute globules of liquid, the lightest of which may be carried into the flame, when the salt of zinc, as wTell as the organic matters, would be partially decomposed, forming brown spots of oxysulphide of zinc, or only of carbon. This is avoided by passing the gas through a tube filled with cotton or asbestos, before it reaches the small end at which it burns. It is better, in all cases, to decompose the arsenuretted hydro- gen which accompanies the hydrogen in Marsh’s apparatus, by passing it through a small tube, heated to redness for about 1 de- cimetre pf its length, so that arsenic may be deposited in the form of a narrow brilliant ring, at a short distance beyond the heat, and thus collected on a small surface. The best arrangement of the apparatus is that represented in fig. 260. The bottle A, in which the hydrogen gas is evolved, should be rather small, unless large quantities of liquid are to be acted on, and yet should be large enough to hold all the liquid to be tested, and still leave about one-fifth of its capacity empty. The zinc and water being introduced into the bottle, it is closed with a cork pierced with two holes, into one of which is inserted the tube mn, of about 1 cm. (J inch) in diameter, for pouring in the liquid, and which dips a little way into the water. To the second tube is fitted a bent tube abc, having a bulb at b, in which the greater part of the water carried over condenses. A glass tube ARSENICAL TESTS. 289 Fig. 260. cd, filled with asbestos, retains the particles of the solution car- ried over by the current of gas. Lastly, a narrow tube efg, of 3 or 4 dec. (12-16 in.) length, and drawn out to a point g, ter- minates the apparatus. Hydrogen is first evolved, to expel the air from the apparatus ; the tube dfg is then heated, for about 1 dec. of its length, by live coals on a chafing-dish. The glass tube should be difficult of fu- sion, or else surrounded by a sheet of tinsel, to prevent its bending. A screen e protects the part fg of the tube from the heat. The gas being lighted at the orifice g, its disengagement is continued for some time, to observe whether a deposit takes place in the part fg, or whether spots can be obtained on a porcelain saucer, in order to ascertain whether the reagents themselves are entirely free from arsenic. This being done, the suspected liquid is introduced, and a gentle evolution of hydrogen kept up by adding a suitable quantity of sulphuric acid, so that the flame cannot attain a length of more than 5 or 6 mm. (| in.) The greater part of the arsenic is deposited at/, a short distance beyond the screen, but as there is almost always a small quantity of arsenuretted hydrogen which escapes decompo- sition and burns in the flame, a portion of it is carefully collected on saucers, and examined for the characteristic reactions of arsenic. If the liquid contained antimony, a brilliant metallic ring is also obtained, in the tube/7 (fig* 260); but it is sufficiently distinguished from that produced by arsenic, by its want of volatility, and other characters to be described when treating of antimony. § 237. The processes above described are of easy execution, and admit of our detecting, with perfect certainty, the smallest quan- tities of arsenic, when it exists in the state of arsenious or arsenic acid, or even of a sulphide; for the latter can be readily trans- formed previously into arsenic acid, by means of nitric acid. But the problem is less simple when it is required to detect the presence of a small quantity of arsenic in large masses of organic matter, as most frequently happens in cases of poisoning. The process to be then pursued will now be succinctly described. If a portion of the food supposed to have been poisoned still re- 290 ARSENIC. main, we must examine if there be not, at the bottom of the vessel, a deposit of arsenious acid, as a white powder, which can be im- mediately recognised by the tests above given. A similar examina- tion should be made of the matters vomited. If these researches are fruitless, the food or matters vomited should be strained through a piece of clean linen, previously washed in distilled water, whereby they are separated into a liquid and a solid portion, which are to be first separately and then conjointly treated. The liquids are evaporated in a porcelain capsule, but as they frequently contain organic matter in solution, they become too viscous to be introduced directly into Marsh’s apparatus, where they would produce too much froth, and the experiment could not be accurately conducted. Moreover, since the presence of these organic matters changes re- markably the reactions by which the arsenic might be recognised, they are destroyed by concentrating the liquids highly, and then adding a quantity of oil of vitriol proportioned to the organic mat- ter supposed to exist in the solution. Upon evaporating to drive off the sulphuric acid, organic matter is destroyed, and assumes the form of a spongy charcoal, which is sprinkled with concen- trated nitric acid, and again heated to drive off this acid—reddish fumes being copiously given off. The arsenic, if present, is con- verted into arsenic acid, which dissolves very readily in water. The residue is therefore treated with a small quantity of boiling distilled water, and filtered, and, if the carbonization has been care- fully performed, a liquid is generally obtained free from colour, or nearly so, which is easily managed in Marsh’s apparatus. The solid matters remaining in the linen should also be carbon- ized by sulphuric acid, by sprinkling them with about one-fifth of their weight of concentrated sulphuric acid, and heating them. When the whole mass becomes fluid, the sulphuric acid is driven off by heat, the residue sprinkled with nitric acid, which is also driven off, and, lastly, treated with boiling distilled water. A limpid liquid is obtained by filtration, presenting the same ap- pearance as that resulting from the treatment of the liquid portion. The two liquids are mixed, and treated together in Marsh’s apparatus. When there is a considerable quantity of arsenious acid in the matter subjected to experiment, we may effect its carbonization by sulphuric acid and the successive evaporations in porcelain cap- sules ; but, if the proportion of ,poison be small, it is always to be feared that some of the arsenious acid may be carried off at the high temperature required for expelling the sulphuric acid. This danger is especially imminent when the substances contain chlo- rides, because chloride of arsenic, which is very volatile, may be formed. In all cases, it is better to effect the carbonization in a glass retort connected with a cooled receiver, for collecting the distilled liquids, which may afterward be examined for arsenic. If the chemist be required to investigate a case of poisoning, ARSENICAL TESTS. 291 after the death of the patient, he should examine the contents of the stomach, and the urine in the bladder, in the manner pointed out above. Lastly, long after the decease of the victim, it may become his duty to examine a corpse in a more or less advanced stage of de- composition. He must then operate on what remains of the stomach, and on the viscera, such as the liver, heart, spleen, etc., which are generally attacked by the poison. They are carbonized in the same way, by sulphuric acid, in a glass retort, after having divided them into small pieces. Animal matters may also be decomposed by suspending them in water, after having ground them in a mortar, and passing a current of chlorine through the liquid, until the organic matter is deposited in the form of colourless flakes, and the liquid is saturated with chlorine. The bottle is then corked, allowed to stand for 12 hours, when the odour of chlorine should still be distinct, then filtered and concentrated in a retort adapted to a receiver. The small quantity of concentrated liquid remaining in the retort is treated in Marsh’s apparatus, and, if necessary, the liquid con- densed in the receiver is also examined for arsenic. It is unnecessary to say that all the chemical reagents used in these processes should be pure, and previously tested with the greatest care, to ascertain that they do not contain the slightest trace of arsenic. The chemist may then have entire confidence in the result of his experiments, if they have been properly con- ducted. But, as it is essential that the judges should share this confi- dence, and that no doubt can hang on the result of the experi- ments, if it show the presence of arsenic, the chemist should be required to perform, contemporaneously with the actual experi- ments, similar operations without the suspected matters, with the same reagents, in the same quantity, and in exactly similar appa- ratus. He should deliver to the judge, on the one hand, the tube dfg (fig. 260) of Marsh’s apparatus, in which he has finally ob- tained the result of his experiments of the suspected matters, as well as the saucers on which he has endeavoured to produce spots; and, on the other, the analogous tube of the other Marsh’s appa- ratus, in which he has finally obtained the result of the operations performed on the reagents alone, as well as the saucers on which he has endeavoured to produce spots. Such a comparison of the results can leave no doubt on any one’s mind.* * For further details, see the report made to the des Sciences, on arsenic in cases of poisoning.—Comptes Eendus de V Academic des Sciences, tome xii. p. 1076. 292 BORON. Equivalent B = 10.9 (136.15, 0 = 100). § 238. Boron* is found in nature combined with oxygen, in the state of boracic acid; which either exists alone, or in combination with bases. In order to extract boron from boracic acid, the acid is first fused at a red-heat in a platinum crucible, to drive off the water it contains, then reduced to a fine powder, and introduced with potassium or sodium into a glass tube, closed at one end, well dried, and heated by a few coals. A slight detonation takes place at the moment of reaction. The potassium seizes upon the oxygen of a portion of the boracic acid, and is converted into oxide of potassium or potassa, which combines with the undecomposed boracic acid, and forms borate of potassa. By treating the mass when cold with water, borate of potassa is dissolved, and boron floats in the liquid, in the form of a very fine brown powder, which is collected on a small filter, and washed with distilled water until a drop of the wash-water, evaporated on a clean watch-glass, leaves no perceptible residue. Boron forms a brown powder, which does not fuse when heated to redness in a current of hydrogen, or any other gas which exerts no chemical action on it. Heated in contact with the air, it burns and is converted into boracic acid; but it is difficult to oxidize it completely in this manner, for, as fast as the boracic acid forms, it fuses, and forms a glazed coating, which protects the yet unaltered boron from contact with the air. COMBINATION OF BORON WITH OXYGEN. Boraci.c Acid, B03. §239. Only one compound of boron with oxygen is known— boracic acid—which is found in nature, either free or in combina- tion with soda, forming a salt known in the arts by the name of borax. In certain volcanic districts of Tuscany, called the Maremmce of Tuscany, jets of gas and vapour constantly exhale from fissures in the soil, which are called suffioni (suffumes), and which contain small quantities of boracic acid. Small lakes of water (lagoni) have formed around the fissures, through which the jets of vapour and gas escaping, throw up liquid cones, and then pass into the air in whitish clouds. * Boron was discovered simultaneously, by Davy in England, and Gay-Lussac and Thenard in France. BORACIC ACID. 293 Around these centres of eruption, basins of clay have been built in rough masonry, in which two or more suffioni terminate; and the water of surrounding springs is conducted into the uppermost lagoon A (fig. 261). After 24 hours, during which the waters Fig. 261. have been constantly agitated by the current of subterraneous va- pours, the liquid in the basin is allowed to run into another, B, where it remains for the same length of time, and becomes charged with an additional quantity of boracic acid. It is then passed successively into the lagoons C and D; the liquid which has run out of a lower basin being immediately replaced by that of an upper one. The solution in the last basin D is conveyed into reservoirs E, F, where it is allowed to remain for 24 hours, and in which are de- posited the greater portion of the earthy substances held in sus- pension. The supernatant liquid is drawn off and passed succes- sively into a series of shallow leaden pans, or evaporators, G, arranged as in fig. 261, over a flue in mason work, through which the hot vapours of a suffione are constantly passing, and afford sufficient heat to evaporate the liquid. After remaining 24 hours in the first evaporator, the liquid is diminished by evaporation to one-half, and is then conveyed into the evaporator immediately below, where it remains for the same length of time ; thus descending from pan to pan, until, when it reaches the last, it is so concentrated that boracic acid crystallizes on cooling in the crystallizers A (fig. 262). The crystallized acid is collected in baskets C, where it is allowed to drain ; and then 294 BORON- Fig. 262. Fig. 263. dried in an oven (fig. 263) with a double bottom H, through which the vapour of a suffione circulates. The boracic acid thus obtained is far from being pure, as it con- tains 18 to 25 per cent, of foreign substances, from which it is purified by solution in boiling w’ater and crystallization. Boracic acid is often prepared in the laboratory from the borax of commerce, which is very pure, by dissolving 1 part of borax in 2| pts. of boiling water, and adding chlorohydric acid until the liquid strongly reddens litmus. On cooling, the boracic acid crys- tallizes in thin plates, which are allowed to drain, and then washed with a little water. If absolutely pure boracic acid be required, it must be again dissolved in boiling water and recrystallized. Crystallized boracic acid forms colourless scales, containing 43.6 per cent, of water of crystallization. Subjected to heat, it first melts in this water, which is then disengaged, and, if it be heated to redness, it fuses into a colourless liquid, which, on cooling, pre- sents the appearance of a perfectly transparent vitreous mass. Between the states of perfect liquidity and complete solidity, bo- racic acid passes through all the intermediate stages, and, like all substances possessing this property, it does not crystallize by fusion, so that it remains perfectly transparent after solidification. But its transparency is not permanent, for, even when preserved in hermetically sealed tubes, it ultimately becomes opaque, from its molecules at common temperatures tending to aggregate, according to the laws of crystallization, which govern them at this tempera- ture, so that a multitude of small cleavages result, which soon destroy its transparency. Exposed to the air, the fused acid is soon covered with a pulverulent substance, produced by its absorb- ing water from the air, and changing into a hydrated acid. 100 parts of water dissolve 2 pts. of the crystallized acid, at the temperature of 50°, and 8 pts. at 212°; so that a solution, satu- rated at the boiling point, deposits f of its acid when it descends to ordinary temperatures. Its solution is slightly acid, reddening litmus, but, like a feeble acid, it produces a purplish-red colour; and yet, in the cold, it ex- pels carbonic acid from its compounds. In the dry way, it expels BORACIC ACID. 295 the most powerful acids, owing to its great fixedness, for it does not boil even at a white heat. But yet, at this temperature, the tension of its vapour is sufficient to allow the acid to evaporate en- tirely, in a short time. At a red-heat, it expels sulphuric acid from the sulphate. The composition of boracic acid has been determined by ascer- taining, experimentally, the increase in weight of 1 gramme of boron, when heated in the air so as to convert it into boracic acid. It has been found to consist of Oxygen 68.T8 Boron 31.22 moo It is difficult to give the formula proper to boracic acid, for the number of definite compounds containing boron is still very limited; and the rules we have applied for determining the equivalents of simple bodies are inapplicable to it. Some chemists adopt for it the formula BOs; in which case the equivalent of boron is obtained by the proportion 68.78 : 31.22 : : 48 : x, whence x — 21.8. Others adopt the formula B03, whence the equivalent of boron is given by the proportion 68.78 : 31.22 : : 24 : x, whence x = 10.9. The acid, crystallized by solution, as mentioned above, is com- bined with 43.6 per cent, of water, which contains a quantity of oxygen equal to that which exists in the anhydrous acid. The formula of the crystallized acid will then be According to the first hypothesis B06-f-6H0 “ “ second “ BCL+3HO COMBINATION OF BORON WITH CHLORINE. § 240. This compound is obtained by heating boron in a current of chlorine, or, more readily, by heating an intimate mixture of boracic acid and carbon in a porcelain tube, while a current of dry chlorine is passed through it. Chloride of boron is a colourless gas ; gives off dense fumes in a moist atmosphere ; has a density of 4.035 ; by contact with water is decomposed into chlorohydric and boracic acids. It formula is, therefore, that of boracic acid in which the oxygen is replaced by an equivalent quantity of chlorine. 1 volume of the gas contains vols. of chlorine, thus Chloride of Boron, BC13. 296 BORON, Boron 0.375 9.28 1J yoI. chlorine 3.660 90.72 4.035. .100.00 COMBINATION OF BORON WITH FLUORINE. Fluoride of Boron, BF3. § 241. A gaseous compound of fluorine and boron is obtained by heating, at a very high temperature, in a small porcelain retort, a mixture of 2 parts of fluor-spar and 1 pt. of fused boracic acid. A portion of the acid is decomposed, its oxygen combining with calcium to form lime, which yields borate of lime with the unde- composed boracic acid; while the fluorine and boron combine to form fluoride of boron. The reaction may be represented by the following equation: 2B03+3CaF=BF3+B03,3Ca0. Fluoride of boron is a colourless gas, with a suffocating odour, and a strongly acid taste; its density is 2.37; it is extremely soluble in water, and has so great an affinity for it, that it carbon- izes organic substances, like oil of vitriol (§ 134). In consequence of its great affinity for water, it fumes copiously when exposed to the air. The composition of fluoride of boron corresponds to that of bo- racic acid, its formula being BF3. Water dissolves 700 to 800 times its volume of fluoride of boron, and the solution is easily obtained, in a concentrated form, in the following manner: Equal parts of fluor-spar and borax are fused together, pulve- rized, and heated in a glass retort, with concentrated sulphuric acid; an acid liquid distils over, which is a very concentrated solu- tion of fluoride of boron in water. If the solution be diluted with a larger quantity of water, it decomposes into boracic acid, which separates, and a peculiar acid which has been called borofluohydric acid. The latter is probably analogous to the silicofluohydric, of which we shall presently treat, and which has been more thoroughly examined. 297 SILICIUM. Equivalent Si = 21.3 (266.7 O —100). § 242. Silicium* is one of the most widely diffused bodies in nature; for its combination with oxygen, silicic acid, is one of the most common substances on the surface of the globe. Silicic acid, heated with potassium, yields silicium and silicate of potassa; but the decomposition is difficult, and does not afford pure silicium. It is preferable to employ the potassium for decomposing a compound of fluoride of silicium and fluoride of po- tassium, the preparation of which will be given hereafter. The two substances are introduced into a dry glass tube and heated: Double fluoride of silicium and potassium. Fluoride of potassium. Silicium Fluorine' Potassium... ( Fluoride of silicium... Fluoride of potassium. 3KF,2SiF3+6K=9KF+2Si. The product is treated with cold water, which dissolves the fluoride of potassium; the silicium is collected on a small filter, and washed with distilled water, until the washings leave no per- ceptible residue on a glass plate. Silicium is a brown powder, infusible when heated in a close vessel, takes fire when heated in the air, and is converted into silicic acid. COMBINATION OF SILICIUM WITH OXYGEN. § 243. Only one compound of silicium and oxygen is known— the silicic acid—which is generally known by the name of silex, and is one of the most common substances in nature. Isolated, it constitutes rock crystal, quartz, quartzose sands, sandstone, etc. Combined with alumina, potassa, or soda, lime, and the oxide of iron, it constitutes many minerals, which are aggregated into granites, slates, etc. In short, all rocks which are not calcareous are silicious. Colourless rock crystal exhibits crystallized and pure silicic acid. The general form of the crystals is a six-sided prism, terminated a six-sided pyramid (fig. 58), belonging to the third or hexagonal system of crystallization. Rock crystal is a very hard substance, which scratches glass,, and has a density of 2.6. Silicic Acid, Si03. * Silicium was first obtained in a pure state by Berzelius. 298 SILICIUM. The highest temperature of our furnaces does not melt rock crystal; but it fuses into a vitreous globule in the flame of a mix- ture of oxygen and hydrogen. At ordinary temperatures, it is not affected by contact with any reagents, except fluohydric acid, which acts upon it rapidly. Caus- tic potassa has a similar effect at a high temperature. When silicic acid is obtained in a disaggregated state, it presents more marked characters. To obtain it in this state, 1 part of finely powdered quartz, and 4 of carbonate of potassa or soda are melted in a platinum cru- cible, whereby a portion of the carbonic acid is driven off, and silicate of potassa formed. Treated with water, the mass dissolves entirely when subjected sufficiently long to a high temperature. If the liquid be diluted with a large quantity of water, and chlo- rohydric acid be added until a strongly acid reaction is manifest, the silicic acid is separated from its combination with the potassa, but remains suspended in the liquid, in the state of transparent jelly, and cannot be separated by filtration. If the alkaline mat- ter be dissolved in a small quantity of water, and chlorohydric acid added to the dense solution, the silicic acid forms a gelatinous, flocculent precipitate, which can be filtered. Nevertheless, its complete separation only takes place by eva- porating the liquid supersaturated by the acid to dryness, and treat- ing the residue with boiling water. The silex then separates in the state of a stiff jelly, which is completely arrested by the filter. It is then probably in the state of a hydrate, but soon parts with its water by drying, and assumes the appearance of a light, white, mealy powder, which becomes very hard by calcination. It is sometimes deposited in the form of a transparent jelly, when certain substances containing it are allowed to decompose spon- taneously and slowly. Thus, silicic ether,, kept in a badly corked bottle, gradually loses all its ether, while the silica remains in the form of a perfectly transparent jelly, which, in time, becomes very hard, without losing its transparency. § 244. The composition of silicic acid is deduced from the analy- sis of the chloride of silicium, which will soon be described. Chlo- ride of silicium is decomposed, by contact with water, into silicic and chlorohydric acids. Silicic acid is therefore obtained from chloride of silicium, by substituting an equivalent quantity of oxygen for its chlorine. By analyzing chloride of silicium, the composition of the acid may be at once ascertained. By following closely the method described (§214) for ascertaining the composi- tion of phosphorous acid, it will be found that silicic acid is com- posed of Silicium 47.06 Oxygen 52.94 100.00 CHLORIDE OF SILICIUM. 299 The same difficulty is experienced in establishing the formula of silicic, as that of boracic acid, for silicium, like boron, affords but few definite compounds. The majority of chemists admit for it the formula Si03, analogous to that of sulphuric acid; from which the equivalent of silicium is then given by the proportion 52.94 : 47.06 : : 24 : x, whence x — 21.3. Others write the formula SiOa, which gives for the equivalent of silicium, 52.94 : 47.06 : : 16 : x, whence x = 14.2. Lastly, some adopt the formula SiO, when the equivalent be- comes 7.1. We shall adopt the formula Si03; not that it is the most con- venient, but simply because it has hitherto been most generally adopted. We therefore take 21.3 as the equivalent of silicium.* COMBINATION OF SILICIUM WITH CHLORINE. Chloride of Silicium, SiCl3. § 245. If silicium be heated in a current of chlorine, it takes fire, and a colourless volatile liquid is formed, which is the chloride of silicum SiCl3. It may be more easily obtained by passing chlo- rine over a mixture of silex and carbon heated in a porcelain tube (fig. 264). Chlorine alone will not expel oxygen from silicic acid, even at the highest temperature; hut the decomposition is easily effected in presence of carbon, which combines with the oxygen of Fig. 264. * The late investigation of Kopp on the difference of the boiling points between the bromide and chloride of silicium, and that of Pierre on the substitution of sulphur for chlorine in the chloride of silicium, strongly confirm the older view, that silica is Si03, and hence that silicium = 21.3.—J. C. B. 300 SILICIUM. the silicic acid to form oxide of carbon ; the chloride of silicium is collected in a refrigerated receiver. The silica used in this ex- periment should be the very finely divided, such as that obtained by decomposing silicate of potassa by an acid; for quartz, even when reduced to an impalpable powder, affords only traces of chloride of silicium. The best method is to mix the silex intimately with an equal weight of lampblack, and sufficient oil to form a paste, and to form it into small balls, which are rolled in powdered charcoal, and cal- cined in a close crucible. The balls, thus rendered porous, are put into the porcelain tube. To make a larger quantity of the chloride, the porcelain tube is replaced by a stoneware retort C (fig. 265), holding about a litre Fig. 265. (a quart), with a tubulure a, to which a smaller porcelain tube b is fit- ted, and carried to the bottom of the retort. The current of dry chlo- rine is passed through this tube. A tube passing through a con- denser is adapted to the neck of the retort, and is succeeded by a U-shaped tube D plunged in a refrigerating mixture contained in an inverted tubulated bell-glass. A straight tube is attached to the lower part of the U-shaped tube, which passes through the tubulure of the bell-glass into a dry bottle, in which the liquid chloride of silicium is collected. When thus made, it is of a yellow colour, owing to an excess of chlorine, which it holds in solution, and of which it is deprived by shaking it with a small quantity of mercury, and is then obtained perfectly pure by distillation. FLUORIDE OF SILICIUM. 301 Chloride of silicium is a colourless, volatile liquid, of the density 1.52; it boils at 138°, and gives off acid fumes in the air. By contact with water, it is decomposed into chlorohydric and silicic acids, which proves its correspondence to silicic acid, the oxygen of the acid being replaced by an equivalent quantity of chlorine. Advantage is taken of this reaction to deduce the com- position of the acid from the analysis of the chloride, as it presents fewer difficulties than the direct analysis of the acid. We thus find that chloride of silicium is composed of Silicium 16.71 Chlorine 83.29 moo We may, therefore, express its formula SiCl3, if we admit Si03 for silicic acid. SiCl3 “ “ Si03 “ “ SiCl “ “ SiO “ “ The density of its vapour has been found to be 5.9. 1 volume of the chloride contains 2 volumes of chlorine; for if, to twice the density 2x2.44 of chlorine, we add the corresponding quantity of silicium, which is calculated by the proportion 83.29 : 16.71 : : 4.88 : x, whence x — 0.98, we find 2 vols. chlorine 4.88 Silicium 0.98 5JS6 which does not differ sensibly from the density of the gaseous chloride found by experiment. COMBINATION OF SILICIUM WITH FLUORINE. Fluoride op Silicium, SiF3. § 246. This compound is obtained by heating together, in a glass flask, equal parts of fluor-spar and pounded glass, with 6 or 8 parts of the strongest oil of vitriol. (See fig. 199.) The silicic acid of the glass yields its oxygen to the calcium of the fluor-spar, forming lime, which combines with the sulphuric acid, and the fluorine is united to silicium, to form the gaseous fluoride of sili- cium. Supposing only the silicic acid of the glass to be present, the reaction may be represented by the following equation: 3CaF+Si03+3S03=3(Ca0,S03)+SiF3. The apparatus employed in the experiment should be previously 302 SILICIUM. dried with the greatest care, since fluoride of silicium is readily decomposed by contact with water. Fluoride of silicium is a colourless gas, which must be collected over mercury, as water instantly decomposes it; its density is 3.57 ; exposed to the air, it gives ofl1 very dense acid fumes. Its composition corresponding to that of silicic acid, its formula is SiFs. § 247. When fluoride of silicium is decomposed by water, gela- tinous silica is deposited, and the liquid contains a peculiar acid compound, called silicofluohydric acid. Reaction takes place be- tween 3 equivalents of fluoride of silicium and 3 equiv. of water. But, of the 3 equiv. of the fluoride, only one is decomposed, pro- ducing 3 equiv. of fluohydric acid, which combine with the 2 equiv. of undecomposed fluoride to form silicofluohydric acid. The reac- tion is therefore represented by the following equation : 3SiF8+3HO=3HF, 2 SiF3+Si03. The formula of silicofluohydric acid is, therefore, 3HF,2SiFs. When this acid is saturated by a base, the hydrogen of the fluo- hydric acid is alone replaced by an equivalent quantity of the metal of the base ; thus potassa gives the following reaction: 3HF,2SiFg+3KO=3KF,2SiF8-F3HO. The silicofluohydrate of potassa is therefore a double fluoride of potassium and silicium, with the formula 3KF,2SiFg. The gelatinous silica which is deposit- ed during the decomposition of the flu- oride by water would soon obstruct the orifice of the tube conveying the gas, if it be dipped into water, and might burst the apparatus. The end of the tube is therefore plunged into a stratum of mercury (fig. 266), about an inch thick, at the bottom of the test-glass, before pouring in the water; so that the gas meets with no water to decom- pose it until after having passed through the stratum of mercury.* The fluoride may likewise be prepared in a glass retort (fig. 267), the neck of which connects writh a re- Fig. 266. * Even in this case, the rapid passage of bubbles of gas upward will form tubes of gelatinous silica to the top of the water, through which the gas would then escape into the air. To avoid this inconvenience, it is necessary to break the tubes of silica by stirring with a glass rod.—J. G. B. FLUORIDE OF SILICIUM 303 ceiver containing water, without passing through a cork, so that the flask may be easily turned around the neck of the retort, and its sides kept constantly moist. The fluoride, being heavy, on the surface of the liquid in the receiver, and a pellicle of gelatinous silex forms, which would soon prevent the action of the water if the flask were not frequently turned. When a sufficient quantity of fluoride has been decomposed, it is filtered through a cloth, and the residue expressed. To render the liquid more transparent, it should be passed through filtering paper, but even then a small quantity of silica remains in sus- pension. Silicofluohydric acid forms a very acid solution, and combines with bases forming double fluorides, the composition of which has been indicated above. Some of these compounds are insoluble; among others, that which it forms with potassa. We have already taken advantage of this property of the acid to precipitate potassa from its solution (§170). If the acid solution be evaporated to dryness with the gelatinous silica deposited during its preparation, the whole substance dis- appears ; water and fluoride of silicium being disengaged. Heat thus produces the inverse reaction of that which takes place in the cold between fluoride of silicium and water; so that now we have, Fig. 267. If the evaporation be performed in a glass vessel, it remains uninjured and retains its transparency; but if, on the other hand, silicofluohydric acid alone, separated by filtration from the depo- sited silica, be evaporated in a glass vessel, it disappears entirely, and the sides of the vessel are affected, for they must afford suffi- cient silica to transform the silicofluohydric acid into fluoride of silicium. 3HF,2SiF3+Si03=3SiF3+3H0. 304 CARBON. Equivalent C = 6 (75 O = 100.) § 248. Carbon appears under very different aspects. It is found in nature, perfectly pure and crystallized, in the diamond; which is met with in alluvial formations, resulting from the disin- tegration of older rocks whose detritus has been carried down by water and has covered extensive valleys and plains. Its principal localities are India, the Island of Borneo, and Brazil. Diamonds are rarely discovered among this detritus, and, in order to find them, it is necessary to wash and sort large quantities of sand. The surface of a crude diamond is generally rough and slightly translucent. Its crystalline form is sometimes well defined, be- longing to the regular system of crystallization, and its primitive form the regular octahedron (fig. 20); but the octahedron is most frequently modified by secondary planes, and the crystal presents the appearance of fig. 27. The crystalline faces of the diamond are rarely plane, but more or less convex, so that the edges them- selves are curved. The curvature is especially evident in those crystals presenting the general appearance of the regular octahe- dron ; but they are really trisoctahedrons (fig. 27), that is, octa- hedrons the faces of which have been replaced by low triangular pyramids. The edges of the pyramids being often completely destroyed by the friction the crystal has undergone during trans- portation with the detritus, it only retains the general aspect of an octahedron with convex faces. It is generally colourless, but is sometimes found tinged with various hues, the most frequent of which are yellow and a more or less dark brown: blue, rose, and green diamonds have also been found. The density of the diamond varies from 3.50 to 3.55. The diamond is the hardest of all known substances, scratching all without exception; and its natural facets are harder than those produced by cutting. The latter property is very common among minerals. Glaziers use diamonds to cut glass in any given direc- tion : they select diamond sparks presenting natural curved sur- faces, and mount them on a suitable handle, to make the instru- ment known as a “glazier’s diamond.” To separate a strip of any width from a pane of glass, they lay a rule along the line to be fractured, and then slide the diamond along the rule, tracing on the glass a very fine line, which renders the glass frangible in this direction; so that, by bending it, it cracks neatly along this line.* * There is considerable skill required to sever a common glass pane with neatness and confidence. In tracing the line, the diamond should always be held CARBON. 305 The diamond can be cut only by its own dust. This operation is begun by rubbing two rough diamonds against each other, and carefully collecting the fine powder which falls down. An outline is thus made of the form the diamond is to receive, and to com- plete the form and polish it, it is fastened to a copper cup held by steel pincers. It is rubbed on a plate of soft steel, spread with some diamond-dust and olive-oil, and made to revolve very rapidly horizontally around its centre. The various faces to be cut are successively presented to it. The rough diamonds which are re- jected are ground in a mortar, and the dust used for diamond cutting. The diamond being pure crystallized carbon, many attempts have been made to crystallize carbon artificially, in the hope of producing it, but without success. Carbon is completely infusible at the highest temperature which can be generated in our furnaces, so that we cannot hope to crystallize it by means of fusion. And, on the other hand, as we know no solvent for it, it cannot be crys- tallized by means of solution. Cast-iron may, indeed, at a very high temperature, dissolve a greater portion of carbon than it can retain at a lower temperature; and, on cooling, it parts with a portion, which assumes crystalline forms. But those are very bril- liant, black laminae, frequently quite large, but in no wise re- sembling the diamond. This crystallized carbon is called graphite. The diamond, placed between the two charcoal cones of a very powerful battery, attains an excessively elevated temperature, and becomes so brilliantly incandescent as to be painful to the eye. But, if observed through a smoked glass, it is seen to swell considerably and separate into several fragments. After cool- ing, it has entirely changed in appearance, having become of a metallic gray-colour, friable, and precisely similar to the coke arising from bituminous coal. This experiment seems to prove that a high temperature is not favourable to the existence of carbon in the state of the diamond, and that its formation did not take place at a very high temperature. § 249. Nature also affords us carbon in a crystalline state en- tirely different from the diamond, in the state of very fine spangles of a metallic gray-colour, which are often extremely small, and aggregated together, forming shining masses, easily divided by a knife, and leaving a leaden-gray streak on paper. This is the substance known in the arts under the name of plumbago, graphite, and blacklead, of which lead-pencils are made. Organic bodies are, as we have frequently said, composed of carbon, hydrogen, oxygen, and nitrogen. When subjected to a in the same position from the beginning to the end of the cut, commencing the cut near to one edge, and terminating it lightly at the further edge. The best method of breaking is to hold the pane by both hands, one on each side of the end of the cut, and to bend and pull it slightly apart at the same time.—J. C. B. 306 CARBON. high temperature, the hydrogen, oxygen, nitrogen, and a portion of the carbon are driven off; and that portion of the carbon which remains presents various appearances, according to the nature of the organic substance. Thus, if a piece of wood be calcined, the coal which remains is black, and exhibits in its fracture the struc- ture of the wood from which it was derived. If sugar or an ani- mal substance be calcined, an extremely light, black, brilliant, swollen coal is obtained, presenting the appearance of fusion. It was not, however, the carbon that fused, but the organic matter which, beginning to melt at the first accession of heat, became more and more doughy as decomposition advanced, and swelled up from the disengagement of gases. Pit-coal, or bituminous coal, calcined apart from the air, gives a coal, called coke, which varies according to the quality of the coal. Fat coal undergoes incipient fusion before being decom- posed, and produces a swollen coke of a brilliant metallic gray. The anthracites, which lose but a small quantity of their weight by calcination, afford a coke having the shape, and generally the appearance, of the original piece of anthracite. Certain organic matters, burning in the air, undergo only an imperfect combustion, emitting a smoky flame, which deposits carbon in the form of an extremely fine black powder. A deposit of this kind is obtained, when a plate of glass is held in the upper part of the flame of a candle. This pulverulent carbon is known in the arts by the name of lampblack, and is generally prepared by burning rosin or tar. The apparatus generally used con- sists of a cylindrical chamber of stone or brick, large enough to allow a sheet-iron cone, having a hole at its apex, and which acts as a chimney, to slide up and down. The walls of the chamber are hung with coarse cloth, which facilitates the deposition of the flakes of lampblack. A cast-iron pot, containing the rosin, is heated by a furnace without, and the entrance of the air is regu- lated by the working-holes. The incomplete combustion of the combustible vapours produces a considerable quantity of lamp- black, which is deposited on the interior of the cone, and chiefly on the walls of the chamber. When the operation is terminated, the cone is allowed to descend: being of a diameter exactly to fill the chamber, it scrapes the sides of it, and throws down all the lampblack on the floor. Lampblack thus made is always mixed with empyreumatic oils, and, when used as carbon in the laboratory, it must be cal- cined in a crucible, apart from the air.* * Lampblack is made in Philadelphia by setting fire to rosin or coal-tar con- tained in a shallow cast-iron vessel, of some 5 feet diameter, which is placed at the outer end of a horizontal semicylindric flue of masonry, of 25 to 40 feet in length. These flues open into a large chamber, with brick or stone walls, and CARBON. 307 Carbon, in these various states, presents very different physical properties : its specific gravity ranges over a wide space ; thus, The density of the diamond is 3.50 That of natural graphite 2.20 That of powdered coke varies from ... 1.60 to 2.00 The density of charcoal varies, according to its porosity. At first sight, it seems lighter than water, on the surface of which it floats; but it is easy to show that this property depends on its containing cavities into which the water cahnot penetrate, for, if pulverized, its powder sinks to the bottom. Ordinary charcoal is a bad conductor of heat, so that a piece of it, lighted at one end, may be held by the fingers very near the burning portion, without communicating much warmth. It is also a bad conductor of electricity; but becomes a good one, when vividly calcined. Thus the half-burned coals from our fireplaces are used to surround the end of lightning-rods, to facilitate the discharge of the electricity into the earth. § 250. The very porous varieties of charcoal possess remarkable powers of absorption, which have been usefully applied in the arts. If a red-hot coal be plunged into mercury, in order to extinguish it, apart from the air, and, without removing it from the mercury, it be then passed into a bell-glass containing any gas, a consider- able quantity of the gas will be absorbed, the quantity varying according to the nature of the gas and that of the coal. A mea- sure of charcoal from boxwood absorbs 35 measures of carbonic acid gas, and 90 measures of ammoniacal gas. If a porous charcoal be left for some time in an atmosphere of sulphuretted hydrogen, so that a large quantity of it is absorbed, and be then passed into a bell-glass filled with oxygen, the coal becomes heated, sulphur separates, and water and sulphurous gas are formed. The combustion is sometimes so sudden that explo- sion ensues. Similar phenomena takes place with other combus- tible gases. Charcoal also absorbs colouring matters dissolved in water. If red wine be shaken for a few moments with certain pulverized porous charcoals, it loses its colour entirely. Charcoal likewise an iron (or sometimes board) roof, and containing 150,000 to 350,000 cubic feet. The whole building is either closed as tightly as practicable, or a portion of smoke is allowed to escape through a chimney or windows covered with coarse wire-gauze. The lampblack deposits on the walls and floor, from the former of which it soon detaches itself, and the whole is collected in a thick layer on the floor. There being two flues, with their separate iron pans, doors, &c., as soon as one is burned out, and a little draft allowed to enter the building, the second is fired,- and the operations are thus continued day and night. The quantity made in three establishments, when in active operation, is nearly two tons daily.—J. C. B. 308 CARBON. absorbs many odorous matters; thus, stagnant waters, exhaling an infectious stench, lose it by contact with charcoal; and it is for this reason that the inside of wooden water-tanks for ships are always slightly charred. The different kinds of charcoal possess very different powers of absorption. In graphite and the bituminous coals they are null, but are strongly marked in wood charcoal, and powerful in pro- portion to the number of the pores in the coal. The charcoal derived from the calcination of bones presents this quality in the highest degree. By c’alcining bones in close vessels, the animal matter they contain is carbonized, and a very porous coal is ob- tained, mixed with the earthy matter of the bones, which is called in the arts animal charcoal or boneblack. The bones are cal- cined in large cast-iron cylinders, arranged horizontally in a fur- nace, and having a pipe at one end, which communicates with a refrigerating apparatus, in which the ammoniacal products are collected for future use. When the calcination is ended, the coal is withdrawn, extinguished in an extinguisher, and reduced in suitable mills to powder of different fineness. § 254. Carbon burns in the air and is converted into carbonic acid gas. Its combustion in oxygen is much more vivid. The charcoal is attached to the extremity of an iron wire, ignited in the blowpipe flame, passed through an alcohol-lamp, and quickly plunged into a vessel filled with oxygen, where it burns with great splendour. The formation of an acid gas by the combustion is easily recognised by pouring into the vessel a little blue infusion of litmus, which is reddened. If lime-water be introduced, it be- comes milky, and carbonate of lime is precipitated. The various kinds of charcoals are combustible in an inverse proportion to their density. Thus, wood charcoal burns in the air; compact coke, especially that of anthracite, only burns in a rapid current of air, as that produced by a bellows, or when masses of it are burned together; the diamond and graphite, though heated to ignition, do not continue burning in the air, but their combustion goes on in oxygen. A small diamond being fastened to the end of a pipe- stem, which is attached to a bent wire, it is strongly heated in the blowpipe, (best in the hydroxygen blowpipe,) and when well ignited is quickly plunged into a bottle filled with oxygen, where it continues to burn until it is entirely consumed. It can easily be proved by lime-water that carbonic acid is formed, as in the combustion of ordinary charcoal. Although carbon has a great affinity for oxygen, it is otherwise a very fixed body. These properties render it a very valuable agent in depriving almost all other substances of their oxygen, and it is hence almost exclusively used in metallurgy for the reduction of metallic oxides. CARBONIC ACID. 309 COMBINATIONS OF CARBON WITH OXYGEN. § 252. Carbon forms several compounds with oxygen, of which we shall notice only the three most important: 1. Carbonic acid C03 2. Carbonic oxide CO 3. Oxalic acid C303 The first two are gaseous at ordinary temperatures; and the third has not been isolated, being only known in combination with water or bases. Carbonic Acid, COa. § 253. When carbon burns freely in the air or in oxygen, it is converted into carbonic acid. But the simplest method of obtain- ing this gas in large quantities is to treat carbonate of lime, a mineral widely disseminated through nature, with a strong acid. Our ordinary limestone, chalk, marble, and the shells of shell-fish are essentially composed of carbonate of lime. Statuary marble is very pure carbonate of lime. To procure carbonic acid, pieces of marble, &c. are introduced into a bottle A, with two tubulures (fig. 268,)* a certain quantity of water poured over it, and the bottle shaken for a few moments, to expel the bubbles of air ad- hering to the marble. To one of the tubulures a, is fitted an exit tube, to collect the gas, and to the other b is adapted a larger tube, terminating in a funnel, and descending nearly l to the bottom of the bottle. Chlorohydric acid is poured through the tube b, and as soon as it reaches the marble, a lively effervescence ensues from the disengagement of carbonic acid gas. The reaction is represented by the following equation : Fie:. 268. Ca0,C03+HCl=CaCl+H0+C03. The result is therefore carbonic acid, which is disengaged in a gaseous form, and may be collected over water or mercury; chlo- ride of calcium, which dissolves in the water of the bottle; and, lastly, water, which remains mixed with that already contained in * Or into a wide-mouthed bottle with a cork pierced for the two tubes. —J. C. B. 310 CARBON. the bottle. In order to obtain pure carbonic acid, a considerable portion of the gas must be allowed to escape before collecting it, for it must expel the air contained in the upper part of the bottle, as well as that lodged in the interstices of the carbonate of lime. The gas is known to be pure when it is completely absorbed by a solution of potassa. The chlorohydric acid is added gradually, through the funnel, and only when the effervescence produced by the preceding portion begins to slacken. Sulphuric may be substituted for chlorohydric acid; when the reaction is represented by the following formula: Ca0,C02+S03=Ca0,S03+C03. In this case, carbonic acid and sulphate of lime are formed; and, the latter being only slightly soluble in water, the greater part of it is deposited in the form of minute crystalline scales, which eventually prevent the contact of the marble and sulphuric acid, and impede the reaction. This does not occur when chlorohydric acid is employed, because the chloride of calcium is eminently soluble in water, and leaves the pieces of marble freely exposed to the further action of the acid. § 253 bis. Carbonic acid is a colourless gas, nearly inodorous, having a slightly sourish taste; its density is greater than that of the air, being 1.529 at 32°, under a pressure of 0m.760 (29.92 in.) A litre of it, under the same circumstances, weighs lgm.977. (100 cub. in. at 32° and 29.92 Bar. weigh 50.03856 grs.) Carbonic acid gas liquefies at a temperature of 32°, under a pressure of 36 atmospheres. The pressure of 27 atmospheres will suffice at a temperature of 14°, and at that of —22°, which is easily obtained by a mixture of crystallized chloride of calcium and ice, a pressure of 18 atmospheres effects its liquefaction. When the temperature is greater than that of melting ice, a greater pressure is required, so that at the temperature of 86°, the pressure of 73 atmospheres becomes necessary. It forms a very unstable, colourless liquid, remarkable for its great dilatability, for its co- efficient of dilatation, which varies greatly with the temperature, is greater than that of atmospheric air, and the coefficient of the latter far surpasses that of all the liquids which we are required to examine at ordinary temperatures. The spec. grav. of liquid carbonic acid, compared with water at 32°, is 0.98 at and 0.72 at 80J°. The acid solidifies at about —94°, when it forms a perfectly transparent vitreous mass. Carbonic acid is eminently soluble in water, which dissolves about its own volume of gas at ordinary temperatures. Its solu- bility does not prevent our collecting it over water for ordinary experiments, but, in exact researches, it is better to collect it over mercury. The quantity of the gas dissolved by water at the same tern- 311 perature, increases with the pressure to which the gas is subjected. It has been observed that a volume of water dissolves its own volume of carbonic acid, whatever be the density of the gas; in other words, whatever the pressure to which it is subjected. Thus, a litre of water dissolves nearly a litre of carbonic acid gas, under the pressure of 1, 2, 3 ... 10 atmospheres; but, as the densities of the gas are, in this case, nearly as 1 : 2 : 3 : . . . : 10, the weight of carbonic acid dissolved will be in the same proportions of 1 : 2 : 3 : . . . : 10. A solution of carbonic acid reddens the tincture of litmus, like a feeble acid, producing only a purplish red. Carbonic acid does not support combustion, and immediately extinguishes a lighted taper plunged into it; nor does it support respiration, for an animal immersed in an atmosphere of it speedily pei’ishes from asphyxia. It does not, however, exert deleterious influence on the organs, for it may exist in considerable propor- tions in the air, without any inconvenience to animals, provided there be sufficient oxygen to maintain respiration. As carbonic acid is much heavier than the air, it may be poured, like a liquid, from one vessel into another in the open air, provided it be perfectly tranquil. Of two tubes, A and B (fig. 269), nearly equal, A is filled with carbonic acid over water, the opening closed under water with the hand, and the- tube taken out. An assistant handing the glass B, filled with air, the carbQnic acid in A is poured into it, as in the figure. It is easy to prove that it has passed from A to B, for a lighted taper continues to burn in A, and is extinguished in B.* Carbonic acid is formed under variety of circumstances; it is constantly produced by combustion in our fireplaces ; large quan- tities of it are given off in respiration; all organic substances ex- posed to a damp atmosphere are.destroyed by fermentation with a copious disengagement of the gas; lastly, burning volcanoes con- stantly project torrents of it into the air. It is also disengaged from many localities which present no igneous eruptions, but which have formerly been the seat of volcanic activity. In such regions, springs issuing from the earth contain carbonic acid in solution, and their waters effervesce on reaching the surface. They are called gaseous or carbonated mineral waters. Gaseous waters are made artificially, by saturating ordinary CARBONIC ACID. Fig. 269. * The best method of exhibiting its density, as a class experiment, is to fill a bell-glass with it by displacement (see fig. 223, § 167), and to pour its contents into another, filled with air, and haying a burning taper at the bottom. As the dx-y gas descends, the taper is extinguished.—J. C. B. 312 CARBON. water with the gas under strong pressure, and immediately trans- ferring it to well-corked jugs or bottles, to prevent the escape of the gas. If the water has been saturated under a pressure of 10 atmo- spheres, it contains a quantity of carbonic acid ten times greater than if the saturation had taken place under the pressure of a single atmosphere. A considerable portion of the dissolved gas will therefore escape when the gaseous water is poured into a glass. If the gaseous water be exposed to the air, it will ultimately part writh all its carbonic acid, and return to the state of ordinary water. This is a natural consequence of the law of the solution of gases in water—a law developed in § 81. We have seen that water dissolved nearly its own volume of carbonic acid gas, the dissolved gas having the same density as the carbonic acid gas of the atmo- sphere Avhich presses on this liquid. Now, when the solution is exposed to the air, the density of the carbonic acid which enters into the composition of the atmosphere is exceedingly small, and, as it were, null, so that the carbonic acid of the solution must be disengaged until it acquires an equal density, that is, until this disengagement is nearly complete. If the gaseous water be poured into a glass, bubbles of gas will be seen to start from its sides, and particularly from the bottom, if it be more rough; and if a very rough body, such as a piece of bread, be thrown into the liquid, a lively effervescence takes place around it. The following is the reason of this phenomenon: Each molecule of carbonic acid in solution is retained by the molecules of the surrounding water, which are uniformly arranged around the molecules of acid, in the interior of the liquid, or even at some distance from the sides. But, immediately in contact with the side, the molecule of acid is only retained in solution by the aque- ous molecules on one side, and, on the other, by the surface of the side of the glass. Now, it will be readily perceived that this side will retain the molecule of carbonic acid with much less force than the particles of water whose place it usurps. The molecules of carbonic acid in contact with the glass are therefore the first to assume the gaseous form. But, if a certain number of these mole- cules have united to form a small gaseous bubble, the latter, pass- ing through the liquid, will necessarily increase by the addition of other molecules of the gas which it meets on its way. For, if we suppose the bubble of gas to be arrested in any one of its posi- tions, it is evident that the molecules of dissolved carbonic acid which are immediately on the periphery of the bubble, being re- tained only by one-half of the particles of water which keep the molecules of acid dissolved in other parts of the liquid, will escape more rapidly than the latter. In localities where this gas is exhaled from fissures in the earth, it frequently accumulates in low spots, natural excavations, and CARBONIC ACID. 313 grottos, in which the air is not often changed; forming an invisible stratum of variable thickness on the surface of the ground, in which animals perish, if they remain for too long a period. The famous Grrotto del Cane, near Naples, presents a phenomenon of this character. Men may walk there free from danger, wdiile a dog, with his head nearer to the ground, soon falls asphyxiated. § 254. Liquid carbonic acid has been lately used to produce great degrees of cold, in order to liquefy and even solidify many gaseous substances. The apparatus used for this purpose consists of two parts: 1st. The generator, in which the liquid acid is produced. 2d. The receiver, into which it is transferred by distillation, so as to separate it from the other products of the reaction, and in which the products of several successive operations may be accu- mulated. The liquid acid is produced by decomposing bicarbonate of soda by sulphuric acid in the generator. The first portions of acid disengaged assume the gaseous form, but the pressure soon be- comes sufficient to liquefy it. The generator is a vessel closed air-tight, and was formerly made of very thick cast-iron, but the danger of employing cast- iron where great powers of resistance are required, and the occur- rence of a terrible accident from an explosion, have proscribed its use. As now made, it consists of a cylindrical vessel of lead, (fig. 270), covered with copper, and strengthened by rings and bars of wrought iron, and contains 6 or 7 litres (l|-2 gallons). The Fig. 270. Fig. 271. 314 CARBON. copper cylinder surrounding the vessel is made to fit it exactly, and the ends are further strengthened by iron plates, fastened together by bars of the same metal. The generator is suspended between the two points/,/', by a cast-iron stand. The construction of the receiver (fig. 271) is similar to that of the generator. The aperture 0 of the generator is closed by a screw h', having a hole through its axis, and furnished with a stopcock r. The screw is worked with a double handle mn, and a leaden ring, compressed in a double collar, renders the closure perfect. There is an aperture i on the upper edge of the receiver B, to which is adapted a copper tube, descending nearly to the bottom of the receiver, with a stopcock r' on the outside. The receiver and generator can be made to communicate, by means of a fixed copper tube stx. To prepare the liquid acid, the stopper k is removed, and there are introduced into the generator 1800 grammes (about 4 lbs.) of bicarbonate of soda, 4J litres (qts.) of water at 95° to 104°, and a cvlindrical copper vessel uv (fig. 272) containing 1000*“ (2£ lbs.) of oil of vitriol. This cylinder falls into the axis of the generator, and, while the latter remains vertical, the sul- phuric acid cannot come into contact with the bicarbonate of soda. The stopper k being then fixed, and the cock r closed, by inclining the generator below the horizontal line, the acid contained in the copper tube is poured out, and the reaction immediately commences. The generator is made to revolve several times around its axis to mix the sub- stances together, and in about ten minutes, the carbonic acid may be passed into the receiver. A connection be- tween the generator and receiver is effected by the tube stx, and the cocks r' and r opened, when the carbonic acid in the gene- rator distils over immediately, and recondenses in the liquid form in the receiver. The distillation occurs by virtue of the difference in temperature between the generator and the receiver. That of the former not being lower than 86°, with a tension of the acid in it equal to about 75 atmospheres. If the temperature of the receiver be 59°, which may be assumed to be that of the laboratory, the maximum tension of the acid being, for that tem- perature, only 50 atmospheres, distillation must take place by virtue of the difference of pressure of 75—50=25 atmospheres; that is, it will be extremely rapid; indeed, less than a minute is required to allow the carbonic acid in the generator to pass into the receiver. The same operation is repeated with 5 or 6 additional quantities of carbonic acid, so as to accumulate about 2 litres (2 qts.) of liquid acid in the receiver. The cock r' being closed, the generator and receiver are dis- Fig. 272. CARBONIC ACID. 315 connected. The latter is then two-thirds full of liquid carbonic acid, surmounted by a gaseous atmosphere exerting a pressure of 50 atmospheres, if the temperature of the laboratory be 59°. It follows, therefore, that if we open the stopcock r' of the receiver, the liquid acid will be forcibly expelled from the receiver, and, if projected into the open air, will immediately assume the gaseous form, appearing like a white cloud. A considerable degree of cold necessarily exists in the gaseous current, and, if the jet be directed into a bottle, or, better still, into a very thin metallic box, a large portion of the carbonic acid will volatilize, by abstracting the heat necessary for its change of condition from the sides of the vessel, and from that portion of the carbonic acid which re- mained liquid. The temperature will fall below —94; and the remaining acid will condense to a solid, in the form of a white cotton-like snow. Carbonic acid may be preserved in the snowy form much longer than in the liquid state; evaporation being very slow, on account of its bad conducting qualities, and an air thermometer, surrounded by the snow evaporating freely in the air, falls to —110. A flake of snowy carbonic acid may be held on the hand without imparting a feeling of intense cold, because it is constantly kept from im- mediate contact by a current of gaseous acid; but, if the flake be pressed between the fingers, great pain is felt, like that produced by a heated body, and the skin is disorganized as by a burn. Figs. 273 and 274 represent the metallic box generally used to collect solid carbonic acid, and is composed of two parts abed and a'b'e'd' (fig. 273), which may be readily sepa- rated and joined together. The part abed has a tubulure t, in which is inserted the small tube u (fig. 275), pre- viously fastened to the piece x of the receiver (fig. 271). By opening the cock r\ a jet of the liquid acid enters the box almost at a tangent to its periphery, and then meet- ing a tongue m, arranged to produce a gyratory movement, a por- tion of the liquid acid is converted into gas, which, after having passed around the box, escapes through the central tubes cd and c'd', while the remainder is solidified in the form of the white snow, which is removed by opening the box. The tubes cd and c'd' are surrounded by two concentric tubes covered with cloth, so that they can be held in the hand without imparting too great a degree of cold. Fig. 274. Fig. 273. Fig. 275. 316 CARBON. If upon the solid acid a liquid be poured which does not combine chemically with it, and does not congeal at a very low temperature, the acid evaporates more rapidly, because the interposed liquid considerably increases its conductibility, and a very powerful re- frigerating mixture is obtained, which rapidly cools bodies im- mersed in it, without, however, lowering their temperature much more than solid carbonic acid alone. If such a mixture be placed under the receiver of the air-pump, and the evaporation be accele- rated by a vacuum, the temperature falls to —150°. Ether is generally mixed with the snowTy acid, and, by means of this frigorific paste, 1 kilogramme (2 lbs.) of mercury can be frozen in a few moments ; and, if an hermetically sealed tube, con- taining the liquid acid, be immersed in it, it congeals to a perfectly transparent vitreous mass. § 255. The composition of carbonic acid can be readily deter- mined, approximately, by the following experiment:—A flask holding about 1 litre (1 qt.) is filled with oxygen, over the mercu- rial trough, and placed in the inverted position represented in fig. 276. A small piece of char- coal fastened to the end of stout platinum wire, being introduced into the flask, is ignited by concentrating the solar rays by a powerful lens, and is converted into carbonic acid. When the combustion is finished, and the gas is allowed to recover its original temperature, it will be found that its volume has not sensibly changed. We hence conclude that carbonic acid gas contains a volume of oxygen equal to itself. Now, 1vol. of carbonic acid gas weighs 1.5290 1 “ “ oxygen “ 1.1056 The weight 1.5290 of carbonic acid contains, therefore, a weight 1.1056 of oxygen, and a weight 0.4234 of carbon; which gives, for the composition of carbonic acid, Carbon 27.68 Oxygen 72.32 100.00 But this determination is only an approximation; and it may be ascertained with precision by the following experiment: A given weight p of pure carbon, as the diamond, contained in a small platinum dish, is introduced into a porcelain tube ab (fig. 277), arranged in a reverberatory furnace. One end of the tube a is connected with an apparatus which furnishes perfectly dry oxygen gas, and the other end with a series of tubes, as represented in the figure. The tube A is a U-tube, containing pumice-stone impregnated with oil of vitriol. Fig. 276. CARBONIC ACID. 317 Fig. 277. The bulbed apparatus B contains a concentrated solution of caustic potassa. The tube C is filled with large pieces of pumice-stone impreg- nated with a concentrated solution of caustic potassa. Lastly, the tube D is filled with large pieces of pumice-stone impregnated with oil of vitriol. Let P be the exact weight of all the tubes B, C, D. The appa- ratus is arranged by joining the various tubes together by caout- chouc tubes. The apparatus being filled with oxygen gas, slowly passed through it, the tube ab, containing the carbon, is heated to redness, and the gases traverse the series A, B, C, D. The tube A condenses the small quantity of hygroscopic moisture which may be furnished by the interior of the tube ab. The carbonic acid formed is almost wholly condensed in the bulbs B ; but if it be too rapidly generated, (an occurrence that cannot always be avoided,) a portion of it, which might escape from B without being condensed, is arrested by the tube C. The gases passing through the bulbs B and the tube C being perfectly dry, and the solution of caustic potassa contained in the apparatus not being sufficiently concentrated to render its tension of vapour insensible, they have a tendency to abstract from this solution a certain quantity of aqueous vapour, which will diminish the weight of the apparatus to that extent. The last tube, D, remedies this defect, by restoring to the gases their condition of absolute dryness which they possessed before passing into the air. In this combustion of carbon, the formation of a small quantity of oxide of carbon might be feared, which would render the analy- sis inaccurate; and, to avoid such an error, the anterior part of the tube ab is filled with very porous oxide of copper, which is heated to redness during the experiment. The gaseous mixture being obliged to traverse the oxide before reaching the apparatus in which absorption takes place, the small quantities of gaseous oxide of carbon which might be present are necessarily converted into carbonic acid. A plug of asbestos is used to separate that 318 CARBON. portion of the tube containing oxide of copper from that which contains the small dish holding the carbon. When the combustion of the carbon is completed, the disengage- ment of the oxygen is continued for some time, in order to be sure that all the carbonic acid produced has passed through the absorb- ing apparatus. The apparatus being taken apart, the first step is to ascertain whether the carbon in the dish is completely destroyed. Most frequently, a small residue of incombustible earthy matter is found, which was mechanically mixed with the carbon. The residue, which cannot exceed a few milligrammes, is weighed, and its weight it subtracted from the weight p, in order to obtain the exact weight (p—'*) carb°n which has been burned. Upon finding the weight Pr of the apparatus B, C, D, it is evi- dent that (P'—P) will represent the weight of carbonic acid pro- duced. It has therefore been ascertained, that a weight (p—*) of carbon produces a weight (P'—P) of carbonic acid. The apparatus B, C, D should be weighed with special precau- tion, where precision is required. It displaces a considerable volume of air, and, in order to obtain the absolute weights P, P', before and after the experiment, the weight of air it displaces un- der both circumstances must be added. If this air were exactly in the same condition, at the time of both weighings, the additions would be unnecessary, because, being nearly equal in both cases, they would destroy each other in the difference (P'—P). But we can never be sure of establishing this identity of conditions, and it is better to guard against this source of error by the following arrangement, which we have already mentioned (§ 97) for weighing accurately a flask filled with gas. The dishes of the balance used in the experiment should be fur- nished with hooks beneath, to one of which the apparatus B, C, D, is attached, by a long piece of wire, to keep it at a great distance from the point of suspension of the dishes to the beam. To the hook of the second dish, and at the same distance, is fastened a sys- tem of tubes B', C', D', as similar as possible to the tubes B, C, D, and charged in the same manner. The system B', C', D' should nearly balance the system B, C, D, as weighed before the experiment; and then perfect equilibrium is established by additional weights. At the second weighing, the system B, C, D has increased by the weight of carbonic acid which it has absorbed which weight will be immediately given by the weights necessary to restore the equilibrium, under the same conditions in which the first weighing was made. As the two systems B, C, D and B', C', D' displace nearly the same volume of air, it is clear that the result of the weighings, made as directed, will be nearly independent of the small varia- tions which the constitution of the air might undergo between the times of weighing. CARBONIC OXIDE. 319 It is thus found that carbonic acid contains 1 eq. carbon 6.0 27.27 2 “ oxygen 16.0 72.73*’ 1 “ carbonic acid 22.0 100.00 If the number 72.73 be divided by the density 1.1056 of oxygen, and the number 100 by the density 1.5290 of carbonic acid gas, the quotients 65.7 and 65.4, which are nearly equal, lead to the conclusion that carbonic acid gas contains a volume of oxygen precisely equal to its own. The difference between 65.7 and 65.4 is owing to the fact that carbonic acid gas departs remark- ably from Mariotte’s law, even under the ordinary pressure of the atmosphere. These quotients would be much more nearly equal, if, instead of dividing the numbers 72.73 and 100 by the respective densities of oxygen and carbonic acid under the pressure of 0m.760, (29.92 in.) we were to divide them by the densities of these same gases under less pressure, as that of 0m.100 (3.937 inches). Carbonic Oxide, CO. § 256. Carbonic oxide gas is obtained, by slowly passing a cur- rent of carbonic acid through a long porcelain or strong glass tube, containing charcoal, and heated to redness. The carbonic acid combines with a quantity of carbon equal to that which it already contains. It is, however, easier to heat finely pulverized carbonate of lime intimately mixed with charcoal, in a stoneware retort, in a re- verberatory furnace. Carbonate of lime alone is decomposed at a red-heat, giving off its carbonic acid; but the gas at this tem- perature meets with carbon, and is converted into oxide of carbon. It is necessary to agitate the gas collected in the bell-glasses for a few moments with a solution of caustic potassa, in order to absorb the small proportion of carbonic acid which may have escaped decomposition. But carbonic oxide gas may be still more readily obtained, by decomposing with oil of vitriol oxalic acid, which is the third combination of carbon with oxygen. The formula of crystallized oxalic acid is C203+3H0, and it will easily part with 2 equiva- lents of water without being decomposed, but the third cannot be abstracted without decomposing it into carbonic acid and oxide; for we have C303=C03+C0. The decomposition takes place when the crystallized acid is heated with a substance which powerfully attracts its water, such as an excess of oil of vitriol. The oxalic acid is introduced into a small flask, and 5 or 6 times its weight of oil of vitriol added. To the flask is fitted an exit tube which carries the gas into a bell-glass over water or mercury. 320 CARBON. When first heated, the oxalic acid dissolves in the sulphuric; but an effervescence soon ensues, arising from the decomposition of the oxalic acid into its two gaseous products, carbonic acid and oxide, which are evolved in equal volumes. To the mixed gases collected in a bell-glass are introduced a few cubic centimetres (1 or 2 fl. dr.), of a solution of potassa, which absorbs the carbonic acid, and leaves the carbonic oxide pure. The gaseous mixture, as it is disengaged, may also be passed through a washing-bottle (fig. 278) containing caustic potassa, and the small quantity of carbonic acid which escapes absorption in the bottle may be ab- sorbed in the bell-glass. Carbonic oxide gas is colourless, inodorous, and has not yet been liquefied. It burns in the air with a characteristic bluish flame, and is then converted into carbonic acid. Its specific gravity is 0.967. Water dissolves only about one-sixteenth of its volume of the gas. It does not act upon litmus, nor combine with the acids or bases. Whenever charcoal is burned in our furnaces, and not supplied with a sufficient quantity of oxygen, a large proportion of carbonic oxide is formed. It thus happens, when a laboratory furnace is filled with ignited coals piled up so as to form a burning heap of several decimetres (a foot or more) in height. The lower strata are at first converted into carbonic acid, from the oxygen of the air passing through the lower bars of the grate, and here the temperature is highest. In the upper strata, combustion being supported only by the highly heated gaseous current which has traversed the lower ones, carbonic acid is converted into oxide of carbon, and the temperature is much lower. Lastly, when the gaseous mixture again comes in contact with the air in the upper part of the furnace, if the temperature is sufficiently high, carbonic oxide is ignited, and burns with a blue flame. In wind and blast furnaces, employed in metallurgy, and often of considerable elevation, combustion obeys the same laws; but, as the fuel and minerals are thrown in cold at the top of the fur- nace, the temperature in that part is always low, and the combus- Fig. 278. CHLOROXICARBONIC GAS. 321 tion of the oxide of carbon only ensues if designedly inflamed; when it continues indefinitely. Oxide of carbon is not merely passive in not supporting respira- tion, but is active as a poison ; for an animal perishes if left for some time in an atmosphere containing a few per cent, of this gas. To its presence must be attributed the uneasiness and head- ache experienced by remaining in a badly ventilated apartment, near a furnace containing burning charcoal, the products of which do not immediately pass up the chimney. If a large proportion of carbonic oxide gas be present in a closely shut room, death ensues from asphyxia. § 257. Carbonic oxide is readily analyzed by burning it with oxygen in the eudiometer. Let us suppose that we have introduced into the eudiometer 100 volumes of carbonic oxide 75 “ of oxygen, 175 By passing an electric spark through it, the bulk of gas, after the explosion, is reduced to 125 volumes. If a little potassa be now introduced into the eudiometer and shaken in it, the carbonic acid produced is absorbed, so that the remaining gas will only measure 25 volumes, and prove to be pure oxygen. The bulk of carbonic acid gas produced contains, therefore, 100 volumes ; that is, it is equal to that of the carbonic oxide operated on, and the volume of oxygen consumed is 75—25=50. One volume of carbonic oxide therefore consumes a J volume of oxygen, and produces 1 volume of carbonic acid. Now, 1 volume of carbonic acid gas contains 1 volume of oxygen; and, conse- quently, 1 volume of carbonic oxide gas only contains a \ volume. If, therefore, we subtract from the density of Carbonic oxide 0.9674 the density of oxygen 0.5528 We have 0.4146 which is the weight of carbon combined with a weight 0.5528 of oxygen, to form a weight 0.9674 of oxide of carbon. Oxide of carbon is therefore composed of 1 eq. of carbon 6.0 42.86 1 “ oxygen 8.0 57.14 1 “ oxide of carbon 14.0 100.00 Chloroxicarbonic Gas, CO,Cl. § 258. Chlorine and carbonic oxide combine under the influence of solar light; and, in order to obtain the compound, a dry flask 322 CARBON. is exhausted of air as completely as possible, and filled with dry carbonic oxide until the pressure of the gas is equal to one-half the pressure of the atmosphere. The flask being closed, chlorine gas is introduced until the internal pressure is exactly equal to that of the atmosphere, so that the flask contains equal volumes of chlorine and carbonic oxide. The gases are merely mixed if the chlorine has been introduced in a room illuminated only by diffused light, or, better still, by the light of a candle ; but, if the flask be exposed to the direct rays of the sun, combination imme- diately ensues, and the greenish colour of chlorine entirely dis- appears. Combination will also ensue in the diffused light of day, but after a greater lapse of time. Under all circumstances, how- ever, after combination has been effected, if the flask be made to communicate with the manometer which measured the internal pressure, the latter will be found to be only one-half of that of the atmosphere. We hence conclude that 1 volume of chlorine has combined with 1 volume of carbonic oxide to form 1 volume of the new gas, which is called chloroxicarbonic, (also chlorocarbonic acid, phosgen gas.) The density of the gas is obtained by adding to the density of chlorine 2.440 “ “ carbonic oxide 0.967 Density of chloroxicarbonic gas 3.407 and its formula is COC1. It may be regarded as carbonic acid C02, or C0,0, in which one of the equivalents of oxygen is re- placed by an equivalent of chlorine. Chloroxicarbonic gas is colourless, and has a peculiar, suffocating odour. It is decomposed by contact with water, an equivalent of each body producing chlorohydric and carbonic acids. Thus, C0,C1+H0=C0SHC1. Oxalic Acid, C303.. § 259. Oxalic acid exists in a great number of vegetables. It is prepared artificially by boiling sugar with slightly dilute nitric acid, which, by yielding a portion of its oxygen, evolves deutoxide of nitrogen and carbonic acid, while oxalic acid remains in the liquid, from which it crystallizes on cooling. Six parts of nitric acid, of the density 1.2, are employed for 1 part of sugar, and about \ oxalic acid is obtained. The crystals deposited in the liquid always retain some nitric acid, from which they are purified by redissolving in boiling water, and again crystallizing them. Nine parts of water, at the ordinary temperature, are required to dissolve 1 part of oxalic acid; but a much smaller quantity of boiling water will suffice. The formula of the crystallized acid is C203+3H0. If it be OXALIC ACID. 323 heated to 212° in a current of dry air, or if exposed for a long time to a dry atmosphere, it loses about 28 per cent, of its weight, corresponding to 2 equivalents of water. But the last equivalent of water cannot be abstracted except by combining the acid with a base. The endeavour to deprive it in any other way of the last equivalent of water decomposes it into carbonic acid and carbonic oxide. Advantage was taken of this reaction to obtain carbonic oxide. Oxalic is a powerful acid, combining with bases, and producing well-defined salts. It readily expels carbonic acid from all its compounds. § 260. Oxalic acid is analyzed in the following manner:—Let us first suppose that it is required to analyze crystallized acid con- taining 3 equivalents of water, according to the formula C303-f 3IIO. One gramme of the acid reduced to powder is accurately weighed, and mixed with 20 or 30 times its weight of recently calcined and perfectly dry oxide of copper, and the mixture intro- duced into a strong glass tube, 5 or 6 decimetres (18-22 in.) in length, open at one end a, and drawn out to a fine point at the other end b. Pure oxide of copper being poured upon the mixture, so as to fill the tube to within 3 or 4 centimetres (1J-1J in.) of its opening a, the tube is arranged in a sheet-iron furnace, made as represented in fig. 279. The series of tubes A, B, C, arranged as Fig. 279. directed for the analysis of carbonic acid (§ 255), are connected with the tube in the furnace by a cock. Lastly, the end of the tube C is connected with an aspirator furnished with a tube con- taining sulphuric pumice (not represented in the figure), the object of which is to prevent the entrance of moisture into the tube C, from the external air. The tube A having been weighed alone, let P be that weight. The tubes B and C having also been weighed together, let P' be their joint weight. When the apparatus is arranged, that portion of the tube ab containing oxide of copper alone is first heated to redness, and when it appears red for the length of 1 or 2 decimetres (4-7 in.), live coals are carefully approximated to that part of the tube containing the mixture of oxide of copper and oxalic acid. The decomposition of this acid soon commences, and the oxide of cop- 324 CARBON. per yielding sufficient oxygen to burn the carbon into carbonic acid, water becomes free, and the mixture of carbonic acid gas and aqueous vapour passes successively through the tubes A, B, C. The tube A retains all the vapour of water, while the carbonic acid is dissolved in the bulbs B and the tube C. The operation is con- tinued until the fire entirely covers the tube, when the combustion of the oxalic acid is terminated. The evolution of gas ceases, and as the absorbing action of the solution of potassa continues in the bulbed apparatus B, the pressure in the interior becomes less than that of the atmosphere, and the solution in the bulbs ascends toward the tube A. It might even be projected into this tube, unless the precaution were taken to give the bulbs the position represented in fig. 281, instead of that in fig. 280, which last it maintains during the combustion. There is then nothing to fear from absorption; for the potassa can only half fill the bulb e, and, if the rarefaction of the interior gas continues, atmospheric air enters the apparatus by the tube C, traversing the bulbs B in the form of bubbles. The carbonic acid and water arising from the combustion of the acid are not, however, entirely absorbed, for a portion of them remains in the combustion tube, and must also be passed through the absorbing tubes. In order to effect this, the coals surrounding the end b of the combustion tube are withdrawn, and when this part is cooled, the fine point b is broken, and a tube immediately fitted to it, by a caoutchouc connecter, containing pieces of caustic potassa. Water being run out of the aspirator at the same time, the external air is drawn into the apparatus, being first deprived of its moisture and its small content of carbonic acid, by traversing the tube containing potassa, just appended to the apparatus. As it traverses the combustion tube and the absorbing tubes A, B, C, it deposits in the latter the water and carbonic acid still remaining in the combustion tube. When 1 litre (1 qt.) of water has been drawn off, we may be certain that all the products of the combustion of oxalic acid have been concentrated in the absorbing tubes. The escape of water is stopped, the tubes detached and separately weighed. Let Q be the weight of the tube A, containing sulphuric acid, which has condensed the water, and Q' the joint weight of the tubes B and C which have condensed the carbonic acid. It is evident that the water produced by the combustion of lgm of oxalic acid weighs (Q—P), and the carbonic acid from the same (Q'-P'). Fig. 280. Fig. 281. OXALIC ACID. 325 If the experiment has been accurately performed, then will (Q—P )=0gm.249 Now 0gm.429 of water contain 0gm.0476 of hydrogen, while 0gm.698 of carbonic acid contain 0gm.1905 of carbon. Now, since oxalic acid contains only carbon, hydrogen, and oxygen, the com- position of lgm is (Q,-P/)=0gm.698. Hydrogen 0.0476 Carbon 0.1905 Oxygen 0.7619 1.0000 and, consequently, of 100 parts, Hydrogen 4.76 Carbon 19.05 Oxygen 76.19 100.00 In order to ascertain the ratio of the equivalents of the three elements in oxalic acid, it is only necessary to divide the propor- tional weight of each by its chemical equivalent; which gives H=nr=4.760 C=^=3.!75 0=^r=9.524 These numbers being to each other as 3 : 2 : 6, the formula of the crystallized acid is either C3H306, or its multiple. Now, the formula C3H306 gives 3 eq. of hydrogen 3.0 2 “ carbon 12.0 6 “ oxygen 48.0 1 “ crystallized oxalic acid 63.0 § 261. Having observed that oxalic acid, heated to 212° in dry air, lost a certain quantity of water, it is necessary to ascertain this proportion exactly by experiment. An accurately weighed quantity of oxalic acid is introduced into a glass tube having the form represented in fig. 282. The tube is first weighed empty, the pulverized acid poured in, taking care that none remains in the vertical leg ab, and again weighed. The increase in weight represents exactly the quantity of matter intro- duced, which suppose to be = lgm.000. The ap- paratus abed is connected by the end d with an aspirator, filled with water (fig. 283), and, by the end a, with a U-tube filled with Fig. 282. 326 CARBON. Fig. 283. sulphuric pumice-stone. The apparatus abed is placed in a vessel of boiling water, if the substance is to be heated to 212°; or in a saturated solution of salt, if a temperature of 230° is required; or, again, in an oil-bath, if the heat is to rise to 390°. A mercu- rial thermometer indicates the temperature, which may be main- tained nearly uniform by a proper management of the fire. In the present case, the temperature of boiling water is sufficient. By drawing off water from the aspirator, the external air tra- verses the apparatus, being first dried in the tube A, containing sulphuric acid, and then passing over the heated matter, which loses its water. When the aspirator is emptied, the tube abed is accurately weighed again, and the difference between the two weights of the tube shows the quantity of water lost. But it is necessary to ascertain whether the substance, if subjected for a longer time to the heat of 212°, might not lose an additional quan- tity of water; and, in order to determine this, the tube abed is replaced in the apparatus, the aspirator again filled and drawn off, and the tube abed weighed once more. If the same weight be found as before, it is a proof that the substance had parted with all the water it could lose at 212°; but if the weight be less, the heated substance must be again subjected to a current of dry air, until consecutive weighings evince no discrepancy. By operating on lgm of the crystallized acid, the loss of weight amounts to 0.826gm, corresponding to 2 equivalents of water. We have, in fact, 1 eq. hydrogen 1.0 2 “ carbon 12.0 4 “ oxygen 32.0 1 “ desiccated oxalic acid.... 45.0 71.43 2 “ “ water 18.0 28.57 1 “ crystallized oxalic acid... 63.0 100.00 The formula of the desiccated acid is Ca04II, which may be EQUIVALENTS. 327 written C303,II0; for 1 equivalent of water may still be elimi- nated, if replaced by 1 equivalent of base. If nitrate of lead be poured into a soluble oxalate, as the neutral oxalate of potassa, a white precipitate of oxalate of lead is formed, the formula of which is Pb0,C303, as may be demonstrated by a direct analysis of the salt. The proportion of oxide of lead is first determined by weighing accurately a certain quantity of oxalate of lead in a platinum cru- cible, and heating it with an alcohol lamp, when the oxalate is decomposed, and oxide of lead remains.* One gramme of oxalate of lead gives 0.742gm of oxide of lead. of the oxalate is also burned in a tube with oxide of copper, like the crystallized acid (§ 260). Water is not obtained, but only 0.315 of carbonic acid, representing 0.086 of carbon. Oxalate of lead is therefore composed of Carbon 0.086 Oxygen 0.172 Oxide of lead 0.742 1.000 Whence we deduce the following composition: 2 eq. carbon 12.0 8.60 3 “ oxygen 24.0 17.19 1 “ oxide of lead 111.5 74.21 1 “ oxalate of lead 147.5 100.00 The above experiments therefore show that the formula of oxalic acid in the salts is C303; that of the crystallized acid from an aqueous solution is Ca03+3H0 ; and, lastly, that of the desic- cated acid is C203+H0. RECAPITULATION of the compounds of carbon and oxygen. §262. The three compounds of carbon with oxygen, which we have studied, are composed as follows: Carbonic oxide Carbon 42.86 Oxygen 57.14 100.00 Determination of the Equivalent of Carbon. * A porcelain crucible or piece of glass tube is preferable, as platinum would be very liable to injury from the reduction of the oxide. The decomposition may, however, be effected safely in platinum, at a low temperature, and with free access of air, which is also necessary to prevent the formation of suboxide.— J. C. B. 328 CARBON, Carbonic acid Carbon 27.27 Oxygen 72.78 100.00 Oxalic acid Carbon 33.83 Oxygen 66.67 100.00 By calculating their composition with reference to the same quantity, 100, of carbon, we have Carbonic oxide Carbon 100.0 Oxygen 133.3 233.3 Carbonic acid Carbon 100.0 Oxygen 266.7 366.7 Oxalic acid Carbon 100.0 Oxygen 200.0 300.0 The proportions of oxygen combined with the same quantity of carbon are, therefore, as 1 : 2 : |. The most simple formulae which can be assigned to the com- pounds are Carbonic oxide CO equivalent=14.0 Carbonic acid C03 “ =22.0 Oxalic acid CO§ “ =18.0 The oxalic is a powerful acid, completely saturating bases, and affording neutral salts, which can be obtained in an anhydrous state. Their analysis has shown that an equivalent of a base (for ex- ample, the weight 111.5 of oxide of lead) combines with 36 of oxalic acid, so that the number 36 represents its true equi- valent. Now, as this number is precisely double of that obtained when the formula COf is given to oxalic acid, its true formula is CS03- Carbonic is likewise an acid, but a feeble one, incapable of neutralizing bases completely. Moreover, with powerful bases, such as potassa and soda, it forms several carbonates; so that it is doubtful which one should be selected as the neutral salt. But, with less powerful bases, as baryta, strontia, lime, and the metallic oxides, it forms but a single series of carbonates; and these are generally regarded by chemists as the neutral carbonates. The analysis of any one of these proves that an equivalent of the base is combined with a weight 22 of carbonic acid. The number EQUIVALENTS. 329 22 is therefore its equivalent, and its formula is, consequently, co3. As carbonic oxide is an indifferent compound, whose reactions are not well defined, its formula is undetermined; and, although we write it CO, we may, on almost equally good grounds, write it c3o, The formulte of the compounds of oxygen and carbon being fixed, it is evident that the equivalent of carbon can be immedi- ately deduced from them, from any one of these three proportions : 57.14:42.86:: 8: x] 72.73 : 27.27 :: 16 : z 1 66.67 : 33.33 :: 24: 2 x J whence, a;=6.0 § 263. It has been shown that 1 volume of carbonic oxide con- tains a J volume of oxygen, and that 1 volume of carbonic acid gas contains 1 volume of oxygen. But we cannot say what is the volume of gaseous carbon or vapour of carbon found in 1 volume of carbonic oxide, or of carbonic acid, as carbon has not yet been vaporized. It is, however, conceivable that its vapor- ization is possible, at higher temperatures than those hitherto produced. If the laws laid down (§ 121) were perfectly demonstrated, it is evident that it would be most frequently possible, the volume of a gaseous binary compound being known, as well as the gaseous volume of one of its elements, to determine, by means of these laws, the gaseous volume of the second element, without finding it directly by experiment, or even without knowing the density of its vapour. This case will particularly occur when the two com- ponents form several gaseous compounds. Let us admit that these laws are exact, and apply them to the composition of carbonic oxide and acid. One volume of carbonic oxide containing a \ volume of oxygen, it should contain, from the laws laid down, either a -J volume of vapour of carbon without condensation, or else, 1 volume of vapour of carbon, condensed to a \ volume; that is, that a \ volume of oxygen, by combining with 1 volume of vapour of carbon, should form 1 volume of carbonic oxide. One volume of carbonic acid gas containing 1 volume of oxygen should contain a \ volume of vapour of carbon; making the con- densation in this case also equal to a J volume. But, if 1 volume of carbonic acid gas, with the densi- ty of 1.5290 contain 1 volume of oxygen with the density 1.1056 we have for the weight of a \ vol. of vapour of carbon.... 0.4234 And for the density of one volume of this vapour, 0.8468. It is evident, that only the first of the two modes of composition 330 CARBON. assumed for carbonic oxide gas is possible, for it is the only one which, with the density of the vapour of carbon just deduced from the composition of carbonic acid, will give the density 0.967 for carbonic oxide gas. Thus, J vol. of vapour of carbon 0.4234 J “ oxygen 0.5528 0.9762 The density 0.8468 for the vapour of carbon can only be con- sidered as an approximative value, because it has been deduced from the density of carbonic acid gas, which, at ordinary tem- perature and pressure, is too great. A more exact value is ob- tained, by starting from the composition which synthetic analysis, founded on weight, has given for carbonic acid, and admitting only the observed density of oxygen gas. It is given by the pro- portion 72.73 : 27.27 :: 1.1056 : J Since the atomic theory admits that carbonic acid is composed of 1 atom of carbon and 2 of oxygen, the atomic formulae of the compounds of carbon and oxygen are the same as their formulae in equivalents, and the atomic weight of carbon is 6.0. whence x — 0.8290. COMBINATIONS OF CARBON WITH HYDROGEN. § 264. The compounds of carbon and hydrogen are very nume- rous. Tavo of them are gaseous at ordinary temperatures, the others liquid or solid. Several of them Avill be described Avhen treating of organic bodies, and our attention Avill now be confined to the principal properties of the tAvo gaseous combinations. Protocarburetted Hydrogen, CsH4. § 265. It is also called light carburetted hydrogen, m distinction from the next compound, and marsh-gas, because it is evolved in large quantities from the waters of stagnant pools. When the muddy bottom of such waters is stirred with a stick, bubbles of gas are observed to arise, which are easily collected in an inverted bottle, filled with water (fig. 284), and a funnel inserted in its mouth. The gas thus obtained is impure, from the admixture of nitrogen and carbonic acid. Pure protocarburetted hydrogen is ob- tained, by heating a mixture of acetate Fig. 281. BICARBURETTED HYDROGEN. 331 of soda and an energetic base, such as caustic potassa or lime, in a glass retort or flask. A mixture of the two bases is generally preferred, and is made by dissolving the potassa in a small quantity of water, and adding powdered lime, so as to form a paste. We will explain hereafter the reaction which, in this experiment, pro- duces the protocarburetted hydrogen. Protocarburetted hydrogen is a colourless, inodorous gas, of the density 0.55^0; burning in the air with a bluish flame, and pro- ducing water and carbonic acid: water dissolves but a very small quantity of it. This gas is abundantly produced in certain mines, and being lighter than the air, it accumulates in the upper part of the shafts, and causes terrific explosions, attended with a great loss of life. Hence, miners call it the fire-damp. It is analyzed by the eudiometer, into which suppose we have introduced 100 volumes of protocarburetted hydrogen and 300 of oxygen. After the passage of the electric spark, the gaseous volume will be reduced to 200, and if a piece of moist potassa be passed into the mixture, the carbonic acid produced by the com- bustion will be absorbed, leaving 100 volumes of oxygen. The 100 volumes of carbonic acid contain 50 of vapour of carbon and 100 of oxygen, and therefore 100 volumes of oxygen have disap- peared, by forming water with the hydrogen of the gas. The latter gas containing 200 of hydrogen, 100 volumes of the carbo- hydrogen are composed of 200 of hydrogen, 50 of vapour of carbon ; which proportion is confirmed by the value of the density of the gas: 2 vol. of hydrogen weigh 0.1382 25.00 J “ vapour of carbon 0.4145 75.00 0.5527 .100.00 The formula of protocarburetted hydrogen is C2H4. Bicarburetted Hydrogen, C4H4. § 266. It is frequently called olefiant gas, and heavy carburetted hydrogen, and is prepared by heating together 1 part, by weight, of alcohol, and 5 or 6 parts of oil of vitriol. The reaction is too complicated to be explained at present; the gaseous products being bicarburetted hydrogen, and carbonic and sulphurous acids. The mixture of sulphuric acid and alcohol is introduced into a capacious glass retort (fig. 285), because it puffs greatly toward the close of the operation, and the gases are passed first through a washing- bottle containing water, and then through a second bottle con- 332 CARBON. Fig. 285. taining a solution of potassa, to absorb the carbonic and sulphurous acids. Bicarburetted hydrogen is a colourless gas, of the density 0.9784; burns in the air with a brilliant flame; is partially decomposed when passed through a porcelain tube heated to redness, carbon being deposited on the sides of the tube. It burns in chlorine, its hydrogen forming chlorohydric acid, and its carbon being deposited. Bicarburetted hydrogen and chlorine also combine in the cold, when the two gases are mixed over water, an oily volatile liquid being formed, of an agreeable, ethereal odour. Its analysis is conducted in the same manner as that of the light carburetted hydrogen. Bicarburetted hydrogen 100 Oxygen 400 being introduced into the eudiometer, 300 remain after the passage of the spark, of which caustic potassa absorbs 200, which is car- bonic acid, containing 100 of vapour of carbon and 200 of oxygen, and the gas remaining in the eudiometer is oxygen. 100 volumes of oxygen have therefore been burned by the hydrogen of the olefiant gas, which gives for the composition of 100 volumes of the gas, 200 of hydrogen, 100 “ vapour of carbon. Noav, 2 vol. of hydrogen weigh 0.1382 14.29 1 “ vapour of carbon 0.8290 85.71 . 0.9672 100.00 which approaches very nearly to the density 0.9784 found by ex- periment. The formula assigned to bicarburetted hydrogen is C4H4. The gas used for lighting buildings is principally composed of carburetted hydrogen gas, and will be treated of under organic chemistry. SULPHIDE OF CARBON. 333 Sulphide of Carbon, or Sulphocarbonic Acid, CS3. COMBINATION OF CARBON WITH SULPHUR. § 267. Sulphur and carbon do not combine when mixed toge- ther and heated under the ordinary pressure of the atmosphere, for the sulphur distils over before the temperature is sufficiently elevated to cause their combination. But if the carbon be heated to redness in a porcelain tube, and vapour of sulphur be passed over it, the carbon burns in this vapour as in oxygen. Now, when carbon burns in oxygen, it is changed into carbonic acid, C03; and when it burns in the vapour of sulphur, it is changed into sul- phuret of carbon, or sulphocarbonic acid, CS3. When burned in oxygen, the latter must be in excess, or, otherwise, carbonic oxide alone is formed; but this result is not to be feared in the combus- tion of carbon in vapour of sulphur, for nothing is ever formed but sulphocarbonic acid, and no less sulphuretted compound of carbon has yet been obtained. To obtain sulphuret of carbon, a porcelain tube is filled with small pieces of coal, and arranged in a reverberatory furnace (fig. 286). The end a of the tube is closed with a cork, and should project far enough from the fur- nace so as not to burn the cork. To the other end b, a curved adapter is filled, the beak of which descends into a very small quantity of water in a receiving- 7 bottle. When the por- celain tube is heated to redness, a piece of sul- phur is introduced at a, and the cork immedi- ately replaced. The sulphur fuses, and the tube being slightly inclined from a to b, it flows toward the hottest part of the tube, where it is vapor- ized, passes in this state over the ignited carbon, combines with it, forming sulphide of carbon, which is deposited in the adapter, and falls in oily drops to the bottom of the water in the re- ceiver. When the vapour has ceased to pass over, another piece of sulphur is introduced, and so on, until the greater part of the carbon in the tube has disappeared. When a larger quantity of the sulphide is re- Fig. 286. Fig. 287. 334 CARBON. quired, the porcelain tube is replaced by a tubulated stoneware retort (fig. 287), into the tubulure of which is fitted a porcelain tube ab, descending nearly to the bottom of the retort, and luted with clay at the tubulure a. The retort being then filled en- tirely with coals, and placed in a furnace (fig. 288), to its neck is Fig. 288. adapted a long tube, which passes through a condenser cd filled with cold water, and communicates with a receiver, as in the pre- ceding operation. The retort being brought to a strong red-heat, pieces of sulphur are successively dropped into the porcelain tube, which is imme- diately closed by a cork. The sulphur, in its descent to the bottom of the retort, is converted into vapour, and traverses the mass of ignited carbon, producing sulphide of carbon, which condenses in the refrigerator and flows into the receiver. In a few hours, several hundred grammes (a lb., more or less) of the sulphide may thus be obtained. The sulphide forms a yellow oily stratum under the water of the receiver, but is not pure, and always contains more or less sulphur in solution. To purify it, it is distilled from a glass retort in a water-bath, the sulphur remaining in the retort, and the sul- phide of carbon distilling under the form of a colourless liquid. The distilled liquid, being kept in contact, for some time, with chloride of calcium, is deprived of its water, and again distilled in a dry apparatus. § 268. Sulphide of carbon is a colourless, very volatile liquid, possessing a peculiar and extremely disagreeable odour; its density is at 32° 1.293 and at 59° 1.271 It boils at 118J°, under the ordinary pressure of the atmosphere, so that, at common temperatures, its vapour has already a con- siderable tension, and the liquid evaporates rapidly, producing a great degree of cold. SULPHIDE OF CARBON. 335 Although it does not dissolve appreciably in water, yet water which has been for some time in contact with it becomes impreg- nated with its peculiar odour. Absolute alcohol and ether dissolve and mix with it in every proportion, whether singly or together. It burns in the air with a blue flame, producing carbonic and sulphurous acids. It dissolves sulphur and phosphorus in large quantities, and if the solutions be allowed to evaporate slowly, they are deposited in regular crystals. It was observed that crystallized sulphur could be obtained in this way, in the form of octahedrons of the fourth system, resembling those found in the solfateme. § 269. Sulphide of carbon is analyzed, by burning it with the oxide of copper, so as to transform the carbon into carbonic, and the sulphur into sulphuric acid. It is necessary, in the first place, to weigh a certain quantity of it accurately, and under such circumstances that, notwithstand- ing its great volatility, it cannot lose by evaporation. To do this, a bulb A (fig. 289), blown between two points a, b, finely drawn out, is weighed, and then filled with the sul- phide of carbon, by inserting one of the points a into the liquid, and sucking at the other point b, until the bulb is nearly filled. The open end b being closed with the finger, the end a is inserted in the flame of an alcohol lamp, and closed hermetically by fusion. The same process is repeated with the end b, so that the sulphide of carbon is hermetically closed in the globe. By again finding the weight of the bulb when filled, its increase in weight necessarily represents the quantity of sulphide introduced. On the other hand, a glass tube is prepared of similar size with that in § 260, drawn out to a point at one end b, and freely open at its end a (fig. 290). Some oxide of copper has been previously Fig. 289. Fig. 290. calcined in an earthen crucible, and allowed to cool where it could not attract the moisture of the air. The tube having been perfectly dried, a scratch is made with a fde on one of the ends of the bulb, which is then broken olf, and the liquid exposed. The bulb and 336 CARBON. detached piece of glass being allowed to fall to the bottom of the tube, a depth of 2 or 3 dec. (8 in.) of oxide of copper is immedi- ately poured on it, and the balance filled with oxide of lead or litharge, which should occupy at least 3 dec. (10 in.) of the tube. The tube is placed on a long sheet-iron furnace, and its open end a fitted to the apparatus A, B, C, described in § 260. The tube A, containing sulphuric pumice-stone, has been accurately weighed alone, and the potassa bulbs B, with the tube C, containing frag- ments of potassa, have been weighed together. It is important to arrange the apparatus with the least pos- sible delay, since the sulphide of carbon contained in the open bulb might give off its vapours through the tube, and escape com- bustion at the commencement of the experiment. When all is thus prepared, the anterior portion of the tube containing the oxide of lead is heated rapidly with ignited coals; then progres- sively that portion containing oxide of copper; and, lastly, a coal is placed, with great care, near the bulb, so as to effect a gentle distillation of the sulphide. Its vapour passing over the oxide of copper, its carbon is burned to carbonic acid, and the greater por- tion of its sulphur remains combined with the oxide of copper, in the state of a sub-sulphate of the protoxide. A portion, however, being disengaged in the state of sulphurous acid, would accompany the carbonic acid, if it were not entirely absorbed by the heated oxide of lead which is in the anterior part of the tube. Carbonic acid alone is therefore absorbed in the tubes B and C. When the evolution of gas has ceased, the experiment is concluded in the manner described in § 260. On again weighing the tube A, it will be found not to have appreciably increased in weight, which proves that the substance contained no hydrogen. The increase of weight of the tubes B and C gives the weight of carbonic acid produced, and, consequently, that of the carbon contained in the sulphide of carbon subjected to analysis. As this body contains only carbon and sulphur, it is evident that the difference will give the quantity of sulphur; but the latter can also be directly determined, and thus a complete analysis of the substance executed. To do this, an additional quantity of the sulphide is weighed in a closed bulb (fig. 289), and, after having opened one end of it, it is placed at the bottom of a glass tube, resembling that used in the preceding experiment, but not so long. It is filled entirely with a mixture of oxide of copper and carbonate of soda, and the open end closed with a perforated cork. The mixture of oxide of copper and carbonate of soda is gradually heated, and when this portion of the tube is heated to a dull red, coals are approached near the end containing the bulb. The vapour of the sulphide burns, carbonic acid is evolved, and the sulphur is converted into sulphuric acid, which combines with the soda. cyanogen. 337 When the operation is terminated, and the tube entirely cooled, the mixture is withdrawn from the tube, and thrown into a capsule ; the tube rinsed several times with hot water, which is poured into the same capsule, taking care not to lose a single drop ; and, lastly, the dish, capsule, and its contents, heated for some time. The ex- cess of carbonate of soda and the sulphate of soda having dissolved, the liquid is filtered, and the residue washed with hot water, until it no longer shows traces of soluble matter. All the sulphuric acid produced by the combustion is then found in the liquid, to- gether with a large excess of carbonate of soda: chlorohydric acid is poured into the solution until it becomes highly acid, whereby carbonate is changed into chloride of sodium; and if a solution of chloride of barium be now poured into the liquid, the sulphuric acid will be precipitated in the state of sulphate of baryta. From the weight of the sulphate obtained, we can infer the quantity of sulphur contained in the sulphide of carbon. By combining the results of the two analyses, it is ascertained that the substance analyzed contains only sulphur and carbon in the ratio of 1 eq. carbon 6.0 15.79 2 “ sulphur 32.0 84.21 381) lOOdM) 1 vol. of its vapour contains J vol. vapour of sulphur 2.2180 % “ “ carbon 0.4145 2.6325 The density of the vapour, as found by direct experiment, is 2.67. Sulphide of carbon presents the same formula in equivalents as carbonic acid. As carbonic acid combines with the metallic pro- toxides BO, forming carbonates R0,C02, so, sulphide of carbon combines with the metallic protosulphides IIS to form true salts RS,CS3, which are often isomorphous with the corresponding com- pounds R0,C03. The name of sulphocarbonic acid has therefore been properly given to sulphide of carbon, and the name of sulphocarbonates to its compounds with the monosulphides. COMBINATION OF CARBON WITH NITROGEN. Bicarburet of Nitrogen, or Cyanogen, CaN, or Cy. § 270. Carbon and nitrogen form a very important compound— cyanogen*—the detailed study of which will be more appropriate * The discovery of cyanogen, due to M. Gay-Lussac, has been of great import- ance in chemical science, because it furnished the first example of a compound body performing the functions of an element in its combinations. 338 CARBON. among the products derived from the animal kingdom; but as its compounds with the metals present a complete analogy with the corresponding chlorides, and are frequently used as reagents to characterize metallic solutions, and distinguish them from each other, we shall detail at present its principal properties, as well as those of its compound with hydrogen, or cyanohydric acid, which closely resembles chlorohydric acid. Nitrogen and carbon do not combine directly; but, if a mixture of carbonate of potassa and carbon be heated together in a porce- lain tube, while a current of nitrogen is passed through it, car- bonic oxide is disengaged; and if the residue be treated with water, a considerable proportion of cyanide of potassium is dis- solved. Cyanide of potassium is prepared, on a large scale, by heating in iron vessels mixtures of carbonate of potassa and the carbonaceous residues from the incomplete calcination of animal matters, such as flesh, bones, horn, etc. It will be described more fully when treating of cyanide of potassium. If a hot solution of nitrate of mercury be poured into a hot and concentrated solution of cyanide of potassium, and the mixture be allowed to cool, crystallized cyanide of mercury is separated, which may be purified by recrystallization. By means of the cyanide of mercury, cyanogen and cyanohydric acid are readily obtained. Cyanogen is obtained by heating cyanide of mercury in a small retort, or in a tube closed at one end, and furnished with an exit tube (see fig. 239, § 199), which conveys the gas into a bell-glass over water, or, better still, over mercury (fig. 291). The cyanide Fig. 291. is decomposed into free cyanogen and metallic mercury, the latter condensing in the upper part of the retort. By continuing the heat until the disengagement of gas ceases, it will be observed that the cyanide does not entirely undergo the simple decomposi- tion just mentioned; for a brown substance remains, presenting exactly the same composition as cyanogen, and which has there- fore been called yaracyanogen. The proportion of cyanogen pass- CYANOGEN. 339 ing into this isomeric condition varies according to the manner in which the cyanide is heated; but hitherto it has never been so decomposed as entirely to avoid its formation. Cyanogen is a colourless gas, with a sharp, peculiar smell, re- sembling that of wild-cherry water; its density is 1.86; it is liquefied at common temperatures, under a pressure of 4 or 5 atmospheres, or when cooled to —4°, without an increase of pressure, and is then a colourless, very volatile liquid, of the density 0.9. It burns with a very characteristic purple flame, giving off car- bonic acid, and setting the nitrogen free. Water dissolves 4 or 5 times its volume of the gas, but readily parts with it when the temperature is raised. The aqueous solu- tion, left to itself, even in a well-corked bottle, becomes at last of a brown colour, and deposits, after some time, a brown powder. The decomposition in this case is too complicated to be introduced here; and has not, moreover, been sufficiently explained. Alcohol dissolves 20 to 25 times its volume of the gas. § 271. Cyanogen being a combustible gas, and affording, by its combustion, gaseous products easily separated, it might be sup- posed that it could be readily analyzed by the eudiometer; but if a mixture of cyanogen and oxygen be exploded in the eudiometer, the combustion is always observed to be imperfect. A more per- fect combustion will be obtained by adding to the mixture of oxygen and cyanogen a certain proportion of a detonating mixture of oxygen and hydrogen in the proportions constituting water. Such a detonating mixture is easily prepared by decomposing water by a galvanic battery, and collecting the gases disengaged at both poles in the same vessel. Suppose that 100 of oxygen gas, 250 “ cyanogen “* Total 350 are introduced into the eudiometer, and, in addition, an indeter- minate volume of the detonating mixture, which need not be mea- sured, since combustion will convert it wholly into water: after waiting a few moments, to allow the gases to mix freely, an electric spark is passed through. The detonating mixture is converted into water, and the cyanogen gives off carbonic acid and free nitrogen. The volume of gas is measured, composed of carbonic acid, nitrogen, and the excess of oxygen, and found to be 350. If the gaseous mixture be shaken with a small quantity of a solution of caustic potassa, the carbonic acid is absorbed, and nitrogen and oxygen alone remain, whose volume is found to be 150. 340 CARBON. The 100 of cyanogen have therefore afforded 200 of carbonic acid, containing 100 of vapour of carbon. It being still requisite to analyze the mixture, 150 of nitrogen and oxygen, a certain quantity of hydrogen, say 150 volumes, is introduced into the eudiometer, making the total volume 300, and the electric spark passed through it. After the explosion, the volume of gas remaining is found to measure 150, so that 150 volumes have disappeared by combustion; and they are evidently composed of oxygen and hydrogen in the proportions forming water, that is, 100 of hydrogen, and 50 of oxygen. Therefore, in the 150 of the mixture of nitrogen and oxygen which remained after the absorption of carbonic acid by potassa, there were 50 of oxygen, and, consequently, 100 of nitrogen. It follows, therefore, that 100 volumes of cyanogen contain 100 of vapour of carbon, 100 of nitrogen. 1 volume of cyanogen gas, therefore, contains 1 volume of vapour of carbon, and 1 volume of nitrogen, condensed into 1 volume. The analysis is confirmed by the density of cyanogen gas, which has been ascertained by direct experiment. 1 vol. of vapour of carbon weighs 0.8290 1 “ nitrogen 0.9713 The sum weighs 1.8003 which does not differ materially from the number 1.86 given by the direct determination. The difference between the two num- bers is, however, too great to attribute it to error of observation, and is rather due to the fact that, at ordinary temperatures, the molecules of cyanogen gas are already closer than they should be, if it could be assimilated to the more perfect gases, as nitrogen, hydrogen, etc. The eudiometric analysis just described does not furnish very exact results; because, 1st. The cyanogen gas is measured in a state of anomalous condensation, as just stated, and consequently its observed volume is too small; 2d. In the combustion of cyano- gen with oxygen, in the presence of mercury, a small quantity of protonitrate of mercury is frequently formed, which causes a cer- tain quantity of nitrogen and oxygen to disappear. The composition of cyanogen can be ascertained more accu- rately by burning it with oxide of copper, and collecting the gaseous products of combustion. A glass tube being filled half with oxide of copper, and half with metallic copper, to one end of it is fitted an exit tube for conveying the gases over a mercurial trough, and to the other end, by means of a cork, a small glass retort containing cyanide of mercury. When the tube is heated CYANOGEN. 341 to redness, the cyanide is slowly decomposed by heat. Cyanogen first passes over the oxide of copper, where it is resolved into car- bonic acid and nitrogen, and the mixture of the two gases then passing through the anterior part of the tube containing metallic copper, the latter decomposes any oxides of nitrogen which might have formed by the combustion of cyanogen. After having allowed a small quantity of gas to escape, so as to be sure that the apparatus no longer contains the smallest proportion of air which previously filled it, a portion of it is collected in a gradu- ated tube or bell-glass, and measured accurately. A small quantity of solution of potassa is then introduced, and absorbs the carbonic acid; whereby the gaseous volume is reduced to The experiment proves that, by burning cyanogen with oxygen, it yields a volume of carbonic acid, double that of the nitrogen which is set free, and by combining the result with the known densities of cyanogen and nitrogen, and with the composition of carbonic acid, it gives the composition of cyanogen. For, 2 volumes of carbonic acid contain 1 volume of vapour of carbon, which weighs 0.8290 1 vol. of nitrogen weighs 0.9713 1.8003 Since a weight 1.8003 of cyanogen contains 0.8290 of carbon, 0.9713 of nitrogen, then 100 of cyanogen contain Carbon 46.15 Nitrogen 53.85 100.00 Since the number 1.800 differs so little from the 1.86 found by experiment for the density of cyanogen that it may be attributed to the want of normal elasticity of the gas at ordinary tempera- tures, it may be inferred from the above numbers that 1 vol. of cyanogen contains 1 vol. of vapour of carbon, and 1 vol. of ni- trogen. Cyanogen may also be analyzed by another method, still more exact than those hitherto described, which simply consists in burn- ing with oxide of copper a metallic cyanide, the composition of which is easily ascertained, such as the cyanide of mercury. The quantity of mercury contained in lgm of the cyanide is first ascertained by putting a given weight of it into the bulb A of a curved tube abed (fig. 292), the end a being made to com- municate with an apparatus which slowly disen- gages hydrogen, the end d is drawn to a point. The bulb A being heated by an alcohol lamp, Fig. 292. 342 CARBON. the cyanide is decomposed, and mercury being set free, is carried by the current of gas into the part bed, where it is condensed. When the operation is terminated, which can be ascertained by the mer- curial vapour ceasing to condense, the tube is broken off at b. The weight of the portion bed, with the contained mercury, having been determined, the mercury is removed entirely, and the tube bed replaced in the balance. The weight necessary to restore the equilibrium is exactly that of the mercury obtained. It will thus be found that 100 parts of cyanide of mercury contain 79.36 of mercury and consequently 20.64 of cyanogen 100.00 The composition of the cyanide being known, to determine that of cyanogen, a known weight of the cyanide is burned with oxide of copper, and the weight of the resulting carbonic acid and nitro- gen determined. The carbonic acid is determined exactly as in the analysis of oxalic acid (§ 260), except that, as the substance contains nitrogen, and the production of a little oxide of nitrogen is to be feared, a longer tube is employed, and about 2 decimetres (8 in.) of its ante- rior part filled with metallic copper. The mercury condenses in the tube A, which has been filled with pieces of chloride of calcium, and the increase in weight of the tubes B, C gives the weight of carbonic acid produced. In order to determine the quantity of nitrogen contained in the cyanide of mercury, the apparatus employed to determine the nitrogen in the nitrate of lead is used (§ 108). A quantity of bicarbonate of soda is put at the bottom of the tube ab; above it, a column of 4 or 5 centimetres (1J to 2 in.) of oxide of copper, then a mixture of a given weight of cyanide of mercury and oxide of copper, followed by an additional quantity of pure oxide of copper, and, lastly, a length of 2 decimetres (8 in.) of metallic copper. The operation is conducted exactly as prescribed in § 108, and, when concluded, the vol. of nitrogen which alone remains in the bell-glass is determined, and from it the weight of nitrogen contained in the given weight of cyanide operated on. Cyanohydric Acid, H,C3N, or IlCy. § 272. Cyanogen and hydrogen do not combine directly, and the cyanohydric acid is obtained by decomposing the metallic cyanides by chlorohydric acid. It may be procured in the anhydrous state, or in solution. To obtain the anhydrous acid, cyanide of mercury is decomposed CYANOHYDRIC ACID. 343 by concentrated chlorohy- dric acid in a flask (fig. 293), connected with a tube abc, the first half of which, ab, is filled with pieces of marble, and the second half, be, with pieces of fused chloride of cal- cium. Following the tube abc is a U-tube, surrounded by a frigorific mixture. The chlorohydric acid decomposes the cyanide of mercury, Fig. 293. HgCy+HCl=HgCl+HCy, disengaging gaseous cyanohydric acid, which carries over chloro- hydric acid and aqueous vapour, when the mixture passes through the tube abc. The chlorohydric, being a powerful acid, decom- poses the marble, forming chloride of calcium, water, and free carbonic acid, Ca0,C03+HCl=CaCl+H0+C03. But cyanohydric being, on the contrary, a very feeble acid, does not react upon carbonate of lime. We have, therefore, a mixture of cyanohydric and carbonic acids and aqueous vapour, which penetrates the second half be of the tube, filled with chloride of calcium, where aqueous vapour alone is absorbed, and the mixed acids pass into the refrigerated tube. Cyanohydric acid is con- densed into the liquid state, while the carbonic acid maintains its gaseous condition; but the former acid necessarily contains all the carbonic acid it can absorb under the circumstances. The anhydrous acid is better prepared by decomposing cyanide of mercury by gaseous sulfhydric acid, in a long glass tube ab, to Fig. 294. which is fitted a U-tube cooled by a refrigerating mixture. The end a is connected with an apparatus for disengaging dry sulfhydric acid, which is prepared by decomposing fused protosulphide of iron by cold dilute sulphuric acid in a tubulated bottle. A slow 344 CAKBON. current of sulfhydric gas, which can be regulated at will, is pro- duced, and dried by being passed through a tube cd, filled with pieces of chloride of calcium. The cyanide is decomposed by the sulfhydric acid, forming sulphide of mercury and anhydrous cyano- hydric acid, which remains gaseous in the tube ab, if kept at a temperature above 77°, but condenses in the refrigerated receiver. The decomposition takes place progressively from the extremity a, if the sulfhydric current be slow, and as the white cyanide is con- verted into a black sulphide, the progress of the operation may be easily followed. If, therefore, the experiment be stopped before the whole of the cyanide is decomposed, perfectly pure cyanohy- dric acid is obtained in the receiver. Cyanohydric acid is a colourless, very volatile liquid, solidifying at 5°, and boiling at 79.7°. The degree of cold produced by its evaporation is generally sufficient to congeal the portion remaining liquid. The density of the acid is 0.697; that of its vapour, 0.947. Its odour is very penetrating, resembling that of bitter almonds. § 273. This acid can be accurately analyzed by determining the hydrogen and carbon simultaneously, and then the nitrogen. To perform both operations, the- liquid acid is introduced into a small bulb drawn out at both ends, closed, and accurately weighed. To determine the hydrogen and carbon, a strong glass tube is prepared, about 60 centimetres (2 ft.) in length, open at one end, and drawn out at the other into the form of an open tubulure. It is partly filled with oxide of copper, and the remainder with me- tallic copper, which should occupy at least 2 decimetres (8 in.) of its length. The open end is fitted to the apparatus intended to collect the water and carbonic acid, as described in § 260, and represented in fig. 279. The bulb containing the given weight of acid is fixed, by means of caoutchouc, to the tubulure which terminates the tube, so that the pointed part, which is closed, may enter the tubulure to the distance of about 1 centimetre (J in.). When the combustion-tube is heated to redness, by pressing the point of the bulb against the side of the tubulure, it is broken, and the bulb opened. The acid immediately distils over, and its vapour is burned by the oxide of copper into water, carbonic acid, nitrogen, and deutoxide of nitro- gen, which last is decomposed by the heated metallic copper filling the anterior part of the tube, and converted into nitrogen. The water and carbonic acid are condensed in the apparatus A, B, C (fig. 279). The distillation of the acid may be easily regulated by cooling the bulb. The nitrogen is determined exactly as in the analysis of hypo- nitric acid (§ 120), except, that two-thirds of the combustion tube is filled with oxide of copper, and the remaining third, with metal- lic copper, as in the preceding experiment. It is thus proved, that lgm of the acid yields CYANOHYDRIC ACID. 345 0.333gm of water, 1.629gm of carbonic acid, 412.1C0 of dry nitrogen gas at 32°, and under a pressure of 0m.760 (29.92 in.), corresponding to the weight 0.518gm of ni- trogen. We infer, from these experimental data, that cyanohydric acid is composed of 1 eq. of hydrogen 1.0 3.70 2 carbon 12.0 44.44 1 “ nitrogen 14.0...... 51.86 1 “ cyanohydric acid 27.0 100.00 * Its formula is therefore H,C2N or HCy. Cyanogen and hydro- gen are combined in it in the same manner as chlorine and hydrogen in chlorohydric acid. 1 volume of cyanohydric acid contains a J volume of hydrogen, and a J volume of cyanogen without condensation, for we have J the density of hydrogen 0.0346 J “ cyanogen...., 0.9300 0.9646 and direct experiment has given 0.947 for its density. § 274. Cyanohydric acid should be preserved in hermetically sealed tubes, filled in the manner described for sulphurous acid (§ 129); but it does not long remain unaltered, for in a few days the liquid turns brown, and deposits a brown powder. The che- mical reaction occurring in this imperfect decomposition appears to be very complex, and has not yet been thoroughly investigated. Cyanohydric, commonly called prussic acid, is one of the most violent poisons known. A drop, placed on a dog’s tongue, kills him instantly. We should therefore handle it with great caution, and be particularly careful not to inhale its fumes. It is soluble in every proportion in water; and its aqueous so- lutions are used in medicine. To prepare solutions of the acid, into a flask A, heated by a water-bath (fig. 295), are introduced 1 part of ferrocyanide of potassium, or yellow prussiate of potash (the double cyanide of potassium and iron, 2KCy+FeCy), and 1J parts of oil of vitriol, diluted with 2 pts. of water. A long glass tube abc is adapted to the flask, passes through a condenser DE, through which a current of cold water circulates, and enters the water of the refrigerated bottle B. By introducing into the bottle more or less water, a more or less concentrated solution of prussic acid is obtained. In all cases, it is necessary to ascertain the quantity of acid dissolved in the liquid, which is easily done by pouring into a given quantity of it a solution of nitrate of silver, whereby a precipitate of 346 EQUIVALENTS OF METALLOIDAL ELEMENTS, Fig. 295. cyanide of silver is formed, from the weight of which the quantity of acid can be inferred. We can also obtain a standard solution of this acid, by dissolv- ing a given quantity of cyanide of mercury in water, and passing a current of sulphuretted hydrogen through the liquid; and re- moving the excess of sulfhydric acid by shaking the liquid for some minutes Avith carbonate of lead. A solution of prussic acid is liable to alteration, and should therefore be made only as it is required. REMARKS ON THE EQUIVALENTS OF THE METAL- LOIDAL ELEMENTS. § 275. We have referred the equivalents of the elements to the equivalent of hydrogen, assumed to be 1.0; hut any other ele- ment might have been selected as a term of comparison, as oxygen, chlorine, etc., and would have given rise to other series of numbers, differing greatly in their absolute values from those adopted. They would, however, have always presented the same proportions to each other. Let us assume the equivalent 8 of oxygen as unity or 100, and calculate the numerical value of three of the other metalloidal elements. It is evident that, in order to obtain the equivalent of hydrogen, according to this hypothesis, we must make the pro- portion 8:1:: 100: x, whence, x=12.5. The equivalents of the other elements can be calculated in the same way; and the following series will result: EQUIVALENTS OF METALLOIDAL ELEMENTS. 347 0=100 11=1 Equivalent of oxygen 100.00 8.0 “ hydrogen 12.50 1.0 ££ nitrogen 175.00 14.0 “ sulphur 200.00 16.0 ££ selenium 494.25 39.5 ££ tellurium 802.50 64.2 “ chlorine 443.75 35.5 ' ££ bromine 1000.00 80.0 “ iodine 1575.00 125.0 “ fluorine 237.50 19.0 “ phosphorus 400.00 32.0 “ arsenic 937.50 75.0 “ boron 137.50 10.9 ' “ silicium 266.75 21.3 v “ carbon 75.00 6.0 A glance at the second column of figures shows that, of fifteen elements, the equivalents of ten, or two-thirds, of them are repre- sented by whole numbers, that is they are exact multiples of that of hydrogen, the lightest of them all. They are: Hydrogen Equivalent = 1.0 Oxygen “ 8.0 Nitrogen ££ 14.0 Sulphur <£ 16.0 Bromine ££ 80.0 Iodine ££ 125.0 Fluorine ££ 19.0 Phosphorus ££ 32.0 Arsenic ££ 75.0 Carbon ££ 6.0 If only these ten were known to us, the law would immediately be assumed that the equivalents of the metalloidal elements are exact multiples of the equivalent of hydrogen.* But the other five metalloids form an exception to the law. It must, however, be observed, that great uncertainty still exists as to the true value of the equivalents of these last substances; for many of them are rare, we are not sure of having obtained them in a state of purity, and the numbers found by various experi- * An English chemist, Dr. Prout, first announced this law, about twenty-five years since. His confidence ia the precision of this law was such, that he did not hesitate to change, arbitrarily, the numerical values which direct experiment had assigned as the equivalents of the elements, in order to render them exact multi- ples of that of hydrogen. Prout’s ideas were not generally adopted by chemists on the continent, but M. Dumas, by his accurate determination of the equiva- lents of hydrogen, carbon, and some metallic elements, has again drawn atten- tion to the point, and shown the only manner in which the question can be de- cided. 348 EQUIVALENTS OF METALLOIDAL ELEMENTS. raenters differ often more widely than the corrections which might he required for the equivalents we have adopted, in order to in- clude them in the law advanced. Whereas ten elements which satisfy the law are those of which the equivalents are known with most certainty, and which have been recently determined, by a great number of experiments perfectly corroborating each other. Among the elements which form the exception, there is only one, chlorine, which has been, and quite recently, the object of many experiments, the special design of which was to ascertain if its equivalent could be considered as a multiple of that of hydrogen. Those of the experiments to which chemists attach most confidence have given the number 443.2, that of oxygen being represented by 100. According to the hypothesis of hydrogen being equal to 1.00, that of chlorine, from these experiments, is 35.45: it is, therefore, not an exact multiple of the equivalent of hydrogen. It will be subsequently seen, that the equivalents of a certain number of simple metallic bodies, carefully determined within the last few years, are exact multiples of that of hydrogen, while others do not present equally simple relations. We shall, therefore, not decide whether the foregoing law be admitted for all the elements, or whether it be applicable to only a certain number of them. There may possibly be a group of elements whose equivalents are multiples of hydrogen, and, as regards the others, their equivalents may be multiples of the equi- valent of some other element, or even they may be represented by a sum of which one of the components may be a multiple of the equivalent of hydrogen, and the remainder multiples of the equi- valents of one or several other elements. The attention of chemists is now directed to this important question, and its solution may be soon expected from their united researches.* * Having adopted the hydrogen scale (H=l) in this translation, because of its more general adoption by English chemists, we have also preferred the equiva- lent numbers given in the Annual Report of Liebig and Kopp. Hence, the slight deviations observable in the equivalents of some of the metalloids, which, how- ever, have no material influence on the science at present. Hence, also, the remarks in § 275 of the original work, which were applied to the hydrogen scale, starting from that of oxygen, have been modified in the translation, to apply them from the hydrogen to the oxygen scale.—J. C. B. PART II. THE METALS. § 276. It was stated (§ 55) that the metals are simple bodies, good conductors of heat and electricity, and possessing a peculiar brilliancy, called the metallic lustre. They exhibit great diversity in their physical and chemical properties, and are therefore sus- ceptible of the most varied applications. Some of them possess great malleability and tenacity, and are the only ones used in an isolated state; the others are only valu- able in combination. Some of them have a feeble affinity for oxygen, being scarcely affected by atmospheric air, in which they remain unaltered for an almost indefinite period, provided the air be not saturated with moisture. Others again readily combine with the oxygen of the air, even in the cold, and are converted into oxides. It is evident that the latter, in their metallic state, cannot be ordinarily used. The metals are hence divided into two great classes, according to their applications. First Class.—Metals which, on account of their great affinity for oxygen, are rapidly oxidized in the air, and cannot be used in the arts in their metallic state. They are: Potassium, Sodium, Lithium, Barium, Strontium, Calcium, Magnesium, Glucinum, Aluminum, Zirconium, Thorium, Yttrium, Cerium, Lanthanum, Didymium, Erbium, Terbium. The metalloidal compounds of these metals are used in the arts when they abound in nature, and their separation from their natural combinations is not too expensive. It will be shown that potassium, sodium, barium, calcium, magnesium, and aluminum furnish a host of products of the highest practical value. The other metals comprised in the foregoing list have as yet received no useful application, and possess only a purely scientific interest. 349 350 THE METALS. Second Class.—Metals Avhose affinity for oxygen is so feeble as to render them but slightly alterable in our atmosphere at ordi- nary temperatures. They are: Manganese, Iron, Cobalt, Nickel, Chromium or chrome, Tungsten, Molybdenum, Vanadium, Zinc, Cadmium, Copper, Lead, Bismuth, Mercury, Tin, Titanium, Tantalum or columbium, Niobium, 'Ilmenium, Pelopium, Antimony, Uranium, Silver, Gold, Platinum, Palladium, Rhodium, Iridium, Ruthenium, Osmium. This is the more numerous class of metals, but in order that they may be really useful in the arts, they must satisfy several conditions which singularly reduces their number. Thus, two essential conditions are a certain degree of malleability and tena- city, without which they cannot be worked into a convenient form; and they should possess these properties in such a degree as to render their working not too expensive. Again, the substances in nature from which they are extracted should not be too rare, nor difficult to manage, as otherwise the metal acquires too great a commercial value, and is used only when a cheaper substitute cannot be found. Iron, manganese, nickel, and cobalt, in their metallic state, present nearly similar properties; but iron is much more abundant in nature, more easily extracted from its ores, and is naturally preferred to the other three when it can subserve the same ends. Manganese is more oxidizable than iron, and changes more rapidly in the air; thus affording another reason for prefer- ring iron. Nickel and cobalt, on the other hand, are less oxidiz- able, possess a ductility and tenacity comparable in this respect to iron, and would certainly take its place in many of its applica- tions, were they less expensive. The brittle metals are not employed in the metallic state, but are frequently combined Avith the malleable metals, forming alloys Avhich present peculiar physical properties. The metals which possess sufficient malleability to be used in the metallic state, are: Manganese, Iron, Cobalt, Nickel, Zinc, Cadmium, GEOLOGY. 351 Copper, Lead, Mercury, Tin, Silver, Gold, Platinum, Palladium, Iridium. Several of them, however, have not yet been applied in the arts, because their ores are too rare, and difficult to manage, or because their properties resemble those of other metals more readily and cheaply obtained. § 277. State of the Metals in Nature.—Metals exist in various states in nature. Some are found isolated, and are then called native. Those which, having a very feeble affinity for oxygen, do not change under atmospheric influence, belong to this class: such are, gold, platinum, rhodium, iridium, palladium, silver, mer- cury, and bismuth. Many others are found in combination with oxygen, sulphur, or arsenic; such as manganese, iron, cobalt, nickel, chrome, tungsten, molybdenum, vanadium, zinc, cadmium, copper, lead, bismuth, mercury, tin, titanium, antimony, uranium, and silver. Some of this division are found in the state of insoluble salts, chiefly in that of carbonates or silicates. The metals of the first class which, as will be remembered, have a great affinity for oxygen, are formed in the state of salts, especially in that of in- soluble silicates or carbonates; they are, however, sometimes met with as soluble salts, dissolved in the waters of the ocean or of salt-springs. A knowledge of the natural situation of the different metals is highly important to the chemist and metallurgist; and we shall be careful to indicate it, when describing each particular metal. But, in order to give our indication some value, it is necessary to pre- mise a few elementary remarks on geology, or the science which treats of the nature and mode of aggregation of the various ma- terials which compose our globe. GENERAL REMARKS ON THE CONSTITUTION OF THE EXTERIOR CRUST OF THE GLOBE. § 278. That portion of the crust of the globe which is accessible to us is composed of mineral substances of various character, which when aggregated in masses are called rocks. Rocks differ from each other, either in the chemical nature of the minerals which compose them, or only in the manner in which these mine- rals are united, whereby they receive a different structure. In some rocks, minerals are distributed with a certain degree of regularity, being stratified in parallel layers, which may be traced to a great extent. The stratification is often evinced by parallel fissures in the rocks, separating them into layers or 352 GEOLOGY. courses analogous to those seen in edifices constructed of hewn stone. At other times, the stratification is recognised by the ten- dency of the rock to divide in parallel layers, as in the slates. These are called stratified rocks. Others do not present this characteristic, for the fissures which traverse them are irregular, and their fracture shows that the minerals are indiscriminately arranged, without any appearance of symmetry. To distinguish these from the former, they are called compact, or non-stratified rocks. Non-stratified rocks are composed of crystalline minerals, and their appearance is that of a mass of heterogeneous mineral sub- stances, which, after having been fused, are allowed slowly to cool. The chemical elements composing the mass are then grouped according to their reciprocal affinities, and different compounds result, which segregate by crystallization. The mass, after cool- ing, presents the appearance of an agglomeration of different crys- tals, arbitrarily scattered, and without any appearance of regular arrangement. Non-stratified rocks are, therefore, often called Plutonic rocks, or rocks of igneous origin, which tacitly admits that they were originally fluid, and assumed their present form by solidifying during a slow process of cooling. §279. Stratified rocks, on the contrary, present an appearance similar to that of deposits still forming at the bottom of seas and rivers, and the large quantity of remains of aquatic animals con- tained in the majority of them renders the analogy still more striking. Geologists admit that these rocks have been formed under water, and therefore call them Neptunian, or sedimentary rocks. The deposits which form at the bottom of seas and rivers natu- rally take the form of nearly horizontal layers, and the inferior layers are, evidently, first deposited. The same must have taken place with the sedimentary rocks found on the surface of the globe ; so that the order in which these rocks have been superim- posed on each other gives* a certain index of the periods of their formation and their relative age. We may thus establish a chro- nological scale of their formation. In level countries, stratified rocks are nearly horizontal (fig. 296), but in mountainous regions they are generally observed to be inclined (fig. 297), often assuming a vertical direction, and are Fig. 296. Fig. 297. GEOLOGY, 353 sometimes even overturned, and lean in a contrary direc- tion, as in fig. 298. It frequently happens that inclined strata are covered with horizontal layers, with a different direction of stra- tification from the former; in which case the latter are said to be unconformable (fig. 299). It is evident that, between the deposit of the two series of strata, some great revolution has taken Fig. 298. Fig. 299. place on the surface of the globe, which has remarkably altered its original aspect. An attentive study of the constitution of the globe has shown that this effect upon the strata has been produced by the upheaval of a more or less considerable mass of non-stratified rocks. The latter does not always force its way to the surface, and the stratified rocks have been merely upheaved, as in fig. 299. But the non-stratified rock has frequently pierced the sedimentary rocks, and formed the projecting spire of a range of mountains, both sides of which are covered by the edges of sedimentary strata (fig. 300). Fig. 300. When sedimentary rocks are in immediate contact with igneous masses upheaved from the interior, they are frequently and deeply modified. Their texture becomes crystalline, as if the materials composing them had undergone fusion, or, at least, as if they had been sufficiently softened to allow their molecules to aggregate in the form of crystals. Rocks thus modified are called metamorphic. The upheaval of rocks must have remarkably changed the rela- tive size and shape of the continents and seas which existed at the 354 GEOLOGY. time of its occurrence. It may have entirely changed the direc- tion of the marine currents which transported the sedimentary matter; and the new strata deposited horizontally on the old, more or less altered from their original position, are often com- posed of materials of a very different nature. The upheaval of older strata is well defined only in the vicinity of the upheaving igneous matter. At a short distance, the same strata may be horizontal, and present, consequently, a stratification more conformable to the newer strata. Every sudden change in the composition and nature of the two superimposed layers, even in conformable stratification, must have coincided with some revolution occurring on the surface of the globe, which changed the direction of the marine currents. But this revolution may have taken place at a great distance from the location of these strata, and, in that case, exercised no influence over their direction. It is equally conceivable that, in submerged localities where, at a certain period, the waters were sufficiently calm to deposit the substances they held in suspension, these waters might, in conse- quence of one of these revolutions, become greatly agitated, and, far from forming new deposits, carry away even those already formed, and transport them to other localities where the current was more feeble. In this way, excavations in older rocks have been formed (fig. 301). Sometimes, the waters becoming more tranquil, Fig. 301. these cavities have received new deposits, and horizontal strata have formed, filling the basins existing in the former (fig. 302). Fig. 302. It is therefore clear that the order in which the strata are depo- sited enables us to judge of the period at which they were formed, and establishes, as it were, their geological age. GEOLOGY. 355 § 280. The geologist is guided by a character of another order in determining the periods of formation of the strata on which he founds their classification. The majority of the sedimentary rocks contain the remains or bear the imprint of animals and vegetables which lived on the surface of the globe when these rocks were formed. Now, animals and vegetables have under- gone, at these various geological epochs, frequent and often well- marked changes; so that the comparative study of animal fossils, known by the name of paleontology, furnishes valuable data to the geologist. § 281. The series of stratified rocks is rarely complete in the same locality, one or more terms being often wanting, and groups, widely separated in the geological scale, being in immediate con- tact. Such gaps, repeated at various stages of the sedimentary formation, prove that the strata are only deposited locally in the parts at that time covered by the waters, and that continents have undergone, at different periods, partial submersions and emersions, before reaching the condition in which we now behold them. When two systems of strata are observed, in any locality, rest- ing on each other unconformably, it may be asserted that an upheaval has taken place between the deposit of the two systems. If the two unconformable systems follow each other immediately in the geological scale, the epoch of upheaval is clearly defined; but if they are widely separated in the scale, in consequence of the absence of intermediate strata, the epoch of upheaval becomes more uncertain. An attentive study of the same upheaval, wher- ever it has exerted its influence, generally points out a portion, at least, of the missing strata in other localities, and the uncer- tainty as to the epoch of the upheaval is confined to narrower limits. The successive upheavals which have modified the primitive form of the globe have produced the various chains of mountains which now exist, so that the epochs of their formation may be referred to the chronological scale furnished by the succession of stratified rocks. In this way their relative age may be determined. Reciprocally, the entire series of sedimentary strata may be sub- divided into several groups, each of which is separated from that which precedes and that which follows it by the phenomena of two mountain chains, which have upheaved the strata existing at the time of their formation ; so that our principal mountain chains form very valuable landmarks accurately dividing the sedimentary strata. § 282. What were the physical causes which produced these successive upheavals, and thus changed the form of continents and seas ? Imagination here finds a vast field over which to wander; and hence there is no lack of theories. Without attempt- ing to unfold the various hypotheses which have been proposed, 356 GEOLOGY. we shall be content to indicate one physical cause which has cer- tainly exerted a great influence over all these revolutions, if it alone did not produce them all. Geodetic measurements have shown the earth to be a spheroid, flattened in the direction of its axis of rotation. This is precisely the form assumed by a fluid globe subjected to a rotary motion ; and it is easy to conceive in this fluid a density varying with the distance from the centre, such that, when influenced by the same rotary movement, the heterogeneous liquid globe would become flattened like the terrestrial globe. This circumstance renders it very probable that our globe was originally in a state of fusion, caused by a very high temperature; that the temperature fell gradually, in consequence of the radiation of heat into space; that the surface naturally cooled more rapidly than the interior, and, at any point of time, the different solidified strata have pre- sented a temperature decreasing from the centre to the circum- ference. If our hypothesis be correct, this condition of things must exist at the present day. And, in fact, all observations hitherto made in mines, or in boring Artesian wells, have shown that, at a certain distance from the surface, the temperature re- mains constant throughout the year, uninfluenced by the variation of the seasons; and that, in starting from this stratum of invari- able temperature, the temperature increases regularly as we descend. The most accurate observations have shown that the increase of temperature is about 1° centigrade for 30 metres, (98|- ft. for 1° C., or ft. for 1° Fahr.) Now, if this increase of temperature continue in the same manner below the strata hitherto accessible to us, the temperature ought to be 1000° C. (1800° F.), at a depth of 30,000 metres (98,430 ft. = 18§ miles), and 2000° C. (3600° F.) at a depth of 60,000 metres, (37 miles); and as the earth’s radius is at least 6,366,200 metres in length, at a depth less than of the earth’s radius, the temperature ought to be 2000° C. (3600° F.)—sufficient completely to fuse all the substances composing the superficial crust of the earth. The internal mass, being fluid, may present nearly everywhere the same temperature. We shall not attach to the numbers just given a value they do not deserve, but regard them as only a probable approximation, sufficient to give great probability to the hypo- thesis advanced. While the temperature of the surface of the earth was very high, the sea-water, and a portion of the substances composing the secondary formations, were diffused through the atmosphere in a gaseous state. But when the surface had cooled sufficiently to allow the water of the atmosphere to remain on it, seas were formed, whose agitated waters broke down the primitive rocks, and drifted their detritus to a greater or less distance, to deposit them, in the form of stratified layers, in localities where the cur- GEOLOGY. 357 rent was less rapid. This was the origin of the first stratified rocks, which were necessarily deposited in nearly horizontal layers. The earth continuing to cool, and consequently to contract, the external solid crust which was formed while the whole mass occu- pied a larger space, being no longer supported on all sides, split in directions where it found the least resistance. The fluid matter of the interior, escaping through the fissures, produced in their vicinity linear upheavals of the layers already formed, and, by following the direction of the fissures, formed the mountain chains, the sides of which are flanked by the edges of the strata, and through the apices of which the fluid matters of the interior fre- quently escape. If the sides of the chain are still under water, new sedimentary deposits will form, but their horizontal layers will not be parallel to those previously deposited, at least where the latter have been affected by the upheavals. At such points, the stratification of the two systems of strata will therefore be unconformable. Subsequently to the deposition of the newer strata, the globe has again split, most frequently in another direction, and has effected an arrangement of the first two systems of strata in a di- rection differing from their former ; and, if still newer strata were superimposed, we should again have unconformable stratification in the vicinity of the new upheaval. We know not Iioav animals and vegetables were developed on the surface of the globe ; but it is evident that no living being existed on the earth, except when the temperature was sufficiently low, and it cannot, therefore, be surprising that their remains are not found in the first sedimentary deposits. They appear only at a later period. It may also be imagined that the great revolutions in the surface of the globe occasioned by the upheaval of a chain of mountains must have instantaneously destroyed the beings ex- isting upon them, and buried their debris among the sedimentary deposits. Equilibrium restored after some time, a new reign of tranquillity began; life reappeared, but under other influences; new species peopled the continents and the seas, and new sedi- ments were again deposited on the line of new shores. This reign of tranquillity was closed by a new catastrophe, which was itself followed by a new period of calm. But, as new animal or vege- table species replaced those which disappeared in these great revo- lutions, their forms became modified, their organization developed and perfected, and creation, generally more simple in the older rocks, ascended gradually to man, of whom no remains are found in any sedimentary strata properly so called, and who, placed on earth at a comparatively recent period, when things were nearly in the state we now behold them, appears to have witnessed only local and more limited revolutions, the traces of which are still 358 GEOLOGY. visible on the surface of the earth, and the remembrance of which lives in the annals of all nations. § 283. Geologists give the name of rock to every agglomera- tion of mineral substances, whether it be hard and firm, as the granites, sandstones, and limestones, or loose, as the sands. The name of formation is given to every system of superimposed rocks in which a certain analogy of structure is recognised, and is chiefly applied to a collection of rocks forming one of the great geological subdivisions. The different rocks constituting the external crust of the earth were first divided into two great classes, primary and secondary ; the primary composed of the non-stratified rocks, the secondary comprising all the sedimentary rocks. The last were then sub- divided into the transition, the secondary, properly so called, and the tertiary. The name transition rocks was given to the lower stratified layers, which often contain crystalline minerals; the more modern stratified layers were called tertiary rocks, and the appellation of secondary rocks wras assigned to the intermediate layers. But the limits separating the various formations not being accurately defined, each one fixed them at pleasure, and great confusion ensued. Geologists now divide stratified rocks into a certain number of groups, the formations' of which are separated by the upheavals which gave birth to our principal mountain chains, and wThich are distinguished from each other by the unconformable stratification of their layers in the vicinity of the upheavals. They have thus made 14 groups of strata, which will be presently enumerated. PRINCIPAL KINDS OF ROCK. § 284. Primary rocks are formed by the agglomeration of dif- ferent crystallized minerals, the most abundant of which are, quartz, feldspar, mica, hornblende, augite, and chrysolite. Quartz is silicic acid. Feldspar is composed of the silicates of alumina, lime, potassa, or soda; mica of the silicates of alumina, potassa, lime, and the oxide of iron. Hornblende, augite, and chrysolite are formed by the silicates of alumina, lime, and the protoxide of iron. Granite, which constitutes the greater portion of the primary formation, is formed by the aggregation of three minerals, feld- spar, mica, and quartz. It presents various shades of colour, owing to the presence of a small quantity of oxide of iron or man- ganese. The proportion of these three minerals varies in every granite. When the feldspar greatly predominates, the rock is called porphyroidal granite. The porphyrys are granites in which the quartz and mica are entirely wanting, and are composed of a feldspathic paste, with imbedded crystals of feldspar. GEOLOGY. 359 The plates of mica scattered through the granite sometimes lie parallel to the same plane, giving the rock a slaty or band-like appearance, when it is called gneiss. The trachytes are volcanic products, of ancient date, and which do not appear to have been always fluid, for they frequently arise from the bosom of the earth in a pasty condition, and form rounded mountains. At other times, they are extended over a horizontal plane, in the form of thick layers. The paste of the trachytes is feldspar, containing many crystals of feldspar, often of large size, and presenting well-marked crystalline faces. The basalts are the result of volcanic eruptions of more modern date than the trachytes. They are composed of augite (silicate of lime, magnesia, and iron), and of labradorite (a species of feldspar, with a base of alumina, lime, and soda). These crystals, being extremely delicate, give the rock a compact appearance. Basalt sometimes pierces the sedimentary strata, spreading over their surface in horizontal layers, as shown in fig. 303, which repre- sents a section of Mount Meissner, in Hessia. Having pierced the secondary strata, in the form of a nearly vertical column BB, it has spread itself over the top of the mountain. Secondary rocks are deeply modified by the contact, or in the vicinity of basalt. Thus, in the stratum c, formed of a tertiary combustible, brown coal, the latter is changed into coke in the neighbourhood of the basalt. Basalts ordinarily form gigantic prisms, joined together, and presenting an appearance of regularity, which is owing to their splitting during the process of cooling. This arrangement in pris- matic columns gives to basalt, where it is exposed to view, a peculiar appearance, such as the columns (fig. 304) of the famous cave of Fingal, in the island of Staffa, north of Scotland. The term lava has been applied to the fluid mineral substances ejected by our modern volcanoes, and spreading in thin layers over their sides. The name slate, or schist, has been assigned to rocks presenting a foliated texture. Pudding-stones are rocks formed by an aggregation of rounded pebbles, imbedded in siliceous cement; they are often of extreme firmness and hardness. Sands are formed by small particles of disaggregated quartz. Fig. 303. 360 GEOLOGY. Fig. 304. When the grains of sand are united together by a quartzose cement, the rock takes the name of sandstone. Sandstones, some- times colourless, are often tinged red or gray by the presence of certain metallic oxides. Calcareous rocks, or limestones, are composed of carbonate of lime, and vary according to the state of aggregation of the sub- stance ; being crystallized in marble, compact and often very hard in the Jura limestone, and friable in chalk. Clay is principally composed of silicate of alumina, almost always associated, however, with a small quantity of silicate of potassa. Argillaceous rocks are characterized by being impervious to water, and retain all the waters which pass through superincum- bent rocks, forming large aqueous reservoirs on their surface. Clays are often mixed with considerable proportions of car- bonate of lime, and are then called marls. Anhydrous sulphate of lime, or anhydrite, and the hydrated sulphate of lime, or gypsum, sometimes form actual strata in secondary formations, while at other times they only form a kind of flattened lenses in the midst of other formations. § 285. Secondary rocks are sometimes formed at the expense of the primary, which have been broken down and drifted by wa- ter ; but, at the same time, the substances composing them have been chemically altered by the joint action of Avater and air. Thus, feldspar becomes changed into clay and into alkaline salts; mica produces clay and calcareous salts; quartz furnishes sands and sandstone. The presence of organized beings, vegetable or animal, must necessarily have exerted a great influence over these chemical changes. The carbon which Ave find in combustible mi- nerals, in the bosom of the earth, probably existed in the atmo- sphere, in the state of carbonic acid, Avhich vegetables decomposed, CLASSIFICATION OF FORMATIONS. 361 as they now do, assimilating to themselves the carbon, and disen- gaging the oxygen. Through animals, calcareous salts .have been principally changed into carbonate of lime; and such is pro- bably the origin of the calcareous layers which abound in various formations. They have been formed by the detritus of shells, often entirely disaggregated; while at other times the shells have preserved their original forms, so that certain calcareous rocks are actual collections of shells, the species of which can be determined at this day with perfect accuracy. In modern times, several silicious rocks have been ascertained to be entirely formed of the silicious skeletons of certain micro- scopic insects. § 286. The following table exhibits the series of divisions of the formations now admitted by geologists, with the principal rocks which compose them, and the system of upheaval which charac- terizes them. They are arranged in the descending order, that is, commencing with the most modern. GEOLOGICAL DIVISION OF THE FORMATIONS. first group.—Contemporaneous or Recent Formation. ' Alluvial deposits filling the valleys of rivers. Modern volcanoes, both extinct and burning. The great volcanoes of the Andes arose during this epoch. RECENT FORMATIONS. second group.— Upper Tertiary. (Pliocene and Miocene.) Strata of ancient sand and alluvium; boulders, drift; tufa, (breccia,) con- taining fossil bones. The eruption of the majority of trachytes and basalts correspond to this epoch. System of the prin- cipal chain of the Alps TERTIARY FORMATIONS. third group.—Middle Tertiary. System of the Western Alps Fresh-water limestone with burrstones; sometimes containing lignite Sandstone of Fontainebleau Upper Eocene. fourth group.—Lower Tertiary. System of the islands of Corsica and Sar- < dinia Marls with gypsum;fossil re- mains of the mammiferae. Coarse limestone. Middle Eocene. Plastic clay with lignite. Lower Eocene. 362 GEOLOGY. fifth group.— Upper Cretaceous. System of the chains of the Pyrenees and - Apennines ' Extensive limestone stratum, called chalk, with layers of silex inter- posed. SIXTH group.—Loiver Cretaceous. System of Mount Yiso. Tufaceous chalk of Touraine. Sand, or sandstone, generally green, and hence called green sand. Ferruginous sands. seventh group.—Oolitic or Jurassic. Calcareous strata, more or less com- pact and marly, alternating with layers of clay. They are divided into several sub-groups, the upper bearing the name of oolite, and the lower being called lias. Sandstone below the lias. System of the Cote d’Or SECONDARY FORMATIONS, eighth group.—Trias. Marls of various colours, called va- riegated marls, (Keuper,) often containing masses of gypsum and rock salt. Limestone, very fossiliferous, and hence called muschelkalk. Variously coloured sandstone, termed variegated sandstone, (Buntersand- stein, gres bigarrd.) System of the Thurin gerwald ninth group.—Sandstone of the Vosges. System of the Rhine. Conglomerate and sandstone. tenth group.—Permian. Stratum of limestone mixed with slate and called zechstein. Stra- tum of conglomerate and sand- stone, termed new red sandstone, (Rothtodtliegendes.) System of the Low Countries and of Wales METALLIC VEINS. 363 eleventh group.—Carboniferous. System of the North of England Sandstone, slates with seams of coal and carbonated iron, (clay-iron- stone.) Carboniferous or mountain limestone with seams of coal. TWELFTH group.—Devonian. TRANSITION EARTHS. System of the balloons of the Vosges, and the hills of the fo- rest of Normandy... Heavy beds of sandstone, called old red sandstone, containing small seams of anthracite. thirteenth group.—Silurian. Limestone, roofing-slate, coarse- grained sandstone, called gray- wacJc. fourteenth group.—Cambrian. Compact limestone, argillaceous shale or slate. These rocks have often a crystalline texture. System of Westmore- land and Hundsruck < in Scotland fifteenth group.—Primary Rocks. PRIMITIVE EARTHS. Granite and gneiss forming the prin- cipal base of the interior of the globe, accessible to our means of observation. METALLIC VEINS. § 28T. It has been shown that the gradual cooling of the globe must have produced a great number of fissures in the solidified crust, which were not always sufficiently large to allow the con- tained fluids to reach the surface. The strata have frequently only be£n split in different directions, and the rents subsequently filled with very different substances, which have reached them either in the state of vapour arising from the interior, or in solu- tion in water coming from the surface or the interior. These fissures have received the names of veins, feeders, or lodes. They often contain only earthy matters, as carbonate of lime, sulphate of baryta, quartz; and these possess but little interest. They are, however, frequently filled, either wholly or in part, with metallic substances, when they become of great importance. Me- tallic veins are generally found in primary rocks, or the most an- cient stratified formations, of which the transition contains the principal veins which have been worked. A metallic lode is rarely found isolated; several being most frequently observed in the same locality, when they pre- sent a nearly parallel direction. Fig. 305 exhibits a transverse section of one of these systems of metallic veins. The simi- larity of the mineral contents of the lodes in the same sys- tem demonstrate their common origin. One system is often traversed by another (fig. 305), affording very different mineral matter from the for- mer ; the latter are called intersecting lodes. A lode is rarely found filled with metalliferous minerals, which most frequently form a net- work abedefg, more or less irregular, amidst a stony crystalline substance, filling the vein (fig. 306). The thickness of a metalliferous fila- ment varies at different points of the lode, being sometimes considerable, at others very small,and sometimes entirely disappearing. The stony minerals which separate the metalli- ferous substance from the sides of the rock constitute the gangue or matrix of the ore. When a metallic vein reaches the surface, it manifests itself either by a line of bold re- lief, when the substance which forms it is harder than the adjoining rock; or by a line of depression, in the contrary case. The head or levels of a metallic vein are often modified by the chemical changes which have affected the substances composing it. § 288. Numerous cavities have been formed in certain stratified formations, proba- bly by the dissolving action of subterranean waters. They are found in all parts of the formation, and have been generally filled at a later period with new substances very different from the surrounding rock. They are called deposits. 364 GEOLOGY. Fig. 305, Fig. 306. Fig. 307. PHYSICAL PROPERTIES. 365 Thus deposits of rock salt are found in the muschelkalk and the variegated marls (fig. 307). Fig. 308. Fig. 308 represents deposits C, C of carbonate of zinc which have been formed at the upper part of a stratum of transition limestone. § 289. Before entering upon the study of each particular metal, we shall succinctly define the general physical and chemical pro- perties of the metals and of their chief compounds. This will facilitate our progress when we arrive at the special history of each metal. PHYSICAL PROPERTIES OF THE METALS. § 290. The physical properties of the metals most deserving of study are, their opacity, lustre, colour, crystallization, malleability and ductility, tenacity, their power of conduction and capacity . for heat. § 291. Opacity.—Metals are very opaque, and do not allow the transmission of light even when reduced to exceedingly thin laminae. Gold, however, in the form of gold-leaf, as produced by the goldbeater, admits of the passage of a considerable quantity of light of a beautiful green colour. The peculiar physical quali- ties of this light show that it has really passed through the metal, and not merely through the small fissures made in the leaf by beating. § 292. Lustre.—Metals aggregated by hammering or fusion, present a peculiar lustre, familiar to every one, but very difficult of definition. When reduced to very fine powder, or in the con- dition of chemical precipitates, their lustre disappears, but it im- mediately reappears if the substance be rubbed with a burnisher, or any hard and polished body. § 293. Colour.—The majority of metals are of a more or less gray colour, in the form of powder, and become whiter when in masses and polished. Some, however, possess well-marked colours: thus, copper and titanium are red; gold is yellow. The alloys formed by white or gray metals are themselves white or gray. Those composed of a coloured metal are tinged by its hue, when 366 METALS. it exists in any quantity. Thus, an alloy of f of copper and J of zinc, called brass, is of a beautiful yellow colour; and an alloy of 90 parts of copper, and 10 parts of tin, called bronze, is also yel- low. The metal of the reflectors of a telescope is made of 67 parts of copper, and 33 of tin, and are white. The white metals reflect the various simple rays of the spectrum in proportions which are nearly the same as those in which these rays compose white light. But, as these proportions are not ex- actly the same, except in white light, and vary with the incidence of the luminous rays, such metals present a peculiar tinge, which may be ascertained by delicate experiment. Coloured metals reflect certain simple rays of the spectrum more copiously than the others, and the proportion of simple rays re- flected varying with the angle of incidence, it follows that the shades of these metals change, according as they are seen more or less obliquely. All the metals reflect in the same proportion the various simple rays which fall on their surface at very small angles of incidence, so that they all appear white when sighted along their surface; but, as their reflecting power for different simple rays varies more and more as the angle of incidence increases, they then become evidently coloured. Their discoloration will be more marked, if, instead of causing the ray of light to be reflected once from their surface, it be reflected several times; in which case, those which generally appear colourless become very sensibly tinged. In order to make this experiment, two mirrors, formed of the metal, are set parallel to each other, and a ray of light observed, which is reflected several times successively from their surfaces, at an . angle approaching 90°. After a single normal reflection, copper presents an orange-red colour, but of the reflected light is white light, so that the hue appears very faint. After 10 successive reflections, the copper as- sumes an intense red colour, which is mixed with only of white light. Bell-metal has a pale orange-yellow tinge, but after 10 succes- sive reflections, the light is of an intense red, and only contains of white light. The light reflected once from the surface of polished brass is evidently yellow, but after 10 reflections, it becomes orange, but is still mixed with A of white light. Silver appears perfectly white when the light is reflected only once from its surface; but, after 10 reflections the light assumes a marked red tinge, although feeble, because it is mixed with $ of white : its hue nearly resembles that of bell-metal after a single normal reflection. Zinc is white after one reflection, but it becomes indigo-blue after 10: the hue is, however, always feeble, because A of white light remain. PHYSICAL PROPERTIES. 367 Steel, after 10 reflections, becomes of a violet hue, but always feeble, because it is mixed with 0.97 of white light. The metal of mirrors is white after one reflection, but becomes evidently red after 10. It is important to be acquainted with the modifications of hue, experienced by light when reflected several times from the sur- face of metals, inasmuch as they explain the various shades of colour seen on the inside of a polished and shallow metallic vessel. The hue assumed by white light when reflected several times from the surface of polished metals, also enables us to assume with a good deal of certainty the colour which they would present by transmitted light, if they could be made sufficiently thin to become transparent. This colour would necessarily be the com- plement of that which would prevail in the light when reflected a number of times from their surface. Thus, light reflected 10 times from the surface of polished gold, is of a beautiful red colour. The complemental colour of red is green; and, in fact, very thin gold-leaf exhibits a bright green colour by transmitted light. § 294. Crystallization of metals.—All the metals are susceptible of crystallization, but it is not easy to place them always under conditions in which they will assume regular forms. Those found in the native state are often well crystallized; thus, gold, silver, and copper are frequently met with in this form. Some metals crystallize when allowed to cool slowly after fusion. The crystals may be isolated by the process described for sulphur (§ 125). Bismuth, in this way, affords very regular crystals. Antimony, lead, and tin also crystallize in this manner, but not so readily. The crystallization of the metal sometimes occurs in the midst of a solid mass, when the latter is maintained for some time at a high temperature. Thus, we frequently find crystals in the in- terior of the large masses of iron which enter into the construction of our metallurgic furnaces. Many metals crystallize when slowly separated from a solution, principally under the influence of feeble electro-chemical forces. If we plunge, for instance, into a solution of sulphate of copper, two plates of copper communicating with the two poles of a bat- tery, the plates of the negative pole become covered with crystals of metallic copper, whilst that of the positive pole gradually dis- solves. Sometimes the crystals are so small as only to be dis- cernible by the microscope; at others, they are larger. The tenacity of metals is greatly influenced by their crystalline structure; for when it is strongly marked, their tenacity is gene- rally very feeble, and they are, most frequently, brittle. Almost all metals which have cooled slowly after fusion exhibit, either in their interior or on their surface, marks of crystalliza- 368 METALS. tion; but their texture is much modified by the manipulations they have undergone. When forged or rolled, their molecules are made to assume forced positions, whereby their physical proper- ties are remarkably modified, and often to great advantage in mechanical applications. The most common crystalline form of the metals is the regular octahedron or cube ; but antimony crystallizes in a rhombohedron. We shall indicate the crystalline form of each metal under its ap- propriate head. § 295. Malleability and ductility.—When metals are subjected to blows with the hammer, some flatten out into sheets, and others fly into fragments; the former are called malleable, the latter brittle metals. Metals are reduced into sheets, either by beating with a ham- mer, or by passing them through rollers. The rollers consist of two metallic cylinders, placed horizontally, one above the other, and are made to revolve with equal rapidity, in the directions indicated by the arrows in fig. 309. The cylinders may be set at various distances apart from each other, but, once fixed, they maintain a uniform dis- tance, which must be somewhat less than the thick- ness of the metallic plate to be rolled. The plate, after being bevelled on one of its edges, is inserted between the cylinders, and being obliged to follow their motion, is extended so as to be enabled to pass through. It is again passed through cylinders now more closely set, and thus reduced to any given degree of thin- ness. Some metals can be rolled when cold; others require a high de- gree of temperature. During this forced flattening of the sheet, the metal undergoes a remarkable change in its molecular arrangement, which fre- quently alters greatly its physical properties, and especially its malleability. It becomes more hard and brittle, and if the rolling be continued, the sheets would inevitably tear. The metal is then said to be hammer-hardened, but its original ductility is restored by annealing it, wdiich consists in heating it, and allowing it to cool slowly. Under the influence of heat, the molecules assume their respective normal positions, and the sheets may again be passed through the rollers. The malleability and ductility of only such metals have been determined which have been obtained in a state of compactness and purity; for the presence of any foreign body, even in the smallest quantity, singularly alters their malleability. The following are those whose malleability and ductility have been well determined:— Fig. 309. 369 PHYSICAL PROPERTIES. Silver, Cadmium, Cobalt, Copper, Tin, Iron, Mercury, Nickel, Gold, Palladium, Platinum, Lead, Potassium, Sodium, Zinc. Gold and silver are exceedingly malleable, as is shown by the extremely thin leaves manufactured by the goldbeater, which are so thin as to require more than 10,000 to form the thickness of a millimetre, (250,000 to 1 inch.) § 296. Wire-drawing.—Certain metals may be drawn out into very fine wire. The malleable metals are the only ones which possess this property; but they must have, in addition, a certain tenacity, in order to prevent them from breaking. The wire-plate consists of a steel plate pierced with circular holes of various diameters, the edges of which are sharpened. The metallic rod intended to be drawn is made rather larger than the hole, No. 1 of the plate, and one of its ends is pointed so as to allow it to pass through hole No. 1, when this end is seized with a pincers, and the whole rod drawn forcibly through the hole. It is necessarily elongated and lessened in size. It is then passed successively through holes No. 2, 3, 4, etc., the diameters of which gradually decrease. Metals become hardened in this operation, as in rolling, and it is necessary to anneal them from time to time, to restore their original ductility. Pure metals and certain alloys can thus be drawn out into very fine wire, but not of extreme tenuity; for at a certain point they no longer possess sufficient tenacity, and break under the traction necessary to draw them through the wire-plate. Much finer threads, however, can be obtained, by resorting to different con- trivances, a single example of which will be given, by describing a process whereby platinum wire has been made as fine as a spider’s web. A cylinder of silver is bored in the direction of its axis, with a hole 1 or 2 millimetres Qg or A inch) in diameter, into which a platinum wire is inserted, of the same diameter, and then the cylinder drawn through the wire-plate. A very fine silver wire is thus obtained, in the centre of which there is a platinum thread still more delicate. The compound wire is then treated with dilute nitric acid, which dissolves the silver, and leaves the platinum thread untouched. The following table exhibits the order in which metals pass with greatest facility through 370 METALS. The rollers. 1. Gold. 2. Silver. 3. Copper. 4. Tin. 5. Platinum. 6. Lead. 7. Zinc. 8. Iron. 9. Nickel. The wire-plate. 1. Gold. 2. Silver. 3. Platinum. 4. Iron. 5. Nickel. 6. Copper. 7. Zinc. 8. Tin. 9. Lead. The two series will be seen to differ remarkably from each other. § 297. Tenacity.—The tenacity of metals is that property by virtue of which they resist attempts to break them, and varies greatly in different metals. In order to test their tenacity, wire of the different metals is made of the same diameter, by passing them through the same hole in a wire-plate. Equal lengths of these wires are attached to a fixed point, and to the other end is suspended a dish to receive weights, by which the smallest weight which will effect the breakage of the wire, can be ascertained. This weight may then be considered as a measure of their resist- ance to rupture, or their tenacity. Metals are thus proved to possess very different degrees of tenacity. The following table exhibits the smallest weights which have broken a wire of 2 millimetres (0.079, or inch) in dia- meter. It contains only the malleable metals, which are ranged in the order of decreasing tenacity: Kilogr. Lbs. avoir. Iron 250 = 551 Copper 137 = 302 Platinum 125 = 275J Silver 85 = 187J Gold 68 = 150~ Kilogr. Lbs. avoir. Zinc 50 = 110^ Nickel 48 = 99£ Tin 16 = 35£ Lead 12 = 26J The tenacity of metals has a great influence upon their ap- plication in the arts; and it frequently varies considerably in the same metal, according to its purity and mode of preparation. When a metallic wire has been extended by the addition of successive weights, it is elongated in proportion to the weight it supports; and if the weights are gradually removed, the wire re- covers the length which it formerly had under the same load. But this proposition is true for each wire only tofca certain amount of weight, beyond which the wire retains a permanent elongation after the removal of the load. It is then said to have exceeded the limit of its normal elasticity. This maximum weight is often much less than that which breaks the wire. In building, therefore, a wire, or a metallic beam, should never be loaded to ACTION OP OXYGEN ON THE METALS. 371 this point; for it would soon change under the prolonged effort of traction, and shortly break under a lighter load than it origin- ally would have easily supported. § 298. Conduction of Heat.—Metals conduct hoat more or less readily; and the study of this property is important in some of their applications, as, for example, in the construction of evapo- rators and steam-boilers. The quantity of liquid evaporated in a given time, depends necessarily on the conducting power of the metal of which the vessel is made; for, with equal thickness, similar vessels, formed of different metals, will transmit, in the same time, quantities of heat in proportion to their powers of conduction. The following table contains the metals arranged in the order of their decreasing conduction : Gold 200 Silver 195 Copper 180 Iron 75 Zinc 73 Tin 61 Lead 36 § 299. Capacity of Heat.—Very different degrees of heat are required to heat equal weights of different metals to the same number of degrees. Thus, the quantity of heat necessary to heat 1 kilogramme of water from 32° to 212°, being represented by 1.000, that which will effect the same elevation of temperature in 1 kilogramme of the various metals, is represented by the follow- ing numbers: Iron 0.1138 Nickel 0.1086 Cobalt 0.1070 Zinc 0.0955 Copper 0.0952 Palladium 0.0593 Silver 0.0570 Cadmium 0.0567 Tin 0.0562 Antimony 0.0506 Platinum 0.0324 Gold 0.0324 Lead 0.0314 Bismuth 0.0308 CHEMICAL PROPERTIES OF THE METALS. § 300. We shall now make some remarks on the manner in which the metals behave with the metalloids, and on the general properties of the compounds which they form with those bodies. Action of Oxygen on the Metals. § 301. Although all the metals have been obtained combined with oxygen, their affinities for it are very different. Some, such as potassium and sodium, combine with it directly, even at the lowest temperatures; others, as gold and platinum, have so feeble 372 CHEMICAL PROPERTIES OF THE METALS. an affinity for it, that they do not combine directly with it under any circumstances, and their oxides are obtained only by indirect methods. The former retain oxygen at the highest temperatures, while the latter part with it readily when their oxides are heated. The affinity of the metals for oxygen may be ascertained, 1st. From the manner of their behaviour to oxygen gas at va- rious temperatures. 2dly. From the greater or less facility with which their oxides are restored to the metallic state. 3dly. From the decomposing action which they exert upon the same oxide under various circumstances; water being the oxide usually made use of. Certain metals decompose water, even at the temperature of 32°; others decompose it only at a tempera- ture greater than 122° or 140° ; some require a temperature of 212°; others react on the vapour of water only at a red-heat, or at a still higher temperature; and, lastly, some do not decom- pose water at any degree of heat attainable by our laboratory furnaces. 4thly. From their decomposing influence upon water in the pre- sence of powerful acids. Many metals decompose water, in the cold, in the presence of sulphuric acid; while others do not even when the temperature is elevated. This property does not depend alone upon the greater or less affinity of the metals for oxygen; but depends especially on the basic affinity of the metallic oxide for the acid (§ 69). The metals have, therefore, been divided into six sections, based upon the above properties. Section 1.—Those which absorb oxygen at all temperatures, even at the highest, and decompose water even at the lowest tem- peratures, producing a copious evolution of hydrogen gas. They are— Potassium, Sodium, Lithium, Barium, Strontium, Calcium. The first three are called alkaline metals; the last three, alkalino- eartliy metals. Section 2.—Those which absorb oxygen at the highest tempera- tures, and whose oxides are indecomposible by heat alone: they do not sensibly decompose water at very low temperatures, but readily above 122°. They are— Magnesium, Manganese, Aluminum; to which may, probably, be added the following metals, whose de- composing action on water has not been yet studied with sufficient care: CLASSIFICATION OF THE METALS. 373 Glucinum, Zirconium, Yttrium, Thorium, Cerium, Lanthanium, Didymium, Erbium, Terbium. Section 3.—Those which absorb oxygen at a red-heat, but do not yield it up again by heat alone, which decompose water only at a temperature superior to 212°, but below a red-heat, and decompose cold water in the presence of powerful acids. They are— Iron, Nickel, Cobalt, Chromium, Vanadium, Zinc, Cadmium, Uranium. The temperature at which these metals combine with oxygen, and that at which they decompose water, depends greatly on their state of division. Aggregated iron, even in the state of filings, combines with dry oxygen only at a dull red-heat; while the same metal very finely divided, which can be done by reducing the oxides of iron by hydrogen gas at the lowest temperature possible, takes fire when thrown into the air, and oxidizes, consequently, at ordinary temperatures. Aggregated iron decomposes the vapour of water only at a red-heat, while pulverulent iron decomposes it at a temperature of about 392°. Section 4.—Those which absorb oxygen at a red-heat, and whose oxides are irreducible by heat alone. They readily decompose the vapour of water at a red-heat, but do not decompose water in the presence of powerful acids. This is owing to the fact that their oxides are but feeble bases; while they form, on the con- trary, with oxygen, bodies which exhibit strong acid properties with reference to energetic bases. Hence, the greater part of them decompose water in the presence of potassa with an evolu- tion of hydrogen gas. This 4th section comprises— Tungsten, Molybdenum, Osmium, Tantalum, Titanium, Tin, Antimony; to which we may probably add— Niobium, Ilmenium, Pelopium. Section 5.—Those which absorb oxygen at a red-heat, whose oxides are not decomposed by heat alone ; which decompose water only at a very elevated temperature, and then very feebly. They 374 THE METALS. do not decompose water either in the presence of strong acids or of powerful bases. They are— Copper, Lead, Bismuth. Section 6.—Those whose oxides are reducible by heat alone, at a more or less elevated temperature, and which do not decompose water under any circumstances. They are— Mercury, Silver, Rhodium, Iridium, Palladium, Platinum, Ruthenium, Gold. § 302. It is useful to remark, that all the metals whose oxides are irreducible by heat alone, can decompose water at a higher or lower temperature; which is due to the fact that water itself is separated into its two elements at an extremely elevated tempera- ture. If a small ball of .platinum, affixed to the end of a wire of the same metal be heated to whiteness by the hydro-oxygen blow- pipe, and plunged rapidly into water, small bubbles of gas, formed of a mixture of hydrogen and oxygen, are disengaged. The water has been therefore decomposed by heat alone, for the metal has seized on neither of its constituent gases. A similar decomposition takes place when a platinum wire immersed in water is heated in- tensely by passing through it the electric current of a powerful battery. § 303. The direct combination of metal with oxygen is a true combustion with disengagement of heat; and when the combina- tion is rapidly effected, the temperature rises sufficiently high to render the substance incandescent. The combustion is more active when the metal is finely divided, because it then presents a greater surface to the action of oxygen; but if the metal be in mass, and the oxide do not fuse at the temperature at which oxidation takes place, the combustion is suddenly arrested, because the metal be- comes covered with a coating of oxide, which defends it from further contact with the oxygen. Thus, finely divided copper pre- viously heated to dull redness, burns readily in oxygen, and is wholly changed into an oxide, while a sheet of copper, under simi- lar circumstances, is only covered with a coating of oxide. Iron, heated to redness, burns freely in oxygen, even when the metal is in the shape of large wire, because the resulting oxide fuses at the temperature of combustion, and keeps the surface of the metal exposed. When the metal is volatile, it may also burn with great energy, Action of dry Oxygen on the Metals. ACTION OF OXYGEN. 375 and even with flame, if it has been previously heated to a proper degree. Thus, zinc heated to redness in a crucible, burns with a very brilliant white flame. In this case, it is the vapour of zinc which burns; and as the oxide of zinc is fixed, its solid particles, suspended in the flame, become luminous, and add great brilliancy to it. Action of moist Oxygen on the Metals. § 304. Metals which do not combine when cold with dry oxygen, frequently oxidize rapidly when exposed to a damp air. Iron preserves its brilliancy in dry oxygen for an indefinite time, while it changes rapidly in moist air, and becomes covered with an ochreous coat, which is the hydrated sesquioxide of iron. Many other metals belong to the same category; but, in some, the change is only superficial, while in others it continues until the whole of the metal is converted into an oxide. An iron bar exposed to a damp air is completely destroyed by rust, while a sheet of zinc soon becomes covered by a pellicle of oxide which'preserves it from further change. The presence of acid vapours in the air greatly facilitates the oxidation of metals. Iron, which remains unaltered in dry oxygen, and even in water deprived of its air by boiling, soon changes when in contact with oxygen and water at the same time; for it then meets with oxygen dissolved in the water, that is, under the most favourable conditions for its combination. Moreover, iron has a certain basic affinity for water, which again facilitates the formation of this oxide, according to the principle laid down (§ 69). For the same reason, iron and zinc, which alone do not decompose cold water, decompose it readily in presence of powerful acids, as if the presence of the acid had increased their affinity for oxygen. The presence of acid vapours in the air greatly facilitates the oxi- dation of a metal, for they increase its affinity for oxygen to a greater degree than water, which only acts the part of a feeble acid. Those metals some of whose oxides play the part of acids with reference to energetic bases, oxidize more rapidly in the air when moistened with an alkaline solution, or in the midst of a moist atmosphere containing ammoniacal vapours. § 305. It is frequently observed that when a certain quantity of oxide has formed on the surface of a metal, its oxidation becomes much more rapid, as if the presence of the oxide increased the affinity of the metal for oxygen. This peculiarity is very evident in iron, and the following experiment demonstrates it clearly. If moistened iron filings be exposed to the air, oxidation goes on very slowly at first, but is soon accelerated, and the iron rusts rapidly. At the same time the fetid odour of hydrogen gas is observed, which occurs when ordinary iron is dissolved in dilute 376 METALS. sulphuric acid. In fact, a sufficient quantity of hydrogen is dis- engaged to allow of its collection after some time, if the experi- ment be made in a suitable apparatus. The oxidation of the metal is, at first, effected by the absorp- tion of the oxygen of the air dissolved by the water which moistens the filings; but the coat of oxide which covers the metal soon forms a voltaic circle, in which iron is the electropositive element. Iron itself is electropositive as regards oxygen; and if it forms the electropositive element of a pile, it becomes still more electro- positive than it naturally is, its affinity for oxygen is increased, and experiment proves that this affinity may be sufficiently great to decompose water at the ordinary temperature. If, on the contrary, a body which becomes the electropositive element of a voltaic circle be brought into contact with iron, the latter, becoming less electropositive than in its isolated state, loses some of its affinity for oxygen: it has become less oxidizable, and may be preserved from oxidation under circumstances in which this would inevitably have ensued had it been isolated. Advan- tage has been taken of this property, in the arts, to render objects made of iron less changeable in the air, by covering them with a thin coating of zinc, which becomes the electropositive element of the circle, and preserves the iron from oxidation. The zinc, on the contrary, oxidizes rapidly; but its oxidation is only super- ficial, for the small pellicle of oxide developed on the surface forms an impervious varnish which preserves the inner layers. Iron, thus protected by a coating of zinc, is called galvanized iron. The same principle has been applied to prevent the oxidation of other metals, such as the copper used in sheathing ships. Unfor- tunately, it has been found that shells will then adhere in greater numbers to the ship’s bottom, and her sailing powers are lessened in consequence of the increased friction. Action of Sulphur on the Metals. § 306. All the metals combine directly with sulphur, when heated with it, or when it is passed in the state of vapour over the heated metal. Some, such as copper, burn in the vapour of sulphur with bril- liancy. Others combine with it, even at ordinary temperatures, if water be present. A mixture of iron-filings and flowers of sulphur slightly moistened, soon disengages a considerable degree of heat, owing to the combination of the iron with the sulphur. Action of Chlorine on the Metals. § 307. Chlorine acts on the metals still more powerfully than oxygen, and converts them readily and entirely into chlorides. ALLOYS. 377 The majority of them combine with chlorine even in the cold. In some, the combination is so energetic that the temperature rises to ignition, and many of them, when pulverized, take fire when thrown into a bottle containing gaseous chlorine. Action of Bromine and Iodine on the Metals. § 308. The action of bromine and iodine on the metals gene- rally resembles that of chlorine; but the affinities are more feeble. Action of Phosphorus on the Metals. § 309. The metals of the first section combine easily with phos- phorus when heated with it; but those of the other sections do not, because the phosphorus volatilizes before the temperature is sufficiently elevated for reaction to take place. Some metals of the third and fifth sections may combine with a certain quantity of phosphorus, when heated to a very high temperature in its vapour. Action of Arsenic on the Metals. § 310. Arsenic combines with the metals much more readily than phosphorus, and several arseniurets are directly obtained by heating a powdered mixture of the metal and arsenic. Action of Boron, Silicium, and Carbon on the Metals. § 311. Some of the metals can combine directly with boron, sili- cium, and carbon: we shall, as we proceed, point out several of these compounds. COMBINATIONS OF THE METALS WITH EACH OTHER, OR ALLOYS. § 312. The majority of the metals can combine with each other, forming alloys endued with metallic properties which partake at once of the nature of both combined metals. By alloying metals with each other, we create, so to speak, new metals, possessing special properties, and more suitable for certain purposes in the arts than the pure metals. The metals used in a pure state in the arts are, Iron, Copper, Zinc, Lead, Tin, Silver, Gold, Platinum, Mercury. Of these, platinum and iron are the only ones exclusively em- ployed in a state of purity. The others are often used alone, but 378 METALS. are more frequently alloyed with each other, or with some other metals, such as antimony and bismuth, which last are never used alone, on account of their brittleness. Copper is a very valuable metal, easily worked by the hammer, but destitute of a great degree of hardness. This quality can be much increased, without greatly diminishing its ductility, by alloying § of copper with J of zinc. An alloy called brass is thus obtained, of an agreeable yellow colour, and admitting a large number of applications. But brass thus made is not easily filed, as it sticks to the file and clogs it; but the inconvenience is remedied by adding to the alloy 2 or 3 per cent, of lead or tin. § 313. For artillery, a metal is required which is hard without being brittle, and which can be cast, and worked in a turning- lathe. Pure copper partly fulfils this condition, but it is too soft, and before the ball leaves the cannon it rebounds several times in the chamber of the piece, soon forming cavities which impair the accuracy of the aim. An alloy of 90 parts of copper and 10 parts of tin presents more hardness and possesses sufficient tena- city. This alloy, called bronze, is used for cannon and many ornamental objects, such as statues, candelabras, etc. By increas- ing the proportion of tin, we obtain alloys still harder, but also more brittle. The alloy of 20 parts of tin and 80 of copper is extremely hard and sonorous, and is used in the manufacture of clocks, cymbals, and tomtoms. The alloy of 67 of copper and 33 of tin, is of a slightly-yellowish white colour, susceptible of a most brilliant polish, and is used for the reflectors of telescopes. Thus, by alloying two metals in different proportions, alloys are obtained differing greatly from each metal in their physical properties, and capable of very various uses. § 314. For printing-types, a metal is required which must sa- tisfy many conditions. It should be very fusible, for the types are cast; it must take the exact impress of the mould, in order that the characters be well defined; and, lastly, it must possess a certain hardness, without being too brittle, for, if the metal is too soft, the types are crushed in the press, and if too hard and brittle, they break. Iron and copper are not sufficiently fusible. Silver, gold, and platinum melt only at a very high temperature, and are, moreover, too expensive. Zinc, antimony, and bismuth are too brittle. Lead and tin are too soft. But a perfectly suitable alloy is ob- tained by mixing 80 parts of lead and 20 of antimony. § 315. Many metals seem to possess the power of combining with each other in indefinite proportions. But when melted alloys are allowed to cool slowly, they generally separate into several others, presenting a definite composition, and often a crystalline structure. This decomposition of the same homogeneous alloy, and several others which separate more or less perfectly, takes ALLOYS. 379 place sometimes when the alloy is exposed for a long time to a high temperature, although less than that producing its fusion. Examples of this will hereafter be adduced. These separations may be easily recognised in alloys fusible at a low temperature, by observing the fall of a thermometer, the bulb of which dips into a certain quantity of fused alloy which is cooling in the air. If the experiment be made upon melted tin, heated to 120° or 140° above its melting point, it will be observed that the temperature falls at first rapidly, but with a decreasing celerity, because the rapidity of cooling of a substance in the air is nearly in proportion to the excess of its temperature over the surrounding medium. But when the temperature reaches 437°, the thermometer stops suddenly, and remains stationary for a longer or shorter time, according to the mass of metal on which we are operating, and then begins again to fall. The point at which the thermometer stops corresponds to the solidification of the tin. The metal, by solidifying, gives off its latent heat of fusion, which compensates at every instant for the loss of heat effected by radiation and the contact of cold air; and the cooling recommences only after the metal is entirely solidified. The same phenomenon is evinced in all homogeneous bodies, whether simple or compound, the constitution of which does not change while cooling slowly after fusion. But, if the same experiment be made on certain very fusible alloys, and principally on the ternary alloys of lead, tin, and bismuth, which, melting at low temperatures, are very suitable for this kind of observation, several points of stop- page are generally observed during their cooling; sometimes as many as three or four. Each of these stoppages corresponds to the solidification of a particular alloy with definite proportions, which is formed at the expense of the elements of the primary homogeneous alloy, and separates in the form of a crystalline powder. After the separation of one or several of these com- pounds, the substance presents the consistence of a sandy paste; and only becomes completely solid after the crystallization of the alloy which remains fluid last. Thus, although we may fuse the three metals together in inde- finite proportions, and obtain apparently homogeneous alloys by rapid solidification, the metals have a tendency to combine in definite proportions, like all other substances in nature; and de- finite compounds are formed whenever, during slow cooling, the molecules have time to obey their elective affinities. § 316. The point of fusion of an alloy is often less than that of the most fusible metal which enters into its composition. Thus, lead melts at 617° bismuth, “ 509° tin, “ 442° 380 THE METALS. The alloy formed of 5 parts of lead, 3 of tin, and 8 of bismuth, melts at 203°, that is, at a temperature much lower than that of its most fusible component. OF THE METALLIC OXIDES. § 317. The metallic oxides present the most diversified proper- ties. Some are more or less powerful bases, which combine with acids forming well marked salts ; others, on the contrary, play the part of acids, and combine with the powerful bases; lastly, some of them combine neither with acids nor bases. In this point of view, the oxides are generally divided into five classes : 1st. Basic oxides, that is, those which combine readily with the acids, and produce definite, crystallizable salts. The protoxides of potassium, sodium, calcium, iron, lead, etc. etc. are basic oxides. 2dly. Acid oxides, which do not combine, or at least very rarely, with the acids, and which form, on the contrary, well defined salts with powerful bases. Chromic acid Cr03, manganic acid Mn03, stannic acid Sn03, plumbic acid Pb02, antimonic acid SbOs, are true metallic acids, which form crystallizable salts with several powerful bases, particularly with potassa. 3dly. Neutral oxides, which at the same time play the part of acids with powerful bases, and that of bases with energetic acids. Alumina A1303 is an oxide of this kind. 4thly. Simple oxides. These oxides combine neither with the acids nor with the bases. Under the influence of acids, they part with a portion of their oxygen, or of their metal, and are converted into protoxides, which combine with the acid. The peroxide of manganese Mn03 is an oxide of this class. When heated with sulphuric acid, it parts with one-half of its oxygen and forms the sulphate of the protoxide of manganese. The suboxide of lead Pb30 is changed, by contact with acids, into metallic lead Pb, and protoxide of lead PbO, which combines with the acid. These oxides frequently undergo decompositions analogous to the bases. Thus, the binoxide of manganese Mn03, melted with caustic potassa, is changed into the sesquioxide of manganese»Mn303, and into manganic acid Mn03, which combines with the potassa: 3Mn03+K0=Mna0a+K0,Mn03. 5thly. Saline oxides. These oxides result from the combina- tion of a basic metallic oxide with a higher oxide of the same metal. They are true salts, in which the electropositive elements of the acid and the base are formed by the same metal. The oxides of iron Fe304, of manganese Mn304, of chromium Cr304, belong to this class, and their formula; should be written FeO, Fe303; MnO,MnaOa; Cr0,Cr303. The brown oxide of chromium METALLIC OXIDES. 381 Cr03, belongs to the same class: it should he written Cr303,Cr03= 3CrOs. The same is true as regards antimonious acid Sb04, the formula of which should be Sb03,Sb05=2Sb04. § 318. Certain metals form a great number of compounds with oxygen, which are included in the five classes just defined, and of which manganese furnishes a remarkable example. The protoxide of manganese MnO is a powerful base. The sesquioxide Mn303 is a very feeble base, but we know as yet of no compounds in which it plays the part of an acid: it forms the limit of the neutral oxides. The binoxide Mn03 is a simple oxide. The oxide Mn304 is a saline oxide, the true formula of which is Mn0,Mn303. Manganic acid, Mn03 and hypermanganic Mn307 are powerful metallic acids. § 319. In general, the oxide of the formula RO is the most powerful base among those which can be formed of the same metal. The oxides R303 are very feeble bases, and frequently play the part of acids with powerful bases; in which latter case, they are ranked with the neutral oxides. The oxides R03 are often me- tallic acids, such as the peroxides of lead Pb03, and of tin Sn03: they are sometimes simple oxides, as the binoxide of manganese Mn03; and sometimes they should be regarded as saline oxides, such as the brown oxide of chromium, Cr03=J(Cr303,Cr03). Lastly, the oxides which have more complex formulae, such as Fe304, Mn304, are saline oxides, which should be written FeO, Fe303 and Mn0,Mn303. § 320. The metallic oxides are obtained by the following pro- cesses:—1. By heating the metal in the air or in oxygen, the various steps of which oxidation have been explained (§ 303). The lower are also changed into higher oxides by calcination. Pro- toxide of manganese MnO heated in the air is converted into the sesquioxide. When protoxide of barium or baryta BaO is heated to 750° in a current of oxygen, it absorbs the gas, and is changed into the binoxide of barium BaOa. A red-heat, on the contrary, decomposes the binoxide of barium, and restores it to the state of a protoxide. 2. The metal may be oxidized by calcining it with a substance which parts readily with its oxygen. By heating antimony with nitrate of potassa, antimoniate of potassa is obtained; and by treating this salt with an acid, antimonic acid is isolated. This process is often adopted to change lower into higher oxides, es- pecially into metallic acids. Thus, by fusing oxide of chrome Cr303 with the nitrate of potassa, this oxide is changed into chromic acid Cr03, and chromate of potassa obtained. 3. The metal, or one of its inferior oxides, may be oxidized by acting on it with nitric acid, and evaporating off the excess of acid. Some metals are thus changed into higher oxides, which remain uncombined; as tin and antimony, which become stannic 382 THE METALS, acid, Sn03, and antimonic acid SbOs. Most frequently, nitrates are formed, which are decomposed by calcination, and that oxide remains which is formed when the metal is heated in oxygen at the temperature at which the decomposition of the nitrate takes place. 4. Some peroxides are obtained by acting on the lower oxides by the binoxide of hydrogen (§ 90). In this way, binoxide of cal- cium and several metallic peroxides are prepared, which cannot be obtained in any other mode. 5. Heat alone converts some higher into lower oxides. The sesquioxides of cobalt and nickel Co303 and Ni203 change at a red-heat into protoxides CoO and NiO. Binoxide of lead PbOa is changed into the protoxide PbO. The binoxide of manganese Mn03 passes into the state of a saline oxide Mn0,Mn303: 3Mn0a=Mn0,Mn303+20. 6. Hydrogen, at a red-heat, reduces a great number of oxides to the metallic state, but reduces some higher oxides only to the state of protoxides. The sesquioxide of manganese Mn303, heated in a current of hydrogen, is converted into the protoxide MnO. 7. The solution of a metallic salt, or of the corresponding chlo- ride, is precipitated by an alkaline base or by ammonia. By pouring a solution of potassa or ammonia into a solution of the sulphate of the protoxide of iron, or of the protochloride of iron, a precipitate is obtained of the hydrated protoxide of iron : Fe0,S03+K0+H0=K0,S08+Fe0,H0; FeCl+KO+HO=KCl+FeO,HO. The same alkaline liquid gives, with solutions of the sulphate of the sesquioxide or sesquichloride of iron, a precipitate of hydrated sesquioxide of iron: Fea03,3S03-f 3K0-f-II0=3(K0,S03)-f Fe303,II0; Fe3Cl3+3KO-f II0=3KCl+Fe303,H0. Potassa is often replaced by ammonia, and sometimes even by carbonate of potassa, when the oxide of the salt does not combine with carbonic acid. The oxides thus prepared in the humid way, generally precipi- tate in the state of hydrates, but heat suffices to transform the majority of them into anhydrous oxides. The hydrates of the oxides formed by the metals of the first section are alone unde* composable by heat. 8. All the carbonates, except those of the metals of the first section, are decomposed by heat, setting their oxides at liberty. Thus, by calcining the carbonates of baryta, strontia, and lime, at ACTION OF METALLOIDS ON THE OXIDES. 383 a high temperature, baryta, strontia, and lime are obtained. The carbonate of lead PbO,COa gives off its carbonic acid at a lower temperature than the preceding carbonates, and the protoxide of lead PbO, remains. When the protoxide which forms the base of the salt has a great affinity for oxygen, it frequently happens that it decomposes the carbonic acid and seizes upon part of its oxygen. Thus, when the native carbonate of the protoxide of iron Fe0,C03, which mine- ralogists call sparry iron, is heated, it gives the saline oxide FeO, Fe303, and a mixture of carbonic acid and carbonic oxide is dis- engaged : 3(Fe0,C02)=Fe0,Fe303+2C03+C0. Action of the Metalloids on the Oxides. § 321. Action of Oxygen.—Oxides which have not reached their maximum of oxidation can sometimes combine directly with a new proportion of oxygen. The combination is sometimes effected in the cold by exposure to the air; but it takes place more readily if water be present, as when the oxide is combined or only moist- ened with water. The hydrates of the protoxide of iron and manganese promptly absorb oxygen from the air, and are con- verted into hydrated sesquioxides. Other oxides combine with oxygen only when moderately heated in the air: thus, protoxide of lead, heated to a temperature of about 750°, absorbs oxygen from the air, and is converted into a new oxide, minium. A higher temperature, on the contrary, decomposes the minium and re- stores it to the state of protoxide. § 322. Action of Hydrogen.—Many oxides are decomposed by hydrogen, which seizes upon their oxygen to form water, but the reaction generally requires a certain elevation of temperature. The oxides of the metals in the first two sections are not decom- posed by hydrogen at any temperature. Those of the metals in the other sections are all reduced to the metallic state by hydrogen, at higher or lower temperatures. Those of the sixth section are all decomposed by hydrogen, at a temperature slightly above that of boiling water ; the others require a red-heat. Hydrogen reduces the oxides of iron at a red-heat, and vapour of water is formed. On the other hand, it was shown (§ 68) that iron, when heated to redness in a current of steam, is oxidized by decomposing the water, and disengaging hydrogen gas. Two en- tirely opposite effects are here produced under circumstances apparently identical. We might infer, from the decomposition of the oxides of iron by hydrogen, that, at a red-heat, the hydrogen has more affinity for oxygen than the iron, while, from the decom- position of steam effected by iron at a red-heat, we would con- clude, on the contrary, that the iron had more affinity for oxygen 384 THE METALS. than the hydrogen. We shall subsequently meet with several analogous phenomena. Chemists explain these apparent contra- dictions, by saying that substances act not only by virtue of the electric affinities, but also according to the respective quantities which are present. So that if two substances are in contact with a third, for which they have slightly different affinities, that which predominates in the sphere of action expels the other. In the two experiments just described, we have had in contact at a red- heat, iron, oxide of iron, vapour of water, and hydrogen. In that in -which the vapour of water is passed over heated iron, the iron may be considered as predominating with reference to the hydro- gen, because this gas, as fast as it is produced, is carried off by the current of steam, of which it constitutes only a small propor- tion, so that the iron will consequently oxidize. In the experi- ment in which oxide of iron is heated in a current of hydrogen, each molecule of the oxide is in the sphere of action of a great number of molecules of hydrogen, and the latter, consequently, seizes on the oxygen. From this it is evident that, for a given temperature, there exists a certain proportion of hydrogen and vapour of water, which exerts no reducing action on oxide of iron, nor oxidizing action on metallic iron. If the proportion of vapour of water is greater, the metal will oxidize; if less, the oxide will be reduced. These proportions in which hydrogen and oxygen should exist, so as to exert no action either on metallic iron or on the oxide of iron, probably vary with the temperature. § 323. Action of Carbon.—Carbon reduces all the metallic ox- ides which are decomposed by hydrogen; and at a very high tem- perature, some oxides are reduced which resist the action of hydrogen. Thus, the oxides of potassium and sodium are entirely decomposed by carbon at a white-heat, and their metals set free. When the reduction of the oxide takes place at a low tempera- ture, carbonic acid is disengaged: if it occur only at a high tem- perature, carbonic oxide is given off. Many metals, in fact, decompose carbonic acid at a red-heat, and convert it into car- bonic oxide. Charcoal produces a similar decomposition. § 324. Action of Sulphur.—At a high temperature, sulphur acts on the majority of metallic oxides. When heated with metals of the first section, a mixture of sulphate and sulphide is formed; but if charcoal be added, a sulphide alone is produced. The oxides of the metals of the second section are not changed when heated with sulphur; but several of them produce sulphides when their oxides mixed with charcoal are heated intensely in a current of vapour of sulphur. The oxides of the metals of the last four sections are changed into sulphides by sulphur, with the disengagement of sulphurous acid; but it is often necessary, for this purpose, to pass the vapour ACTION OF METALLOIDS ON THE OXIDES. 385 of sulphur over the highly heated oxide, and sometimes even to mix the latter previously with charcoal. § 325. Action of Chlorine.—Chlorine acts variously on the oxides, according to its dryness, moisture, and temperature. When cold, or influenced by heat, dry chlorine changes nearly all the oxides into chlorides. We must except, however, the oxides of some metals of the second section, which resist the action of chlorine, even at the highest temperatures. But by taking care to mix the oxide previously with charcoal, and to heat the mixture in a current of dry chlorine, the affinity of carbon for oxygen, combined with that of chlorine for the metal, always effects the decomposition of the oxide; carbonic oxide being given off, and a metallic chloride formed. When the oxides are dissolved or suspended in water, the action of chlorine is, generally, very different from that first mentioned. If a current of chlorine be passed through a solution of potassa, the reaction varies with the state of dilution or concentration of the liquid, and the temperature. If the solution be dilute, and the temperature not allowed to rise, a reaction takes place between 2 equivalents of potassa and 2 of chlorine, forming hypochlorite of potassa and chloride of potassium, as expressed by the follow- ing equation: 2K0+2C1=KC1+K0,C10. If the solution be concentrated, and the temperature allowed to rise, reaction takes place between 6 equivalents of potassa and 6 of chlorine, and a mixture of chlorate of potassa and chloride of potassium is obtained; thus, 6K0+6C1=K0,C10S+5KC1. If the concentrated alkaline solution be kept constantly boiling, chloride of potassium and chlorate of potassa are again formed; hut the proportion of chlorate is less, and oxygen gas is given off. The oxides of all the metals of the first section behave in the same manner. The oxides of a majority of the metals of the second section are not changed by chlorine under the influence of water, even at the temperature of 212°, except magnesia and protoxide of man- ganese. The oxide of magnesium is changed, in this case, into chloride of magnesium and hypochlorite of magnesia: the pro- toxide of manganese behaves, under the influence of moist chlorine, like the protoxides of the metals of the third section. The protoxides of the metals of the third section, suspended in water, are changed by chlorine into chlorides and sesquioxides. With the protoxide of iron, we have, 3Fe0 + Cl=FeCl+Fea03. 386 THE METALS. If the oxide be suspended in an alkaline liquid, the protoxide is completely transformed into a sesquioxide, and the chloride of potassium is produced: 2FeO+KO+Cl=FeaO,+KCl. Chlorine does not act on the sesquioxides of the metals of the third section suspended in water, unless the liquid contains a large quantity of potassa. In this case, the oxide of iron may pass into the state of a compound containing more oxygen than the sesqui- oxide, ferric acid, which forms, with potassa in excess, the ferrate of potassa. Thus, FeaO,+5KO+3Cl=2(KQ,FeO,)+3KCl. The oxides of the metals of the last three sections are changed into chlorides by the action of chlorine in presence of water. The action of bromine and iodine on the metallic oxides is in general analogous to that of chlorine. § 326. Action of the Metals on the Metallic Oxides.—The action of the metals on the metallic oxides may often be foreseen, when we have a very clear idea of the affinity of the metals for oxygen. But it is difficult to generalize upon this action, for the relative affinity of the metals for oxygen varies greatly with temperature. Thus potassium decomposes oxide of iron at a red-heat, whilst at a higher temperature, as a strong white-heat, iron, on the con- trary, decomposes oxide of potassium. METALLIC CHLORIDES. § 327. Many of the metals combine directly with chlorine, es- pecially if heated in a current of the gas, when they are rapidly and entirely transformed into chlorides. This property must he attributed, partly to their great affinity for chlorine, partly to the physical properties of the chlorides. The chlorides are, in fact, all very fusible, and many of them volatile; so that when a metal is heated in a current of chlorine, its surface is always exposed freely to the action of the gas, either because the melted chloride runs off or volatilizes as it is formed. The metallic chlorides are, in general, undecomposable by heat alone, excepting the chlorides of gold, platinum, and, probably, those of several other metals of the sixth section, which are re- duced to the metallic state by an elevated temperature. Many metallic chlorides are obtained by dissolving their metals in chlorohydric acid; the protochlorides of metals of the third section being thus readily obtained. The chlorohydric acid is decomposed, chloride formed, and hydrogen disengaged. Fe+HCl=FeCl+H. METALLIC SULPHIDES. 387 The metals of the fifth section do not decompose chlorohydric acid, even at the boiling point; but a chloride is formed when nitric acid is added, that is, when the metal is treated with aqua regia. The metals of the third section are changed, in this case, into perchlorides. Action of the Metalloides on the Metallic Chlorides. § 328. Action of Oxygen.—Oxygen does not act on the chlorides of metals of the first section ; but readily oxidizes those metals in the second, third, fourth, and fifth sections, when their chlorides are heated in a current of oxygen. The chlorides of the sixth section, which are not decomposed by heat, are not changed when heated in oxygen; while those, on the contrary, which are decom- posed by heat alone, give off their chlorine without combining with the oxygen. § 329. Action of Hydrogen.—The chlorides from metals of the first two sections are not decomposed at any temperature by hy- drogen ; but those from the last four sections are decomposed by it at various temperatures. This behaviour offers a convenient method for obtaining several metals in a state of purity, but is dif- ficult of application to others, because decomposition takes place only at the highest temperature. An inverted action is, moreover, observed here, exactly resembling that pointed out (§ 322) when speaking of the action of hydrogen on the oxides. Thus, chloride of iron is decomposed by hydrogen at a red-heat, chlorohydric acid being disengaged and metallic iron remaining. On the other hand, metallic iron decomposes chlorohydric gas at the same tem- perature, forming chloride of iron and setting hydrogen free. The anomaly was explained (§ 322). § 330. Action of Carbon.—Carbon exerts no sensible action on the metallic chlorides. METALLIC BROMIDES AND IODIDES. § 331. Metallic bromides and iodides are prepared like the cor- responding chlorides, and their reaction with the metalloids is analogous to that of the chlorides. METALLIC SULPHIDES OR SULPHURETS. § 332. It was stated (§ 306) that all metals can combine with sulphur when heated with it, or still better, when heated to a high temperature in the vapour of sulphur. A great number of metal- lic sulphides can also be obtained by heating the oxides with sul- phur, or by calcining in a crucible covered with damp charcoal, a mixture of metallic oxide, carbonate of potassa or soda, and sul- phur. The alkaline carbonate is then changed into a polysulphide, 388 THE METALS. which itself converts the metallic oxide into a sulphide, while oxygen is disengaged in the state of carbonic oxide. If the metal can form an electronegative sulphide, as happens with the me- tals of the fourth section, this sulphide combines with a portion of the alkaline sulphide which has become a monosulphide, and a sulphosalt is formed, in which the alkaline monosulphide plays the part of a base. > A great number of metallic sulphides can also be prepared by passing a current of sulphuretted hydrogen through a solution of the metallic salts, especially insoluble sulphides from metals of the fifth and sixth sections. Sulphides from metals of the third section may also be pre- pared in the humid way, by pouring a solution of alkaline sul- phide into a saline solution of the metal. Thus, with sulphate of the protoxide of iron and monosulphide of potassium, the reac- tion is Fe0,S03+KS=K0,S03+FeS. If an excess of alkaline sulphide be poured into the solution of a salt formed by a metal of the fourth section, there is formed, at first, a precipitate of the metallic sulphide, but it is subse- quently dissolved in the excess of alkaline sulphide, by producing a sulphosalt, in which it plays the part of an acid. The sulphides of the third and fifth sections have a well-marked metallic lustre. Metallic sulphides resist powerfully the action of heat, there being only a few sulphides of the sixth section which are decom- posed at a very elevated temperature. § 333. Action of Oxygen.—Oxygen acts energetically on all metallic sulphides, at a higher or lower temperature. The sulphides of the metals of the first section, heated in con- tact with oxygen, are changed into sulphates ; the metal and sul- phur both combine with oxygen, and the products of combustion remain combined. The sulphide of magnesium, which belongs to the second section, presents a similar reaction. Those of the third and fifth sections, and the sulphides of manganese belonging to the second, are decomposed by oxygen; but the products of de- composition vary according to the temperature. When the latter is very elevated, sulphurous acid is disengaged, and the metal remains in the state of an oxide. At a lower temperature, at a dull red-heat, for instance, a certain quantity of sulphate is always formed, so that we obtain a mixture of oxide and sulphate. Metallic sulphides of the fourth section are changed into oxides, and the sulphur disengaged in the state of sulphurous acid. Lastly, sulphides from the sixth section, heated in a current of oxygen, Action of the Metalloids on Metallic Sulphides. SALTS 389 are reduced to the metallic state, and the sulphur disengaged in the state of sulphurous acid. Oxygen can also act when cold on the majority of sulphides, principally under the influence of water, whereby many of them are finally changed into sulphates. METALLIC PHOSPHURETS. § 334. The metals of the first section are the only ones which combine easily with phosphorus. Phosphurets of several other metals are obtained by passing a current of phosphuretted hydro- gen through saline solutions, by which an insoluble phosphuret is precipitated. In this way, the phosphurets of copper, lead, and tin can be prepared. But the best mode of obtaining the phos- phurets consists in heating the phosphates mixed with charcoal. The phosphurets of metals of the first section are decomposed by water, and disengage phosphuretted hydrogen, thus readily evincing their nature. METALLIC ARSENIURETS. § 335. A great number of metallic arseniurets may be prepared by heating together the metal and arsenic, both finely powdered. They are also obtained by decomposing the arseniates by charcoal, at a high temperature. The arseniurets, in general, possess a metallic lustre; and when heated with chlorohydric acid, they dis- engage arseniuretted hydrogen. GENERAL CONSIDERATIONS ON THE SALTS. § 336. I give the name of salt to every combination of two binary compounds, one of which acts the part of an electropositive element, or base, and the otjier that of an electronegative element, or acid. The bases, or electropositive binary compounds, always result from the combination of a metal with a metalloid, such as the pro- toxide and protosulphide of potassium. The acids, or electro- negative binary compounds, are most frequently combinations of two metalloids, as sulphuric, nitric, phosphoric, etc. acids; sul- phocarbonic and sulpharsenious acids. But they sometimes result from the combination of a metal with a metalloid, as chromic, manganic, stannic, etc. acids, the sulphides of antimony and tin. The majority of known bases are compounds of a metal with oxygen; the majority of the known acids, compounds of oxygen with a metalloid, or metal. Thus, the most numerous, and by far the most important salts, are the oxysalts. We are now, however, acquainted with a considerable number 390 THE METALS. of sulphosalts, formed by the combination of an electropositive metallic sulphide or sulphobase, with an electronegative metalloidal or metallic sulphide, or sulphacid. We also know some double chlorides, which may be considered as resulting from the combination of an electropositive metallic chloride, or chlorobase, with an electronegative metalloidal or me- tallic chloride, or chloracid. These compounds, called chlorosalts, are not yet very numerous; but others will undoubtedly be found, when the attention of chemists is directed to this point. An oxyacid may possibly combine with a sulphobase or with a chlorobase, and a sulphacid or a chloracid with an oxybase, so as to form a salt; but hitherto, no compound of this kind is certainly known. 337. The oxysalts are, therefore, by far the most important, and the only ones which have hitherto been carefully studied. All our general remarks on the salts, in this chapter, will relate princi- pally to the oxysalts. We should probably make similar remarks on the other classes of salts, were they as well known to us. The oxysalts are divided into neutral, acid, and basic salts. The characters on which this distinction is founded are easily de- fined as regards salts formed by the combination of powerful bases with energetic acids; but they become less clear in salts formed by powerful bases with feeble acids, or by feeble bases with power- ful acids, or, lastly, by feeble bases and acids. The difficulty is still greater when the acid, or base, or resulting salt are insoluble in water. The nature of neutral, acid, or basic salts is generally recog- nised by the changes of colour they produce on certain vegetable colouring matters, called coloured reagents or tests, the most im- portant of which is the tincture of litmus. 338. The blue tincture of litmus is a true salt, resulting from the combination of a mineral base with a vegetable acid, which is red. When the tincture is treated with a string acid, its base is removed and the vegetable acid set free, which then shows its true colour, a bright red. But, if it be treated with a feeble acid, only a portion of its base is removed, and there remains a salt with an excess of vegetable matter, having a purplish tint. A soluble base, on the contrary, changes the reddened tincture of litmus to blue, because it combines with the acid and forms a blue salt. In order that the blue tincture may be as sensitive as possible to the action of acids, it necessarily should not be mixed with an excess of free base ; for, in this case, the first portions of acid added would sim- ply combine with the free base, and there would be no reaction on the tincture after the complete saturation of the free base. Again, in order that the red tincture of litmus may possess its maximum of sensitiveness to the action of bases, the blue tincture should have been decomposed by a quantity of acid exactly sufficient to SALTS 391 isolate the red vegetable acid, and no other free acid should be formed in the liquid. Sulphate of potassa does not react with tincture of litmus, be- cause the sulphuric acid and potassa are combined with so great an affinity that they cannot combine separately, either with the acid or with the base of the coloured tincture, so that the latter remains intact and preserves its colour. But, if a colouring matter existed, the acid of which was powerful enough to remove the potassa from the sulphate of potassa, it is clear that this matter would manifest an alkaline reaction in the presence of the sulphate of potassa. The indications of the coloured reagents are, therefore, not ab- solute, but merely relative. It might even happen that the same substance would evince an acid reaction with one colouring matter and alkaline reaction with another. In this way, boracic acid pro- duces a purplish colour with the blue tincture of litmus, thus mani- festing the reaction of a feeble acid, while it turns hematin blue, and presents, as regards this colouring matter, a basic reaction. In the same manner, the nitrate and acetate of lead redden the tincture of litmus and turn hematin blue. The base of the tinc- ture of litmus removes the acids from the two salts of lead, and the coloured acid being set free, the blue tincture is reddened. The red acid of the hematin, on the contrary, abstracts the oxide of lead from the nitrate and acetate, and a blue salt is produced. 339. Let us now examine the salts which sulphuric acid forms with various bases. Sulphuric acid reddens strongly the blue tincture of litmus, and the reaction is so delicate that part of sulphuric acid thrown into water evinces it in a very marked manner. Potassa, on the contrary, blues the tincture of litmus previously reddened by an acid, and the reaction is as evident as that exerted by the acid on the blue tincture, provided the litmus has been reddened only by the smallest possible quantity of acid. If a dilute solution of sulphuric acid be carefully poured into a solution of potassa, testing with the greatest accuracy the reac- tion of the liquid with the tincture of litmus, a liquid is obtained which no longer manifests an alkaline reaction on the tincture, without presenting, however, the acid reaction; and yet the liquid is such that the addition of a single drop of the acid would imme- diately show an acid reaction. We then say that the alkaline properties of the potassa have been exactly neutralized by the acid properties of the sulphuric acid, that there has been a saturation or neutralization of the acid by the base in their action on the tincture of litmus. If the liquid be evaporated to dryness, a crys- talline salt, the sulphate of potassa, remains. The analysis of this salt shows that it contains quantities of potassa and sulphuric acid, such that the acid contains three times as much oxygen as the base; and, as we have agreed to call the 392 THE METALS. equivalent of potassium the quantity of this metal which combines with 8 of oxygen, the formula of the sulphate of potassa should evidently he written K0,S03. If soda or lithia be saturated in the same manner with sulphuric acid, and the liquid neutral to tincture of litmus be evaporated, a salt is obtained, the sulphate of soda or lithia. In these two salts, the quantity of oxygen contained in the sulphuric acid is again ex- actly triple of that contained in the base. If the same experiment he made in solutions of baryta and strontia, which powerfully blue the reddened tincture of litmus, it will be observed that the first drops of acid added cloud the liquid, and a white precipitate is formed. This insoluble compound will continue to be deposited until the liquid begins to exert a slight acid reaction, when the filtered solution will leave no residue upon evaporation. The insoluble sulphate thus formed does not re- act on the tincture of litmus; but it cannot hence be concluded that the product is really neutral. For, in order that a substance may act on the tincture of litmus, it must be soluble in water, so that the molecules of the salt may come into contact with those of the tincture. The analysis, however, of the sulphates of baryta and strontia thus produced, again shows that the oxygen in the acid is equal to three times that in the base. Chemists have agreed to consider these sulphates as neutral salts, although their neutrality with coloured reagents cannot be directly verified. All the basic oxides of the metals of the other sections being in- soluble in water, it is impossible to ascertain their peculiar action on coloured reagents. By combining them with sulphuric acid, sulphates are still obtained, and, when soluble, they generally red- den the tincture of litmus. Nevertheless, in all these sulphates, the oxygen in the sulphuric acid is treble of that in the base, as in the neutral sulphates of potassa, soda, and lithia. Chemists have agreed to consider as neutral sulphates all the sulphates in which the quantity of oxygen in the acid is treble of that in the base, whatever may be, otherwise, their reaction on vege- table colours. Potassa, soda, and lithia may form salts with sulphuric acid, which contain more acid than the neutral sulphates. If the bases be dis- solved in an excess of sulphuric acid, and the solution be evapo- rated, crystallized sulphates are obtained, in which the oxygen in the acid is six times that in the base. These salts will therefore be acid sulphates, or bisulphates. 340. A solution of potassa, exactly saturated with nitric acid, affords when evaporated a crystallized salt, in which the oxygen of the acid is quintuple that of the base. In the same way, if so- lutions of the metallic oxides of the first section be saturated with nitric acid, soluble salts are obtained perfectly neutral to coloured SALTS, 393 tinctures, and which crystallize after the evaporation of the liquid. In all these nitrates, the oxygen of the acid is quintuple that of the base. But, if the metallic oxides of the other sections be dissolved in nitric acid, nitrates are obtained which crystallize after the evaporation of the liquid, and which present the same proportion of 5 : 1, between the quantity of oxygen in the acid and the base, but their solutions exhibit a strongly acid reaction. We regard as a neutral nitrate every nitrate in which the oxygen in the acid is quintuple of that in the base, whatever may be its reaction on tincture of litmus. 341. Water plays the part of a base with reference to powerful acids. Monohydrated sulphuric acid may therefore be regarded as a true salt, and even as a neutral sulphate, for the proportion between the oxygen of the acid and that of water is as 3 : 1. For the same reason, monohydrated nitric acid will be a neutral ni- trate of water. It may therefore be said that when nitric or sul- phuric acid is combined with bases, these bases are made to react on salts already formed, on nitrate or sulphate of water, and that the base is only substituted in place of the basic water, by virtue of its greater affinity. 342. In the two examples first selected, the composition of the neutral salts was determined by finding the quantities of potassa, soda, and lithia which exactly saturate, with regard to coloured reagents, the same weight of sulphuric or nitric acid. Now, these quantities are found to be such that they contain precisely the same weight of oxygen. The same relation is observed in the crystallized salts which the same acids form with other metallic oxides. This very remarkable law may therefore be advanced: The ponderable quantities of the various bases which form neutral salts with the same weight of nitric or sulphuric acid, contain ex- actly the same quantity of oxygen. If these quantities of the various bases be referred to the weight of sulphuric and nitric acid chosen to represent their equivalents, and be designated by a, b, c, d..., it may be said: If the equivalent A of sulphuric acid form neutral salts with the weights a, b, c, d..., of potassa, soda, baryta, lime, etc., the equivalent B of nitric acid will also form neutral salts with the same weights a, b, c, d..., of these bases; so that these weights a, b, c, d..., which are equivalent to each other as regards the weight A of sulphuric acid, are also equivalent to each other as regards the weight B of nitric acid. 343. Let us now examine the compounds which weak acids form with these bases, and ascertain how chemists proceed in determin- ing the composition of their neutral salts. With weak acids, such as sulphurous, carbonic, boracic, etc., the saturation of the alkaline properties of potassa, as regards coloured reagents, is never completely effected, whatever may be the quan- 394 THE METALS. tity of acid added. The liquid always retains an alkaline reaction, and the character of saturation evinced by coloured reagents can- not be invoked to define the neutral salts. 344. If a current of sulphurous acid gas be passed through a concentrated solution of potassa, until the latter can no longer dissolve it, a crystallized salt is deposited after some time, in which the oxygen of the acid is quadruple that of the base. If this salt be redissolved in water, and a quantity of potassa added equal to that it already contains, a new erystallizable salt is obtained by evaporating the liquid, in which the oxygen of the acid is double that of the base. Which of these salts shall be assumed as the neutral salt? Chemists are governed in their choice by the following considera- tions. By endeavouring to form sulphites with the various metallic ox- ides, two series of salts are obtained with the metals of the first section, which correspond to the two sulphites formed by potassa; but, with the metals of the other sections, only a single series of salts is obtained, viz. that in which the oxygen of the acid is double of that of the base. Chemists have agreed to regard those as neutral sjilphites which exist in the greater part of the metallic oxides. Consequently, the neutral sulphite of potassa takes the formula KOjSOg, and the sulphite containing a double quantity of sulphurous acid is considered as an acid sulphite, or a bisulphite, and its formula becomes § 345. A precisely similar circumstance occurs in the carbonates. If a concentrated solution of potassa be saturated with carbonic acid, a crystallized salt is deposited, after some time, the acid of which contains four times more oxygen than the base. If this salt be redissolved in water, and a quantity of potassa added equal to that it already contains, a new crystallized carbonate can be ob- tained by evaporating the liquid, in which the acid only contains a quantity of oxygen double that of the base. Moreover, both salts exhibit an alkaline reaction to coloured tinctures. Soda and lithia afford two similar carbonates. Baryta, strontia, lime, and magnesia form carbonates frequently found in beautiful crystals in a native state. In all these carbonates, the relation between the oxygen of the acid and that of the base is as 2 : 1. They are in- soluble in water, but dissolve slightly in water charged with car- bonic acid. The latter solution may be regarded as containing carbonates in which the oxygen of the carbonic acid is equal to four times that of the’ base; but these have not yet been obtained in a crystalline form. The liquid, when evaporated, always depo- KO,2SOa. SALTS, 395 sits carbonates, in which the oxygen of the acid is double that of the base. The metals of the other sections also afford only the first series of carbonates. This consideration has induced the majority of chemists to re- gard those as neutral carbonates in which the oxygen of the acid is double that of the base. The formula of the neutral carbonate of potassa is therefore K0,C03, and the second salt becomes a bicarbonate, the formula of which is written KO,2COa. Some chemists, however, even now regard this last salt as a neutral carbonate, because it approaches the neutrality shown by coloured reagents more than the first. They write its formula K0,C304; and the first salt becomes a subcarbonate, or a bibasic carbonate, of which the formula is written 2 K0,Ca04. In this point of view, the formula of carbonic acid is C304, and the wreight of its equivalent is twice as great as we have admitted it (§ 262). § 346. Boracic acid forms two salts with alkalies, both of which have an alkaline reaction. If boracic acid be dissolved in a solu- tion of soda and the liquid be evaporated, a salt is obtained in which the boracic acid contains six times more oxygen than the base. If this salt be melted in a platinum crucible with as much more soda as it already contains, a new salt is obtained which dis- solves* in water and crystallizes upon evaporating the liquid. In this salt, the boracic acid contains only three times more oxygen than the soda. Which of them shall we choose as the neutral salt ? The difficulty is here greater than with the sulphites and carbon- ates, which have been more minutely studied than the borates, so that chemists are not agreed upon this point. Some regard the first borate above mentioned as the neutral salt, and give it the formula NaO,BOe; in this case, the second borate becomes a bi- basic salt, and its formula is written 2Na0,B08. Others, on the contrary, consider the second borate as the neutral salt, and write its formula NaO,BOs: the first salt then becomes a biborate, the formula of which is Na0,2B03. § 347. The definition of a neutral salt presents peculiar diffi- culties with some acids, even very powerful ones, which chemists regard as polybasic, that is, as possessing the property of forming neutral salts, not with one, but with several basic equivalents. We shall give an idea of these difficulties by taking phosphoric acid for an example. It was stated (§ 211) that phosphoric acid can be obtained in three states. That which is obtained by dis- solving phosphorus in nitric acid, differs remarkably in its proper- ties from the acid obtained by the combustion of phosphorus in oxygen ; for the two modifications produce perfectly distinct classes of salts. We are acquainted with even a third modification of the acid, which afford a third series of phosphates, differing from the first two. These facts will be developed more in detail when treating of the phosphates of soda, and it will now be sufficient to 396 THE METALS. consider the salts formed by phosphoric acid, obtained by the solu- tion of phosphorus in nitric acid. If a great excess of a solution of phosphoric acid be poured into a solution of soda, and the liquid be evaporated, a crystallized salt is obtained in which the phosphoric acid contains five times more oxygen than the soda. If this salt be redissolved in water, and as much more soda added as it already contains, a new crystallized salt is obtained by evaporating the liquid, in which the oxygen of thewcid is to that of the base as 5 is to 2. Lastly, if the last salt be dissolved in water, and an excess of soda be added, a third crystallized phosphate of soda is obtained by evaporation, in which the oxygen of the acid is to that of the base as 5 to 3. The first of these three phosphates has an acid reaction on litmus; the other two, on the contrary, have an alkaline reaction. The same modification of phosphoric acid gives therefore three very different phosphates. How shall we decide which of them shall be considered as the neutral phosphate ? Chemists have been induced, by a mass of facts which will be developed when treating of the phosphates of soda, to admit that the three phosphates have the same mode of constitution ; as all those formed of 1 equivalent of the acid combine with 3 equivalents of base. In thef third phosphate, the 3 equivalents of base are 3 equivalents of soda; in the second, there are 2 equivalents of soda and 1 equivalent of basic water; and, lastly, in the first phosphate, the 3 equivalents of base are formed of 1 equivalent of soda and 2 equivalents of basic water. Thus, the three phosphates, although one has an acid, and the other two an alkaline reaction, are all considered as having the same composition; and, if one of them be regarded as a neutral salt, the others are equally so. § 348. The consideration of the water which may act the part of a base in salts has greatly modified the views of chemists on the classification of these salts. The majority of acid salts may be regarded as neutral, the excess of acid being considered as combined with the basic water. Thus, the crystallized bisulphate of potassa contains 1 equivalent of water, which it does not aban- don, by the sole action of heat, without decomposition. We have, therefore, some reason for regarding this salt as resulting from the combination of the neutral sulphates, the sulphate of potassa and sulphate of water, and writing its formula K0,S03 + H0,S03. This reasoning is applicable to the majority of other acid salts. By generalizing it, we are led to regard the same acid as forming only a single series of salts, all presenting the same mode of con- stitution, and differing only in the nature of the bases combined with the acid. § 349. We thought it proper to insist on the definition of the neutrality of salts and their division into neutral, acid, and basic SALTS, 397 salts, because this division is now generally adopted. The some- what prolix discussion in which we have indulged shows these definitions to be vague and full of contradictions; and it would be desirable for chemists to abandon them entirely. § 350. If an oxybase and a hydracid be brought together, there is not a simple combination of the two bodies, but a reciprocal de- composition, the hydrogen of the hydracid combining with the oxygen of the base to form water, and the electropositive element of the base, the metal, combining with the electronegative element of the hydracid, to form another binary compound which corre- sponds in its composition to the oxybase used. Thus, potassa and chlorohydric acid produce water and the chloride of potassium : K0+HC1=H0+KC1. With sesquioxide of iron and chlorohydric acid, water and ses- quichloride of iron are formed: Fea03+BHCl=BH0+Fe3Cl3. The saturation of the hydracid by the base, ascertained by means of coloured reagents, is often as complete as those of a powerful oxacid by the same base. Thus, the solution of chlorohydric acid, which strongly reddens the tincture of litmus, may be rendered perfectly neutral to the tincture by adding the proper quantity of potassa; and if the liquid be then evaporated, only water and chloride of potassium are obtained. § 851. Several chemists assume that, in solution, the hydracid and oxybase are simply combined, and that the reciprocal decom- position takes place only at the moment of crystallization. Many reasons are advanced in favour of and in opposition to this view, which we shall not stop to consider, but admit, with the majority of chemists, that the reciprocal decomposition of the hydracid and oxybase takes place at the very moment when the two bodies are brought into contact. The binary compounds of the metals with those metalloids capable of forming hydracids with hydrogen, present physical pro- perties analogous to those of the salts; and, in a great number of chemical reactions effected in water, they behave like simple compounds of the oxybase with the hydracid. Thus, when chlo- ride of potassium is heated with hydrated sulphuric acid, sulphate of potassa is formed and chlorohydric acid disengaged. The re- action is therefore precisely similar to that which would take place if the sulphuric acid decomposed a salt formed by the direct com- bination of the hydracid with the oxybase, and simply expelled the latter in order to combine with the base. But, in reality, the reaction is more complex ; for the water combined with the sul- phuric acid is decomposed, its oxygen uniting with the metal of the binary compound, its hydrogen with the electronegative ele- 398 THE METALS. ment, and, lastly, the newly-formed oxybase forming a salt with the oxacids: KCl+SOs,HO=KO,SO,+HCl. On account of the great resemblance between this class of binary- compounds and the salts properly so called, in their physical pro- perties, and even in a great number of chemical reactions, many chemists consider them as a peculiar class of salts, which they term haloid salts ; and call halogen bodies, or halogens, those bodies, simple or compound, which form hydracids with hydrogen, and, consequently, haloid salts with the metals. We shall not adopt this view, for it is incompatible with the definition we have given of the word salt, a definition we think proper to preserve with precision. Moreover, the binary compounds we are now consider- ing present no analogy with the salts, except when they are soluble in water, and subjected to chemical reaction in this liquid. § 352. The salts are nearly all solid at the ordinary tempera- ture. Those resulting from the combination of a colourless acid with a colourless base are themselves colourless; those formed of a coloured base with various colourless acids are coloured, and present nearly the same colour when crystallized in water. Salts formed by colourless bases with the same coloured acid, generally approximate to the colour of the free acid. The taste of soluble salts depends most frequently on the base; thus, the salts of soda have a decided saline flavour, resembling that of common salt; the salts of potassa have a slightly bitter, saline taste; those of magnesia are insufferably bitter; those of alumina are sweet and astringent, etc. Sometimes, however, the flavour of the salt is strongly affected by the nature of the acid, as in the sulphites, those formed by metallic acids, sulphosalts, etc. § 353. Many salts may be obtained either in the anhydrous state or combined with a certain quantity of water. A great number of soluble salts, when deposited from solution, retain water in combination, called water of crystallization, the quantity of which is always the same in the same salt, when crystallized at the same temperature and in an identical solution, and presents a simple ratio in equivalents with the equivalents of the acid and base which enter into the composition of the salt. Thus, the water of crystallization of salts follows the laws of combination in definite proportions, which we have observed in all other chemical compounds. § -354. The same salt frequently combines with very different quantities of water, when deposited from the same solution, but at different temperatures. Thus, sulphate of soda takes 10 equi- valents of water, when crystallized in an aqueous solution, at a temperature below 91°; but is deposited in an anhydrous state if SALTS, 399 the temperature of the liquid be above 91°. Protosulphate of manganese, crystallized in an aqueous solution, at a temperature below 43°, has for its formula Mn0,S03+7H0 ; when crystallized between 43° and 68°, its formula is Mn0,S03-f-6H0 ; and, lastly, wrhen crystallized between 68° and 86°, it has only four equivalents of water, and its formula is Mn0,S03+4II0. In these different states of hydration, the crystals of the sulphate of manganese pre- sent very different and incompatible crystalline forms, showing that water of crystallization influences the crystalline form in the same way as the other elements of the salt. The sulphate of man- ganese Mn0,S03+7H0 soon loses its transparency, and at the temperature of 50°, effloresces and falls into powder. In a short time, it contains only six equivalents of water. Thus, even in the solid state, the salt has assumed the composition peculiar to it at this temperature, and with 'which it would have been deposited had it crystallized in a solution at the temperature of 50°. So also, the sulphate Mn0,S03-f 6HO, exposed for a long time to the tem- perature of 86°, falls to pieces, and assumes the composition Mn0,S03-f-4H0. If this last salt be heated to a temperature of about 212°, it again loses three equivalents of water; but it retains the last equivalent, which can be abstracted only by heating it above 392°. Thus, the same sulphate of manganese has hitherto been obtained with the following compositions : Mn0,S03 anhydrous sulphate; crystallized salt heated to 572°, Mn0,S03-f HO crystallized sulphate, heated to 248°, Mn0,S03+4H0 crystallized between 68° and 86°, Mn0,S03-(-6H0 crystallized between 43° and 68°, Mn0,S03-f-7H0 crystallized below + 43°. § 355. The hydrated salts can therefore abandon successively their water of crystallization as the temperature rises. It is na- tural to suppose that the water which is disengaged at the lowest temperature is retained in the compound by a more feeble affinity than that which resists. It is hence evident that it is interesting to study carefully these successive dehydrations of various salts,, in order to assign to each portion of water the part which actually belongs to it. We shall even have occasion, subsequently, to re- mark that a hydrated salt cannot always completely lose its water without an entire modification of its composition and chemical qualities. Thus, the formula of common phosphate of soda crys- tallized at a low temperature is (2NaO),POs-f25HO. It effloresces in the air, losing a portion of its water; and if crystallized at about 86°, it combines with less water, and the crystals, no longer efflo- rescent in the air, present the formula (2NaO)POs-f-17HO. If the same salt be heated to about 300°, a phosphate is obtained, (2Na0)P05-f HO, with only 1 equivalent of water. But, if these 400 THE METALS. variously hydrated salts be dissolved in water, and again crystal- lized at a low temperature, the same primitive salt (2NaO)PO + 25110 is obtained. Thus, the successive dehydrations which the salt has undergone do not prevent it from assuming its original composition wdien brought into contact with water. But if the phosphate of soda be heated to a dull red-heat, it loses its last equivalent of water, and its composition is entirely changed; for, upon solution in water and recrystallization, the ordinary hydrated phosphates are not obtained, but salts entirely different in their forms and chemical reactions. The last equivalent of water in this salt, therefore, plays a much more important part than the others, since it cannot be driven off without entirely changing the nature of the salt. This last equivalent of water is called the water of constitution, and all the others water of crystallization. § 356. Many salts lose a portion of their crystal-water when exposed to the air at ordinary temperatures, if this air is not satu- rated with moisture, and part with it more readily when the air is perfectly dry. The dehydration of a salt may often be pushed very far by keeping it in vacuo under a bell-glass, near a dish containing oil of vitriol. If we wish to ascertain exactly the quan- tity of water lost by the salt under these circumstances, a certain quantity of the finely powdered salt is weighed in a small capsule, and placed under the receiver of an air-pump, over a larger capsule con- taining oil of vitriol. After remaining 24 hours in vacuo, the cap- sule is again weighed, and the difference expresses the water lost. Upon replacing it in the vacuum, and weighing it at the end of 12 hours, if it has not experienced an additional loss of weight, it is certain that the salt has parted with all the water it can lose under the circumstances. But if there has been a diminution of weight, the capsule must be replaced a third time, and so on, until no change of weight between two con- secutive weighings can be observed. § 357. In order to ascertain the quantity of water which a salt gives off successively, at different tem- peratures, a small oil-stove or bath (fig. 310) is frequently used in the laboratory, and consists of a double upper box, with a door on one side, and the space between the sides filled with a fixed oil. The stem of a ther- mometer, passing through the tubu- lure a, has its bulb in the oil-bath, to indicate the temperature. The stove is heated by a small furnace until the thermometer marks the temperature required, which is kept nearly sta- Fig. 310. SALTS 401 tionary by regulating the furnace. The dish, containing an exact weight of the salt to be dried, is placed in the small chamber of the stove, and the door closed. It is difficult to ascertain in this manner the precise temperature at which the desiccation of the salt takes place, since it may differ essentially from that indicated by the thermometer; and, in order to operate with greater precision, the process detailed in § 261, for oxalic acid, must be adopted. § 358. Salts containing a great deal of crystal-water often fuse when heated, experiencing what is called the aqueous fusion; and the fused substance may be considered as a solution of the anhydrous salt in the crystal-water of the salt. By continuing the heat, the water of crystallization gradually escapes ; the sub- stance dries, and may in its turn fuse, if the temperature be suffi- ciently high and the salt can support it without decomposition. The anhydrous salt is then said to undergo the igneous fusion. § 359. Certain anhydrous salts, such as common salt, exhibit slight detonations when thrown on burning coals, and are then said to decrepitate. The decrepitation of crystals is often occa- sioned by a small quantity of water, interposed between the crys- talline laminae, being suddenly converted into vapour by the heat, producing a series of small detonations. Decrepitation is often owing, also, to the bad conducting power of the salt for heat, which results in a host of small fractures in each individual crystal, ac- companied by explosion. § 360. Action of Electricity.—The electric battery readily decom- poses salts, particularly when dissolved in water. If the battery be powerful, the decomposition may be very complex, effecting a separation even of the elements; but if it be feeble, the acid merely separates from the base, seeking the positive pole, while the base repairs to the negative. The decomposition is evident, if the ex- periment be conducted as follows: A solution of a neutral salt, as the sulphate of potassa, is poured into the curved tube abc (fig. 311), and coloured with a small quantity of syrup of violets. The colouring matter of the syrup is reddened by acids and greened by alkalies. The two poles of the battery, terminating in platinum wire, are inserted into the open ends of the U-tube. The liquid becomes red in the leg aby at the positive pole, and green in the leg be, or the negative pole. In a short time the separation is well marked, and continues while the battery is acting ; but if the wires be re- moved, the liquids in the two legs mix slowly, reproducing sul- phate of potassa, and the colouring matter assumes its original violet hue. The same effect would ensue immediately if the tube were shaken so as to mix the liquid in the two legs more rapidly. Fig. 311. 402 THE METALS. SOLUBILITY OF SALTS. § 361. The study of the solubility of salts in various liquids is one of the most important in chemistry. In fact, on the differ- ence of their solubility are founded the processes by which they are separated when mixed together, as well as various modes of preparing them. Water is the most usual and important solvent of salts, as it dis- solves a great number of them, and often in considerable quantity. Some salts likewise dissolve in alcohol and wTood-spirit, and they are generally such as are very soluble in water. The solubility of salts in liquids varying with the temperature, it is necessary to determine it for the different degrees of the ther- mometric scale, from the lowest temperature to that at which the saturated solution boils under the ordinary pressure of the atmo- sphere. It would even be very interesting to study the solubility of salts at more elevated temperatures, by operating in close ves- sels, in which the pressure could be increased at pleasure, and consequently the boiling point of the liquid raised; but this has never yet been done. The solubility of salts generally increases wdth the temperature ; but we shall have occasion to point out some exceptions to the rule. § 362. A saturated solution of a salt at a given temperature may be obtained in two ways. The solvent may be poured on a great excess of salt, so that fragments of the latter may rise above the level of the liquid, and the whole kept for several hours at the temperature required. The decanted liquid then contains all the salt it can dissolve at that temperature, and is said to be satu- rated. The solution of a salt may also be effected at a temperature higher than that at which we wish to ascertain its solubility, and the liquid allowed to cool slowly until it reaches this temperature, when it is kept stationary for 15 minutes. A portion of the salt is deposited during the cooling of the liquid, and that quantity only which it can dissolve at the desired temperature remains in solution. Experience has shown that the same coefficient of solu- bility is obtained for the same salt by adopting either of these pro- •cesses. The second, however, requires some precaution. It has been observed that a liquid, when not in contact with perfectly formed crystals of the salt which it contains, may retain a much larger portion of the salt than corresponds to its normal solubility at that temperature. The saturated solution of certain salts, more soluble in hot than in cold, may be cooled several degrees without depositing crystals; but if a small crystal of the supersaturating salt be dropped into the solution, the excess of the salt crystallizes immediately, and in a few moments the liquid contains only the SALTS 403 normal quantity of the salt it dissolves at that temperature. Such abnormal solubilities are therefore never observed when the liquid is allowed to remain in contact with an excess of the salt. Agitation of the supersaturated liquid, or the introduction of a foreign body, particularly if the latter present projecting points, fre- quently effects the separation of the excess of dissolved salt. The phe- nomenon is analogous to that observed in the congelation of liquids, and may be attributed to the same cause, namely, a certain diffi- culty experienced by the saline molecules of moving in the liquid and assuming an arrangement suitable to crystalline aggregation. In this way water may be cooled several degrees below the ordi- nary temperature of its congelation without becoming solid, when the vessel containing it is in a state of absolute rest; but if a small piece of ice or of pointed glass be thrown in, congelation immedi- ately ensues. § 363. Sulphate of soda presents a remarkable instance of the inertia of saline molecules in solution. Its solubility increases rapidly with the temperature from 32° to 91J, but from 91J it diminishes with the increasing temperature, although more slowly than it had increased from 32° to 91J; and, at the boiling point, the liquid contains a much more considerable proportion of salt than at ordinary temperatures. If a thin stratum of oil, or spirit of turpentine, be poured over a hot saturated solution of sulphate of soda, and the liquid be allowed to cool slowly and quietly, it will not deposit crystals, even at temperature at which the liquid could have originally contained only one-half of the salt, by virtue of its nominal solubility. But if a pointed piece of glass be plunged through the stratum of oil into the saline solution, crys- tallization commences immediately. A still more striking experiment may be made on the same salt. A solution of it, saturated when hot, is poured into a funnel-shaped glass tube (fig. 312) so as to fill about f of ab. Being made to boil for a few moments to expel the air, and a feeble ebullition still maintained, the narrow part e is rapidly closed by the blow- pipe. The tube being allowed to cool, the solution may then be cooled to 32° without crystalliz- ing ; and yet it contains ten times more salt than it could dissolve by its normal solvent power. The tube may even be shaken without crystallization taking place ; but if the narrow portion be suddenly broken, the salt in- stantly crystallizes, and the li- quid becomes solid. At the same Fig. 312. 404 THE METALS. time the tube becomes sensibly warm to the hand. The disengage- ment of heat is due to the fact that all substances give off heat in passing from the liquid to the solid state. Now, the dissolved sul- phate of soda was liquid; and by solidifying into crystals, heat was disengaged. A similar effect ensues whenever a salt crystal- lizes from solution; but is only appreciable when the crystalliza- tion is copious and rapid. If crystallization take place slowly, such as during the gradual cooling of a liquid, the heat disengaged by the solidification of the salt only retards the rapidity of cooling. If crystallization take place by spontaneous evaporation, it is still slower, because the evaporation of the liquid carries off heat, and that given off by the crystals in solidifying cannot be appreciated but by most delicate experiments. § 364. To determine the solubility of a salt in water at a given temperature, we always endeavour to find the quantity of salt con- tained in a saturated solution at that temperature. The solution is prepared by one of the processes indicated above, taking care to keep it for at least half an hour in the presence of an excess of crystallized salt, at the temperature required. About 50 grammes of the liquid are poured into a flask (fig. 313), the neck of which may be about 2 decimetres in length, and its exact weight rapidly ascertained. The liquid is evaporated over a small furnace by boiling, taking care to keep the neck of the flask inclined at an angle of 45°, in order to avoid loss of the salt. When the liquid Fig. 313. is evaporated, the flask is still heated until the salt has lost, not only the water which dissolved it, but also its water of crystalliza- tion. To drive off the last traces of moisture, a glass tube attached to the nozzle of a bellows is introduced into the flask, and air gently blown into it, the current of which completely removes the moisture. The flask being weighed after cooling, gives the weight of the anhydrous salt contained in the solution. Let P be the weight of the solution subjected to evaporation, p the weight of the anhydrous salt obtained; (P —p) will be the SALTS, 405 weight of the water. A weight (P—p) of water dissolves a weight p of anhydrous salt; consequently, 100 parts of water dissolve 100 . jrrj of anhydrous salt, at a known temperature T. If the crystallized salt contains crystal-water, we may inquire what is the smallest quantity of water which can dissolve a given weight of it at the temperature T. Let n be the wejght of crystal- water required by a weighty of anhydrous salt to form (p+*) of hydrated salt: the quantity of water which dissolves the weight (p+7t) of hydrated salt is evidently (P—p—*). Therefore, a weight (P —p—rt.) of water dissolves a weight (p+rf) of hydrated salt to form a liquid saturated at the temperature T. One hundred parts of water will therefore form a liquid saturated at the tem- perature T with a weight 100 . °f crystallized hydrated salt: or, again, 100 parts of crystallized hydrated salt will dissolve in a weight 100 . 1~^7r of water. § 365. The solubility, at various temperatures, of a salt contain- ing water of crystallization may be expressed in two ways ; either by the quantity of water contained in a solution of the salt, satu- rated at those temperatures ; or by the quantity of water required to dissolve a certain weight of hydrated salt, and to obtain a satu- rated solution at the temperature T. In the former case, the so- lubility is referred to the anhydrous salt, and the water of crys- tallization is considered as co-operating in the solution. In the latter case, it is confidently assumed that the salt still exists in the solution in the hydrated state, and that the water added acts only as a solvent. Many hydrated salts fuse in their crystal-water, at a higher or lower temperature, undergoing what is called the aqueous fusion. At the temperature which effects the fusion, it is evident that a weight Jt of water dissolves a weight p of anhydrous salt; but the solubility of the crystallized salt is infinite at this temperature; for a gramme of water would dissolve at this temperature an in- definite quantity of crystallized salt, because the salt dissolves in its own water of crystallization. § 366. It is frequently more easy and more accurate, instead of evaporating a saline solution to determine the proportion of anhy- drous salt it contains, to treat the salt chemically, by forcing one of its constituents into an insoluble combination. Thus, to determine the quantity of sulphate of soda which a liquid contains, a few grammes of this solution may be weighed, diluted with an indefi- nite quantity of water, and treated with an excess of chloride of barium. The precipitated sulphate of baryta being collected on a filter, washed, and weighed after calcination, we can infer, from its weight, the weight of anhydrous sulphate of soda from which it was produced. For let p be the weight of the sulphate of baryta, the composition of which is 406 THE METALS 1 eq. baryta 76-5 1 “ sulphuric acid 40-0 1 “ sulphate of baryta 116-5 A weight p of sulphate of baryta, therefore, corresponds to p. of sulphuric acid. The sulphate of soda contains 1 eq. of soda 81 1 “ sulphuric acid 40 1 “ anhydrous sulphate of soda 71 The weight of sulphate of soda, which corresponds to the weight p. jjjL of sulphuric acid, and consequently to the weight p of sul- phate of baryta, is given by the proportion 40 : 71:: v. AL : x, whence x=p. 116.5 ' 110.5 The same mode of treatment will serve to ascertain the solu- bility of any sulphate whatever. Reciprocally, the solubility of a salt of baryta may be deter- mined by precipitating the baryta by a soluble sulphate, and calculating the proportion of the salt of baryta, of which the com- position is known, from the weight of sulphate of baryta obtained. The solubility of a chloride may be ascertained by precipitating the chlorine in the state of chloride of silver. There are even salts in which this is the only plan that can be used; such as those which are decomposed by heat before reaching the anhydrous state, and which oxidize readily by contact with the air. For ex- ample, chloride of magnesium, dissolved in water, cannot be brought to the anhydrous state without partial decomposition, so that its solubility cannot be accurately determined by the general method, founded on evaporation, and explained in (§ 364). § 367. Let us suppose that we have thus determined the solu- bility of the same salt in water at all temperatures, from the lowest unto that at which its saturated solution boils under the ordinary pressure of the atmosphere: we may represent the ratio of solubility to the temperature, by a mathematical curve, count- ing the temperature on the line of the abscissas, and marking on the corresponding ordinates lengths proportional to the quantity of salts dissolved by the same weight of water. This curve may be constructed with sufficient precision, when a certain number of direct determinations of solubility (8 or 10) have been made at sufficient distances apart in the scale of temperatures, and the curve can afterward be used to ascertain the solubilities at all intermediate temperatures. The annexed plate represents the curves of solubility of a great number of salts. The horizontal line AX is divided into 110 equal SALTS. 407 parts, each of which represents one degree of the centigrade ther- mometer ; the temperature of melting ice corresponding to the zero of the division. 100 equal divisions are marked on the vertical line AY, but are not necessarily equal to those of the horizontal line AX. Let us suppose that it is required to construct the curve of solubility of sulphate of soda in water. Direct experiments have given us the following numbers : Temperature. Anhydrous salt dissolved Crystallized salt dissolved Centig. Therm. in 100 pts. of water. in 100 pts. of water. 0.00° 5.02 12.17 11.67 10.12 26.38 13.30 11.74 31.33 17.91 16.73 48.28 25.05 28.11 99.48 28.76 37.35 161.53 30.75 43.05 215.77 31.84 47.37 270.22 32.73 50.65 322.12 33.88 50.04 312.11 40.15 48.78 291.44 45.04 47.81 276.91 50.40 46.82 262.35 59.79 45.42 244.30 70.61 44.35 229.70 84.42 42.96 217.30 103.17 42.65 210.20 The temperatures inscribed in the first column of the table are marked on the line of abscissas, and on the corresponding ordinates a number of divisions are taken, equal to that which represents the number of grammes of salt dissolved by 100 grammes of water. These numbers are contained in the second column, for the solu- bility of the anhydrous salt; and in the third column, for the solubility of the crystallized, hydrated salt. The numbers in the second column are obtained directly by ex- periment. Those in the third are thence deduced, as follows: The equivalent of the anhydrous sulphate of soda is 71, and the composition of crystallized sulphate is, 1 eq. of anhydrous sulphate of soda 71 10 “ of water of crystallization 90 1 “ of crystallized sulphate of soda 161 Let us suppose, that at the temperature T, 100 grammes of water dissolve p grammes of anhydrous sulphate of soda. Then p grammes of anhydrous salt correspond to p. of crystallized salt, and require p. ff grammes of water to be changed into the 408 THE METALS. crystallized salt; we may therefore say that p. ~ of crystallized sulphate of soda are dissolved in (100—p. ff) grammes of water. Consequently, we can find the weight of crystallized salt dissolved by 100 grammes of water, by making the proportion: 100—p p .l3 :: 100 : x, whence 2=100 p. . —I—. 71 100-p._ 71 § 368. The curve of solubility of sulphate of soda may be con- structed on our plate in the ordinary manner, between the tem- peratures of 0° C. (32° F.) and 25° C. (77° F.), and is represented by the curve line BC, between these extremes of temperature. But, for temperatures above 24° C. (77° F.), 100 parts of water dissolving more than 100 parts of crystallized salt, the ordinates become greater than 100, and can no longer be marked on our plate. Nevertheless, if we suppose a second plate, similar to the first, placed above it, we shall have in all 200 vertical divisions, and the construction of the curve may be continued. Above 30° C. (86° F.), the ordinates become greater than 200, and, if we wish to continue the curve, we must superpose a third plate on the second, and so on. Now, let us suppose that the second, third, and fourth plates, after having been placed end to end above the first, are subsequently superposed on it; the branch of the curve of which the ordinates are comprised between 100 and 200 will then assume the direction DE ; the branch with its ordinates com- prised between 200 and 300 will have the direction FG; the branch with its ordinates greater than 300 will be at HIK; lastly, the branch of which the ordinates are comprised between 300 and 200, and corresponding to the temperatures between 36° C. (96.8° F.) and 111° C. (231.8° F.) will be at LM. In order to obtain the real ordinates of the branch DE, we must add 100 to the ordinates measured immediately on the plate. The real ordinates of the branches FG and LM will be obtained by adding 200 to the apparent ordinates measured on the plate. Lastly, we obtain the ordinates of the branch HIK, by adding 300 to the ordinates measured on the plate. The same mode of construction has been used for the curves of solution of salts Avhose solubility is greater than 100, beyond a certain temperature, such as nitrate of potassa. § 369. The solubility of a considerable number of salts increases nearly in proportion to the temperature, so that their curve of solubility scarcely differs from a right line. Sometimes this right line is only slightly inclined to the line of abscissas; as is the case with chloride of sodium, the solubility of which does not sensibly increase with the temperature. The right lines representing the solubility of sulphate of potassa, chloride of potassium, chloride of barium, and sulphate of magnesia, are more inclined to the line of SALTS, 409 abscissas. The curves of solubility of nitrate of potassa, nitrate of baryta, and chlorate of potassa, turn their convexity toward the axis of the abscissas : the curve of the nitrate of potassa rises very rapidly, in proportion as the abscissas increase. The curve of solubility of sulphate of soda presents a very re- markable form, rising rapidly between 0° C. (32° F.) and 33° C. (91° F.), and at about 33° C. presenting a point of retrogression from which the curve descends towards the axis of the abscissas, always turning its convexity towards this axis. The singular point presented by the curve of solubility of the sulphate of soda for the abscissa of 33° C. (91° F.) corresponds with a remarkable change which takes place at this temperature in the constitution of the salt. In fact, if it be crystallized by evaporation from a liquid maintained at a temperature below 33° C. (91° F.), the salt always crystallizes in the hydrated state NaO,SO3-f-10IIO. But, if the same solution be crystallized above 33° C., the salt is always de- posited in the anhydrous state NaO,S03. Thus, the discontinuity which we observe in the curve of solubility for the abscissa of 33° C., coincides with a change of the constitution of the salt at this temperature. The first branch comprised between the ab- scissas 0° C. and 33° C. relates to the hydrated salt Na0,S03+ 10HO; the second branch between 33° 0. and the abscissa cor- responding to the boiling point of the saturated liquid, refers to another salt, the anhydrous sulphate of soda Na0,S03. §370. A knowledge of the relative solubility of the various salts, at different temperatures, is of deep interest, because it ena- bles us to foretell the order in which these salts crystallize at a given temperature when their solutions are evaporated. Let us suppose a mixture of only two salts, the nitrate of potassa and chloride of sodium. These two salts present equal solubilities at the temperature of 23.6° C. (74.48° F.), which is the abscissa cor- responding to the crossing of their curves. Below 23.6° C. the solubility of the nitrate of potassa is less than that of the chloride of sodium, whilst above this temperature it is more soluble. It therefore follows that, if a solution containing equal proportions of the two salts be evaporated at a temperature below 23.6° C., nitrate of potassa will crystallize first; and, on the contrary, if the solution be evaporated by heat, chloride of sodium will be first deposited. The inversion, in the order of solubility of salts with the tem- perature, frequently determines the double decompositions em- ployed in the arts, of which chloride of sodium and sulphate of magnesia afford a remarkable example. If a liquid containing chloi'ide of sodium and sulphate of magnesia be evaporated at a temperature above 53|-°, the chloride of sodium is deposited in crystals, and the sulphate of magnesia remains in solution. If, on the contrary, the evaporation takes place below 45°, or better still, 410 THE METALS. if the liquid saturated at 59° be cooled to about 32°, crystals of sulphate of soda are deposited, and chloride of magnesium remains in solution. There is, in this last case, a double decomposition. § 371. It is, however, important to remark, that what has just been said of the solubility of salts refers only to their solution in pure water, and that their solubility may be very different in water already containing other salts. Thus, a solution of nitrate of po- tassa, saturated at a given temperature, cannot dissolve an ad- ditional quantity of nitrate at this temperature ; but it will dissolve a considerable quantity if a certain proportion of common salt has been previously added. So that the solubility of nitrate of potassa is greater in salt water than in fresh. The solubility of the nitrate of potassa is, on the contrary, less in a solution of chloride of po- tassium than in pure water, for in dissolving the latter in a liquid saturated with nitrate of potassa, it precipitates a portion of the nitrate in small crystals. Experience has shown that when two salts differ both in their acid and their base, and that a double decomposition can take place, the presence of one of these salts may favour the solubility of the other. In this way, the presence of chloride of sodium favours the solubility of nitrate of potassa, because nitrate of soda and chloride of potassium are formed, which are respectively more so- luble than nitrate of potassa and chloride of sodium, at least at temperatures above 77°. When, on the contrary, the two salts con- tain the same base or the same acid, there can be no double decom- position, and the presence of one of the salts in the solution dimi- nishes the solubility of the other. For this reason, a solution of chloride of potassium dissolves less nitrate of potassa than pure water. We must, however, except the case in which the two salts combine to form a double salt possessing peculiar solubility. §372. Saline solutions boil at higher temperatures than pure water; the difference, with the same salt, is in proportion to the quantity of it in solution. The boiling point of a saline solution should be measured by a thermometer with its bulb kept in the boiling liquid; for if it were placed only in the vapour, at some distance above the liquid, it would indicate a temperature very little above 212°. The following table contains the boiling point of several satu- rated saline solutions: Names of the salts. Proportion of salt in 100 of water. Boiling point. Chlorate of potassa .... 61.5 ... 219.56° Chloride of barium .... 60.1 ... 219.92° Carbonate of soda 48.5 ... 220.28° Chloride of potassium., 49.4 ... 226.94° Chloride of sodium 41.2 ... 227.12° Chlorohydrate of ammonia 88.9 ... 237.56° SALTS. 411 Names of the salts. Proportion of salt in 100 of water. Boiling point. Nitrate of potassa 335.1 .... 240.62° Chloride of strontium 117.5 .... 244.04° Nitrate of soda 224.8 .... 249.8° Carbonate of potassa. 205.0 .... 275.0° Nitrate of lime 362.0 .... 303.8° Chloride of calcium ... 325.0 .... 355.1° §373. The solution of salts, or of any other substances in water, is accompanied sometimes by depression, sometimes by elevation of temperature. A substance which has crystallized from an aqueous solution at a low temperature, and which contains, con- sequently, all the combined water it can assume at this tempera- ture, produces cold by resolution in water, at the same or higher temperatures. The production of cold is owing to an absorption of heat from the disaggregation of the salt which, by dissolving, passes from the solid to the liquid state. The heat may be re- garded as a species of latent heat of fusion of the salt, but is pro- bably very different from the latent heat of fusion properly so called, that is, the heat which the substance absorbs when it under- goes the igneous fusion. We shall give it the name of latent heat of solution of a salt. Sulphate of soda, crystallized at a low temperature, accord- ing to the formula Na0,S03+ 10IIO, produces cold by dissolving in water; and the same takes place with crystallized chloride of calcium CaCl+6HO. The salts which crystallize when cold with- out any water of crystallization, as the chlorides of potassium and sodium, produce likewise a depression of temperature by solution. The quantity of heat which equal weights of various substances absorb by dissolving in water, is often very different, even when they are analogous in the aggregate of their properties. Thus, 50 grammes of common salt, dissolved in 200 centimetres of water, produce a depression of temperature of 3.42°, while 50 grammes of chloride of potassium depress the temperature 24.52° when dissolved in the same quantity of water. Anhydrous salts, which combine with water of crystallization when separating from an aqueous solution at a low temperature, generally produce heat, when dissolved in water in their anhydrous state. Thus, anhydrous sulphate of soda and anhydrous chloride of calcium produce a considerable elevation of temperature by so- lution in water. There is, in that case, a superposition of two effects: 1st. A disengagement of heat due to the combination of the anhydrous substance with water ; 2dly. An absorption of heat produced by the solution of the hydrated body in the same liquid. Accordingly, as one of these effects predominates over the other, there is absorption or disengagement of heat. § 374. Advantage is often taken of the absorption of heat pro- 412 THE METALS. duced by the solution of certain substances in water, to obtain refrigerating mixtures. By effecting the solution in the coldest water we can procure, we can lower its temperature to several de- grees below 32°. Thus, by dissolving 1 part of chloride of potas- sium in 4 parts of water at 50°, a solution is obtained at the temperature of 29J°. If the solvent water is at 32°, the liquid marks 11|° after solution. The depression of temperature is often greater and more rapid when, instead of dissolving the salt in pure water, it is dissolved in an acid liquid. Thus, by dissolving crystallized sulphate of soda in a solution of chlorohydric acid, a depression is obtained of 45° or 55°. On this property a process has been founded for procuring ice at all seasons. The apparatus used, and known by the name of the family ice-box, is represented in figs. 314 and 315. It is composed of a hollow cylinder C, destined to receive the re- frigerating cylinder I, for containing water, which becomes a hol- low cylinder of ice from the effect of the internal refrigerant. Into the refrigerating mixture itself is inserted another cylindrical vessel A, closed at bottom, and turned by a winch, and which, by means of suitable projections, agitates the mixture, and renews the points of contact of the refrigerating body with the inner and outer vessels. If the hollow vessel be filled with water, the latter freezes like the surrounding water. In order to prevent its being warmed by the circumambient air, the space I is surrounded by an envelope of some bad conductor, as cotton or tow. The cover D of the inner cylinder A is likewise stuffed. In order to obtain the maximum effect, it is advisable to cause the Avhole refrigerating mixture to act gradually. 1500gin lb.) of sulphate of soda and 1200sm (2-§ lb.) of chlorohydric acid are first introduced into Fig. 314. Fig. 315. SALTS. 413 the cylinder C, when the temperature of the water to be frozen will fall, in 5 or 6 minutes, from +77° to 32°. The liquid mix- ture is then allowed to flow into the lower vessel Y, by opening the stopper s by means of the lever mn. Another quantity of the mixture, equal to the first, is then introduced into the cylinder C, allowed to act for 15 minutes, and again run off into the vessel Y. The third and fourth quantities of the mixture should like- wise act for 15 minutes. Thus, we have used altogether 6 kil. (13 lb.) of sulphate of soda, and about 5 kil. (11 lb.) of chlorohy- dric acid; the operation has lasted an hour, and 5 or 6 kil. (11-13 lb.) of ice have been obtained. The cold liquid collected in the lower vessel Y may be used for cooling bottles of wine. Many bodies soluble in water, when brought into contact with ice, melt it rapidly, and dissolve in the water thus produced. A considerable depression of temperature is thus obtained, depending at the same time on the latent heat of solution of the salt and the latent heat of fusion of the ice. By mixing pulverized sea- salt and pounded ice, a mixture is obtained which reduces the tem- perature to 6°. By mixing finely powdered chloride of calcium with snow or pounded ice, the temperature falls to —49°. A considerable depression of temperature can also be produced by adding ice to a cold and concentrated solution of chloride of calcium, in which it melts rapidly, and the temperature may fall to -22°. OF THE DECOMPOSING POWER EXERTED BY ACIDS ON SALTS, AND THE BINARY COMPOUNDS RESULTING FROM THE UNION OF ME- TALS WITH METALLOIDS. § 375. The reactions which the various acids exert on salts, and the binary compounds resulting from the reaction of hydracids on oxybases, may be foreseen from certain general laws which observation has proved, and which will now be explained. If the reacting acid is identical with that already in the salt, it often happens that the salt combines with a new quantity of acid, and a salt is formed with an excess of acid. If sulphuric acid be added to sulphate of potassa KO,SOs, the bisulphate of potassa K0,2S03 is formed. So, also, if a current of carbonic acid gas be passed through a solution of neutral carbonate of potassa K0,C02, the bicarbonate of potassa K0,2C03 is produced. If the base of the salt does not combine with a greater quantity of acid, the salt often dissolves in the acid added, especially if the latter be mixed with a large quantity of water. Thus, the nitrate of potassa dissolves in a dilute solution of nitric acid; but if the liquid be evaporated, the nitrate crystallizes unchanged. § 376. If the reacting acid differs from that existing in the salt, decomposition will ensue under several circumstances. 414 THE METALS. Decomposition will ensue when, the salt being soluble in water, the reacting acid can form an insoluble compound with its base. By pouring sulphuric acid into a solution of nitrate of baryta, sulphate of baryta is immediately precipitated, and nitric acid set free in the liquid. If the base of the salt forms a soluble salt with the new acid, and the reaction takes place in sufficient water to dissolve one or the other salt, it cannot, in general, be decided whether a new salt has formed or the first has remained unchanged in the liquid. But, if the new salt is less soluble than the original salt, the decomposition can always be effected by evaporating the liquid to a point when the new salt can no longer remain in solu- tion. The new salt is then deposited by virtue of the principle announced; for it is actually insoluble in the liquid at the degree of concentration given to it. If sulphuric acid be poured into a dilute solution of nitrate of potassa, no signs of decomposition are apparent; but, if the liquid be properly evaporated, sulphate of potassa is deposited, because it is less soluble than the nitrate, especially at an elevated tem- perature. Nitric acid may, on the contrary, decompose sulphate of potassa, if the evaporation takes place at a very low tempera- ture ; for at 32° the nitrate is less soluble than the sulphate. Similar reactions take place between hydracids and salts, or between oxacids and the binary compounds of metals with the me- talloids which form hydracids with hydrogen; and they are deter- mined by the same circumstance of insolubility. By pouring clilorohydric acid into a solution of sulphate of silver, chloride of silver is precipitated, and the liquid contains free sulphuric acid: Ag0,S08+HCl+nH0=AgCl+S0,+(ra+l)H0. Again, if chlorohydric acid be poured into a solution of nitrate of lead, chloride of lead is deposited in small crystalline scales; but if the liquid is much diluted, there is still water enough to dissolve the chloride of lead, and nothing evinces the occurrence of decom- position; it soon, however, becomes apparent, if the liquid be evaporated to the proper degree. § 377. Sometimes decomposition is determined by the insolubility of the acid which exists in the salt. If sulphuric or nitric acid be poured into a concentrated solution of borate of soda, sulphate or nitrate of soda is produced, and boracic acid is precipitated in small crystalline scales. When the liquid is sufficiently dilute to dissolve boracic acid, the decomposition does not manifest itself immediately by visible signs: it is easily seen, however, that de- composition has taken place, even in the dilute liquid. It will be sufficient to remember that boracic acid acts on litmus only like a feeble acid, producing a purplish red, while sulphuric and nitric acids produce a bright red colour. If, therefore, the first drops of sulphuric or nitric acid added have remained free, the liquid 415 SALTS. should produce with litmus the bright red colour; but if they have decomposed a corresponding quantity of borate of soda, by libe- rating boracic acid, the liquid should assume a purplish red hue. Now, it is observed that the tincture becomes vinous red on the addition of the first drops of acid, and preserves this colour until the borate is entirely changed into sulphate. The addition of the least drop of sulphuric acid then changes the tincture to a bright red. Here the reaction has not been produced by the insolubility of the boracic acid, but by the fact that sulphuric and nitric arc much more powerful than boracic acid. § 378. A salt can always be decomposed by an acid less volatile than that ivhich it contains. Carbonic acid is gaseous at ordinary temperatures, and is but slightly soluble in water. Nitric acid dissolved in water has its boiling point above 212°; so that it will readily expel carbonic acid, even in the cold. All the carbonates are, in fact, decomposed by nitric acid. A similar decomposition of the carbonates is effected by the hydracids, such as chlorohydric acid, which is gaseous at ordinary temperatures; but as it is very soluble in water, and its solution boils above 212°, it must drive off carbonic acid. Since aqueous nitric acid boils at some degrees above 212°, and concentrated sulphuric acid above 600°, the latter, under the in- fluence of heat, will readily expel nitric acid from all its com- pounds. Sulphuric and phosphoric are two powerful acids; but as the latter is still less volatile than oil of vitriol, it readily expels sul- phuric acid, by the assistance of heat. It was observed that sulphuric acid decomposes the borates in solution in the cold; but boracic, being a much more fixed acid, decomposes all the sulphates at a high temperature. Silica behaves like a very feeble acid in solutions; for the solu- ble alkaline silicates are decomposed by the most feeble acids, even by carbonic. On the other hand, with the assistance of heat, silicic acid expels all other acids. The reactions exerted by the various acids on a salt depend on the nature of the liquid in which this salt is dissolved; for, the order of solubility may be entirely inverted in passing from one solvent to another. If acetic acid be poured into an aqueous solu- tion of carbonate of soda, carbonic acid is disengaged with effer- vescence. This decomposition may be attributed to two causes: the acetic is a stronger acid than the carbonic, and the latter is gaseous at ordinary temperatures, and at the same time is but slightly soluble in water. On the other hand, acetate of potassa dissolved in alcohol is decomposed by carbonic acid, owing to the insolubility of carbonate of potassa in strong alcohol. The decom- position is therefore caused by insolubility. 416 THE METALS. The state of concentration of an acid and the temperature exert a powerful influence over these reactions. If a solution of sulfhydric acid be poured into a dilute solution of chloride of anti- mony, a precipitate of sulphide of antimony is formed. But, if sulphide of antimony be heated with a concentrated solution of chlorohydric acid, chloride of antimony is formed, and sulfhydric acid disengaged. § 379. When the acid of a salt, and that employed to react on it, are both gaseous, and at the same time but slightly soluble in water, and when, moreover, their affinities for the bases are nearly equal, the acid which is present in excess will expel the other. Thus, by passing a current of carbonic acid gas for some time through the solution of an alkaline sulphide, the latter is entirely converted into a carbonate, and sulfhydric acid driven off. Reci- procally, by passing sulfhydric acid for some time through a solution of an alkaline carbonate, it is entirely converted into sul- phide of potassium. The vapour of water, at a high temperature, expels carbonic acid from alkaline carbonates, when the latter are heated in pla- tinum tubes in a current of steam, and hydrate of potassa is formed. Reciprocally, the hydrate of potassa, heated to the same tempera- ture in a current of carbonic acid, is converted into carbonate of potassa. These facts exhibit the influence of mass, analogous to that already mentioned in § 322. ACTION OF BASES ON SALTS, AND ON THE BINARY COMPOUNDS OF HYDRACIDS ON OXYBASES. § 380. When a salt is brought in contact with an additional quantity of the same base which it already contains, it frequently happens that no reaction ensues, and never occurs when the acid cannot form a salt more basic than the original. If potassa be added to a solution of sulphate of potassa, and the liquid evapo- rated, the original sulphate crystallizes anew. At other times, combination ensues; thus, potassa added to a solution of bisulpliate of potassa produces the neutral sulphate. A solution of neutral acetate of lead can dissolve an additional quantity of oxide of lead and form a basic acetate. If the base added to a saline solution is different from that ex- isting in the salt, the original salt is frequently decomposed, and a new one formed; and the decomposition is determined by cir- cumstances analogous to those which cause the reaction of acids on salts. § 381. Grenerally speaking, a soluble salt is decomposed, when the reacting base can form an insoluble salt with the acid of the salt. If baryta be added to a solution of sulphate of potassa, sul- SALTS, 417 phate of baryta is precipitated, and caustic potassa remains in the liquid. Baryta also decomposes carbonate of potassa in a dilute solution, and carbonate of baryta is precipitated. The state of concentration of the liquid exerts great influence over these decom- positions; for, if carbonate of baryta be boiled with a concentrated solution of caustic potassa, a considerable quantity of carbonic acid is abstracted from it, and carbonate of potassa formed. § 382. The decomposition is often determined by the insolubility of the base which exists in the salt. Thus, potassa decomposes nitrate of lead, and hydrated oxide of lead is precipitated. The same is true for all the salts formed by the insoluble metallic ox- ides. In this case, nevertheless, it is proper to attribute the decomposition partly to the preponderating affinity of the potassa for the acid, for potassa is a much more powerful base than such metallic oxides. § 383. An insoluble metallic oxide sometimes decomposes a salt formed by a base equally insoluble. Thus, oxide of silver decom- poses nitrate of copper in solution, precipitating oxide of copper; and the decomposition is determined, in this case, only by the preponderating affinity of oxide of silver for nitric acid. § 384. When the base of a salt is volatile, it is generally expelled by a more fixed base, particularly when assisted by heat. Thus, lime readily expels ammonia from its compounds. The same de- composition is effected with the assistance of heat, by the insoluble metallic oxides, whose salts, when in solution, are, on the contrary, decomposed by ammonia. Thus, oxide of lead, heated dry with hydrochlorate of ammonia, disengages ammonia, and chloride of lead is formed. Ammonia, on the other hand, decomposes chlo- ride of lead in solution, and precipitates oxide of lead. RECIPROCAL ACTION OF SALTS ON EACH OTHER, AND OF BINARY COMPOUNDS ON EACH OTHER AND ON SALTS. § 385. When two salts are mixed together, several phenomena may ensue: The two salts sometimes combine to form a double salt. Sul- phate of alumina combines with sulphate of potassa, forming a double salt known by the name of alum. Chloride of potassium combines with perchloride of platinum, and produces a double chloride of platinum and potassium. At other times, there is no apparent reaction of the two salts upon each other, and evaporation reproduces the two salts which have been mixed. Frequently, however, the two salts suffer mutual decomposition, which is determined by certain general circumstances,* demanding * The laws which govern the double decomposition of salts, and the reaction of the acids and bases on the salts, are called Berthollefs laws. 418 THE METALS. a careful analysis, for they generally enable us to foretell the re- actions which will ensue. We shall distinguish the case in which the two salts are heated without the contact of water, or the dry way, and that in which they are brought into contact in solution, or by the humid way. § 386. When two salts of the same acid, but of different bases, are heated together, the two salts frequently combine in definite proportions, producing double salts which crystallize on cooling. In this manner a great number of double silicates may be pro- duced which, from their beautiful crystallization, present the cha- racters of definite compounds. In the same manner, we may obtain, in the dry way, double chlorides and several other double salts; but the combination is often destroyed upon dissolving the compound in water, the two original salts crystallizing separately. § 387- When two salts of different acids and bases are heated together, and when, by the mutual interchange of acids and bases, a new salt more volatile than the first two can be formed, its forma- tion is generally determined by this circumstance. If chlorohydrate of ammonia be heated with carbonate of lime, chloride of calcium and carbonate of ammonia are formed, the latter of which is much more volatile than either of the original salts. For the same reason, sulphate of ammonia, heated with the chloride of calcium, produces chlorohydrate of ammonia Avhich volatilizes, and sulphate of lime which remains. It frequently happens that the reactions thus produced in the dry way between two salts, are precisely the inverse of those which take place in an aqueous solution. Thus, we have just seen that, by heating a mixture of chlorohydrate of ammonia and carbonate of lime, car- bonate of ammonia and chloride of calcium are formed; but if carbonate of ammonia be poured into a solution of chloride of calcium, carbonate of lime is produced, and chlorohydrate of am- monia remains in solution. In the first case, the reaction is determined by the volatility of carbonate of ammonia, and in the second by the insolubility of carbonate of lime. Mutual Action of Salts in the dry way. § 388. When solutions of two salts are mixed together, capable of producing an insoluble salt by the interchange of their acids and bases, decomposition always ensues, and the insoluble salt is precipitated. If a solution of sulphate of soda be poured into a solution of nitrate of baryta, sulphate of baryta is precipitated, and nitrate of soda remains in solution: Mutual Action of Salts in the humid way. Na0,S03+Ba0,N05=Ba0,S03+Na0,N05. SALTS, 419 So, also, if a solution of carbonate of soda be added to a solu- tion of chloride of calcium, carbonate of lime is precipitated, and chloride of sodium is formed, which remains in solution: CaCl+NaO,COa=CaO,COs+NaCl. It is not necessary for such reaction between the two salts that a salt insoluble in water should be formed from their elements, but it is sufficient that a salt less soluble than the two original salts can be produced under circumstances realizable at will. Thus, if solutions of chloride of potassium and nitrate of soda be mixed, and the liquid evaporated at a low temperature, the two salts originally mixed separate, chloride of potassium crystallizing first, and nitrate of soda remaining in the liquid. If, on the contrary, the solution be evaporated at the boiling point, a double decom- position takes place, chloride of sodium being deposited, which, at the given temperature, is the least soluble of all the compounds which can be formed by the acids and bases present, and nitrate of potassa remains in the liquid. The decanted liquid deposits crystallized nitrate of potassa on cooling. § 389. By crystallizing liquid at different temperatures, inverse decompositions may frequently be obtained. Supposing sulphuric and chlorohydric acids, soda, and magnesia to exist in solution at the same time, in such proportions that acids and bases exactly saturate each other, it may be presumed that the liquid con- tains : Either chloride of sodium and sulphate of magnesia, Or chloride of magnesium and sulphate of soda, Or both chlorides of sodium and magnesium and sulphates of soda and magnesia. It is impossible to decide in what order the acids and bases have combined in the liquid. If the solution be evaporated at a tem- perature above 59°, chloride of sodium crystallizes, being the least soluble of all the possible products at the given temperature. The greater part of the chloride of sodium may be thus separated; and if the evaporation be continued, sulphate of magnesia is obtained mixed with a small quantity of chloride of sodium. If, on the contrary, the liquid be evaporated at a low tempera- ture, as at 32°, sulphate of soda becomes the least soluble of all possible compounds, and is first deposited, while chloride of mag- nesium remains in the liquid. Thus, with the same solution, we may obtain at will, according to the temperature of evaporation, chloride of sodium or sulphate of magnesia ; or, sulphate of soda and chloride of magnesium; and we can always foretell, by consulting the plate of solubilities, page 420 THE METALS. 407, what salts will be formed at a certain temperature, and in what order they will be deposited. It is, therefore, conceivable that an exact knowledge of the curves of solubility of the different salts is of great importance; but, unfortunately, they are only known for a small number. The deposition of one of the salts can be frequently determined without evaporating the liquid, by merely modifying the nature of the solvent. If solutions of acetate of potassa and of chloride of calcium be mixed, there is no apparent reaction, if the liquids are not highly concentrated. But, by adding a sufficient quantity of alcohol to the solution, chloride of potassium is deposited, and acetate of lime remains in the liquid. § 390. When acids and bases exist simultaneously in solution, it is generally impossible to decide in what manner they are com- bined, and to draw conclusions as to the order in which they will be successively deposited by crystallization; for the order is deter- mined solely by inferior solubility at the operating temperature, and it may be admitted that the less soluble salt is formed at the very moment of its crystallization. Bases, however, exist, in which a probable decision can be given as to the nature of the salts existing in a solution, as in a mixture of two groups of acids and bases, when, one of the bases forming colourless salts with the two acids, the other base forms coloured salts with them, but of different shades of colour. If solutions of the protosulphate of iron and acetate of soda be mixed together, the brown shade of the liquid proves that it contains acetate of iron and sulphate of soda immediately after mixture; because sulphate of iron forms a light green, and acetate of iron a brown solution. Again, a current of sulfhydric acid gas exerts no action on a solution of protosulphate of iron, while it decomposes proto- acetate of iron, producing a deposit of black sulphide of iron. Now, the same precipitate is formed when sulfhydric acid is passed through a liquid in which acetate of soda and protosulphate of iron have been dissolved at the same time. This latter character is, however, less decisive than the colour; for it might be said that the reciprocal decomposition of the two salts takes place only by virtue of the sulfhydric acid, and is determined by the insolubility of the sulphide of iron, which may be formed in the case of reci- procal decomposition, but would not form if no reaction took place in the mixture of the two salts. § 391. An insoluble salt may sometimes be decomposed by boil- ing it for a long time zvith a soluble salt. This occurs whenever the base of the original insoluble salt can form an insoluble salt with the acid of the reacting soluble salt. Thus, the insoluble salts of baryta, strontia, and lime, as the sulphates of baryta and strontia, the phosphates or arseniates of all three bases, are decom- posed when they are boiled with a solution of carbonate of potassa BEHAVIOUR OF SALTS. 421 or soda. Carbonates of baryta, strontia, and lime are formed, and the liquid contains the alkaline base combined with the acid of the original insoluble salt. But, to render the decomposition complete, a large excess of alkaline carbonate must be used. The same decomposition is much more readily effected, by operating by the dry way; and frequent use will hereafter be made of it to recognise the nature of an insoluble salt. For the acid of such a salt forms a soluble alkaline salt, the acid of which may be recog- nised by characters soon to be developed. The base remains in the state of an insoluble carbonate ; but by treating the carbonate with an acid which forms a soluble salt with the base, such as nitric acid, a solution of the base is obtained, in which the chemi- cal reactions characteristic of the base may be ascertained. DISTINCTIVE CHARACTERS FOR RECOGNISING THE ELECTRONEGA- TIVE ELEMENT OF BINARY COMPOUNDS FORMED BY THE METALS, AND THE NATURE OF THE ELECTRONEGATIVE ELEMENT, OR ACID, ENTERING INTO THE COMPOSITION OF A SALT. § 392. A binary compound, formed by a metal and a metalloid, or a salt formed by a metallic oxide, being given, how can the nature of the binary compound, or that of the salt, be ascertained ? The solution of this important question is generally divided into two parts: 1st. The determination of the electronegative element; that is, the metalloid of a binary compound, or the acid of a salt. 2dly. The determination of the electropositive element; that is, the metal of the binary compound, or the base of the salt. At present, we shall consider only the first part of the question, and treat the second part fully in detail under each particular metal. Determination of the Electronegative Element, that is, of the kind of Binary Compounds formed by the Metals with the Metalloids. § 393. Oxides.—The characters employed for deciding whether a binary metallic compound is an oxide are often reduced to the physical characters of these oxides, characters which will be indi- cated with precision when describing each metal. At other times we rely on their property of dissolving in strong acids, such as oil of vitriol, without disengaging any gas or acid vapour, and on our inability to detect in the solution any other acid than the one employed to effect solution. The majority of the metallic oxides are reduced by hydrogen when heated, the metal remaining free, and the vapour of water alone being disengaged. By taking the precaution to use dry hydrogen, the appearance of non-acid drops of water condensing 422 THE METALS. in the anterior and cold portion of the tube in which the substance is heated, is a sure indication that an oxide is operated on. Certain metallic oxides, however, are not reduced by hydrogen, such as the oxides of potassium, sodium, lithium, barium, strontium, calcium, magnesium, aluminum, and of all the earthy metals. But the oxides of potassium, sodium, lithium, barium, strontium, cal- cium, and magnesium, are more or less soluble in water, and exhibit a decided reaction on the tincture of litmus, a property they share only with the corresponding sulphides. Now, sulphides are easily distinguished from oxides, from the manner of their beha- viour to acids which disengage sulfhydric acid abundantly, easily recognisable by its odour. The oxides of aluminum and of all the other earthy metals are not decomposed by hydrogen, nor do they dissolve in water, nor, consequently, exert any action on the tincture of litmus. They are known, both by their insolubility in water, and, when treated with sulphuric acid, by their dissolving without disengaging acid vapours, and by the impossibility of ascertaining in the liquid the presence of any other acid than the sulphuric. § 394. Sutyhides.—Sulphur, like oxygen, frequently forms seve- ral compounds with the same metal, so that we may have mono- sulphides, bisulphides, trisulphides, etc. The monosulphides of potassium, sodium, and lithium are alone soluble in water ; all other monosulphides are insoluble, or, at least, very slightly soluble. The polysulphides of potassium, sodium, lithium, barium, strontium, and calcium, are equally soluble. A monosulphide, heated with dilute sulphuric, or with chloro- hydric acid, disengages sulfhydric acid gas, easily recognised by its odour, and no sulphur is deposited: RS+S03+H0=R0,S03+HS, or RS+HC1=RC1+HS. If it be a bisulphide, or, in general, a polysulphide, sulphuretted hydrogen is also disengaged, but, in addition, a deposit of sulphur is formed. RSa+S03+H0=R0,S03+HS+S or RSa+HCl=RCl+HS+S. Many of the metallic sulphides, are attacked with difficulty by aqueous chlorohydric acid, even at the boiling point, but are always decomposed by nitric acid, or aqua regia. The sulphur is then changed into sulphuric acid, the presence of which may be always recognised by the characteristic properties of the sulphates, to be hereafter explained. When sulphides are heated with a mixture of carbonate and nitrate of potassa, they produce alkaline sulphates soluble in water and easily recognised. BEHAVIOUR OF SALTS. 423 The metallic monosulphides act the part of bases to other sul- phides, forming sulphosalts, which we shall subsequently learn to recognise. § 395. Selenides.—The seleniurets, treated with chlorohydric acid, disengage selenohydric acid gas. Heated with nitric acid, or aqua regia, they produce selenious acid, the presence of which is recognised by sulphurous acid, which precipitates selenium in the form of a characteristic red powder. When heated in the dry way with a mixture of carbonate and nitrate of potassa, they give seleniate of potassa; but if the resulting alkaline salt be boiled with an excess of chlorohydric acid, the selenic is changed into selenious acid, from which selenium may be then precipitated by sulphurous acid. § 396. Phosphurets.—The phosphurets of the alkaline and alka- lino-earthy metals disengage phosphuretted hydrogen gas in con- tact with water, and the gas is instantly recognised by its odour. The phosphurets of the other metals, heated with potassium, yield their phosphorus to it, and it then disengages phosphuretted hydrogen when moistened with water. § 397. Arseniurets.—The metallic arseniurets possess metallic lustre. Treated with nitric acid or aqua regia, they are converted into arseniates, recognisable by characters we shall afterward ex- plain. Heated with nitrate of potassa, they produce a soluble alkaline arseniate. § 398. Chlorides.—The metallic chlorides are nearly all soluble in water, that of silver and protochloride of mercury being the only exceptions. A metallic chloride, treated with oil of vitriol, disengages chloro- hydric acid. Heated with a mixture of peroxide of manganese and sulphuric acid, chlorine is given off, which is easily recognised by its odour and other physical properties. The chlorides, dissolved in water, give with nitrate of silver a white precipitate, which collects into flakes by shaking the liquid. The precipitate is blackened by sunlight, assuming first a violet tinge. The rapidity of the change of colour is proportioned to the intensity of light, and rapidly ensues when exposed to the direct rays of the sun. The precipitate of chloride of silver is insoluble in acids, but readily dissolves in ammonia. § 399. Bromides.—A bromide, treated with oil of vitriol, disen- gages obierohydric acid; but vapours of bromine are constantly disengaged, at the same time imparting a brown colour to the gas. If the bromide be treated with a mixture of sulphuric acid and peroxide of manganese, bromine only is disengaged. A solution of a bromide gives, with nitrate of silver, a light yellowish-white precipitate of bromide of silver, which is insoluble in an excess of acid, and readily dissolves in ammonia. The precipitated bromide is coloured by light like the chloride, but is immediately tinged 424 brown, while the chloride assumes at first a violet hue. The bro- mides, in solution, are decomposed by chlorine, and bromine being set free, colours the liquid brown. § 400. Iodides.—The iodides, treated with oil of vitriol, instantly produce a considerable deposit of iodine; and if the mixture be heated, intense violet vapours are disengaged. The reaction is due to the decomposition of oil of vitriol by iodohydric acid, water and sulphurous acid being formed, and iodine set free. The iodides in solution are decomposed by chlorine, iodine being pre- cipitated, the smallest quantity of which in solution is instantly detected by its imparting to starch an intensely blue colour. A certain quantity of the solution is mixed with a solution of starch, effected in boiling water and cooled, or with ordinary starch-paste, and then a few drops of chlorine-water are added to decompose the iodide and liberate iodine. The mixture immedi- ately assumes a decided blue colour. It is important not to add an excess of chlorine, which would destroy the blue colour by de- composing water, and generating chlorohydric and iodic acids. § 401. Fluorides.—A fluoride, treated with oil of vitriol, disen- gages vapours of fluohydric acid, which may be immediately recog- nised by its property of attacking glass. If silicic acid or pounded glass be added, and the mixture heated, gaseous fluoride of sili- cium is disengaged, which is decomposed by contact with water, affording a deposit of gelatinous silica. Solutions of fluorides are not precipitated by nitrate of silver. § 402. Cyanides.—The cyanides, treated with sulphuric or chlorohydric acid, disengage cyanohydric (prussic) acid, easily recognised by its odour. The most feeble acids, such as the car- bonic, give off the same odour with soluble cyanides, and even the alkaline cyanides manifest it in a damp atmosphere. The cyanides, with salts of protoxide of iron, give a white pre- cipitate which rapidly turns blue in the air. THE METALS. Determination of the Oxacid which enters into the constitution of an Oxysalt. § 403. Nitrates.—Nearly all nitrates are soluble in water, a few sub-nitrates alone being insoluble. Heat decomposes them, afford- ing products which are rich in oxygen and powerfully assist com- bustion. In consequence of this property, the nitrates deflagrate on hot coals, and often detonate when heated with powdered char- coal. The alkaline nitrates, subjected to a gradually increasing temperature, disengage at first pure oxygen, and are changed into nitrites. Heated still further, they are entirely decomposed, evolving nitrogen and oxygen. The other nitrates disengage oxygen and deutoxide of nitrogen, or oxygen and hyponitric acid. When those formed by soluble bases are decomposed by heat, they leave a strongly alkaline residue. BEHAVIOUR OF SALTS. 425 Heated with sulphuric acid, they disengage vapours of nitric acid; and if a small quantity of metallic copper be added to the mixture, deutoxide of nitrogen is immediately disengaged, recog- nised by the reddish vapours it forms in the air. The presence of a very small quantity of nitric acid in a liquid may be ascertained by pouring a small quantity of the liquid into a solution of the protosulphate of iron, acidified by sulphuric acid, and then plunging into it a strip of iron. If the liquid contains nitric acid, it turns red or brown after some time. Influenced by the sulphuric acid, the metallic iron decomposes the nitric acid, and deutoxide of nitrogen is disengaged, which dissolves in the protosulphate of iron and colours the liquid (§ 114). § 404. Nitrites.—The nitrites are decomposed by heat, like the nitrates, fusing on coals, and deflagrate when heated with powdered charcoal. With sulphuric acid, they immediately disengage reddish vapour, which suffices to distinguish them from the nitrates. § 405. Chlorates.—The chlorates are all decomposed by heat. Those of the alkalies and alkaline earths disengage oxygen, yield- ing a residue of chloride which is neutral to coloured tests, while the corresponding nitrates, under the same circumstances, leave a strongly alkaline residue. The chlorates of the other metallic oxides disengage by heat a mixture of oxygen and chlorine, leaving an oxide or oxychloride. The chlorates are energetic supporters of combustion, deflagrate on heated coals, and produce violent detonations when heated with very combustible bodies, such as charcoal, sulphur, and phos- phorus. Treated with sulphuric or chlorohydric acid, they disengage a yellow gas, chlorous acid, recognisable by its colour, peculiar odour, and property of readily detonating on a slight elevation of temperature. The chlorates do not precipitate salts of silver, because chlorate of silver is soluble in water; but the residue left after calcining the alkaline and alkalino-earthy chlorates being a chloride, gives, with a solution of nitrate of silver, a precipitate of chloride of silver, which may be recognised by its characteristic properties (§ 398). § 406. Perchlorates.—The perchlorates behave like the chlorates when subjected to the action of heat, or when heated with com- bustibles, but are easily distinguished from them, because they do not disengage chlorous acid by the action of oil of vitriol, and, consequently, are not coloured, for perchloric acid is merely isolated, without decomposition. Perchlorate of potassa is but slightly soluble in water, and hence the salts of potassa give, with the perchlorates, a granular crys- talline precipitate when the liquids are not too dilute. § 407. Hypochlorites.—The hypochlorites disengage the peculiar 426 THE METALS. and characteristic odour of hypochlorous acid, which they give off copiously when treated with an acid. Their solutions bleach vege- table colours. Only the hypochlorites of potassa, soda, and lime, have been studied. They behave like energetic oxidizing agents, immediately changing sulphurous into sulphuric acid, and peroxid- izing metallic protoxides. § 408. Bromates.—The bromates are decomposed by heat like the chlorates. Those of the alkalies and alkaline earths leave a residue of bromide, which may be recognised by the characters designated in § 399. When heated with sulphuric acid, bromic acid is isolated and decomposed into oxygen and bromine, the latter tinging the gas brown. §409. Iodates.—The iodates are decomposed by heat. The alkaline salts alone leave a residue of iodide. The alkalino-earthy iodates, and those of all other metallic oxides, leave an oxide or an oxiodide, violet vapours of iodine mixed with oxygen being copiously given off. Sulphuric acid precipitates iodic acid from the iodates in a concentrated solution; and if some reducing body, as sulphurous acid, be added to the liquid, iodic acid is decomposed, and iodine precipitated. § 410. Periodates.—The periodates behave, when heated, like the iodates, but are distinguished from the latter by the slight solubility of periodate of soda, even in the presence of an excess of alkali, and the slight solubility of periodate of silver. § 411. Sulphates.—Nearly all the sulphates are soluble in water; those of baryta, strontia, and lead are nearly insoluble; that of lime is slightly soluble. The sulphates of the alkalies, alkaline earths, and of lead are indecomposable by heat alone: the other sulphates are decomposed, and generally yield a gaseous mixture of sulphurous acid and oxygen. Some sulphates, however, are decomposed at so low a temperature that the sulphurous acid and the oxygen remain united, and are disengaged in the state of sul- phuric acid (§ 138). All the sulphates are decomposed by carbon, assisted by heat; the products of the composition vary with the nature of the base and the temperature. The alkaline sulphates, heated rapidly with carbon, at a high temperature leave a residue of monosul- phide ; at a lower temperature they afford a mixture of polysul- phide and carbonate. Those of the alkaline earths, with the exception of magnesia, give similar products. Those of the other metallic oxides, heated with carbon, yield a residue either of sul- phide, or oxide, or even of metal, if the temperature be sufficiently elevated. But the experiment can always be performed with any sulphate, so as to obtain a sulphide, if a certain quantity of car- bonate of potassa be added to the mixture. The alkaline sulphide remaining after calcination is easily recognised, as it gives off sulphuretted hydrogen with acids. The same character evidently BEHAVIOUR OF SALTS. 427 belongs to the salts formed by all the oxacids of sulphur, as well as to the sulphates, but we shall soon learn how to distinguish them from each other. As sulphuric acid does not act on the sulphates, this fact imme- diately distinguishes the sulphates from all salts which, under similar treatment, disengage acid vapours. The sulphates soluble in water give, with the soluble salts of baryta, a white precipitate which is insoluble in an excess of acid; a property entirely characteristic of the sulphates. § 412. Sulphites.—The alkaline and alkalino-earthy sulphites, heated in a close vessel, are changed into sulphates* and sulphides: 4(K0,S0a)=3(K0,S03)+KS. The other metallic sulphites disengage sulphurous acid, and the oxide remains as a residue. Heated with carbon, they give pro- ducts similar to those of the sulphates. Sulphuric acid, poured upon a sulphite, disengages sulphurous acid gas, easily recognised by its odour, and no deposit of sulphur takes place. Concentrated boiling nitric acid changes the sulphites into sul- phates. Chlorine produces the same change on the sulphites in solution. The soluble sulphites also absorb oxygen from the air, and are changed into sulphates. § 413. Hyposulphates.—The hyposulphates are all soluble in water. Those of the alkalies, alkaline earths, and of oxide of lead disengage sulphurous acid when subjected to the action of heat, leaving sulphates. Those of the other metallic oxides are more completely decomposed, and an oxide generally remains. The hyposulphates, treated with cold sulphuric acid, manifest no apparent decomposition; but, when heated with the acid, they give off sulphurous acid. They do not precipitate the salts of baryta, for hyposulphate of baryta is soluble in water. They are readily converted into sulphates by nitric acid, or by an aqueous solution of chlorine, and are then precipitated by salts of baryta. § 414. Hyposulphites.—Nearly all the hyposulphites are soluble, those of silver and lead alone being nearly insoluble. Heat de- composes the alkaline salts into sulphates or sulphides. Chloro- hydric and sulphuric acids, poured into a solution of a hyposulphite, evolve sulphurous acid gas, and cause a deposit of sulphur; but the reaction does not always take place immediately, and often does not ensue for some time, unless the liquid be slightly heated. Highly concentrated nitric acid, chlorine, and solutions of the hypochlorites, cause all the sulphur of the hyposulphites to pass into the state of sulphuric acid. The hyposulphites give, with the salts of silver, a white precipi- 428 THE METALS. tate, which, however, soon blackens from its conversion into a sulphide: K0,SA+Ag0,N0s=K0,S03+AgS+N0s. The alkaline hyposulphites readily dissolve chloride, bromide, and iodide of silver in large quantities. The majority of characters enumerated as distinguishing the hyposulphites, also belong to the monosulphuretted hyposulphates KO,S3Os, to the bisulphuretted hyposulphates K0,S40s, and to the trisulphuretted hyposulphates KO,SsOs. These last salts have, hitherto, been too little studied to allow us to assign to them any distinctive characteristics, and we are obliged to resort to chemical analysis. § 415. Recapitulation.—All the salts formed by the oxacids of sulphur give sulphides when heated with a mixture of alkaline, carbonate, and charcoal, so that the product of calcination disen- gages sulphuretted hydrogen with chlorohydric acid. This cha- racter distinguishes the salts formed by the oxacids of sulphur from all others. They might, indeed, be confounded with the sulphides and the sulphosalts; but these bodies immediately dis- engage sulphuretted hydrogen with the acids. The salts formed by the oxacids of sulphur are easily distin- guished from each other by the following characters, if they are treated with sulphuric acid : No reaction ensues with the sulphates; With the hyposulphates, there is no apparent reaction when cold, but, assisted by heat, sulphurous acid is evolved; With the sulphites, sulphurous acid is disengaged, without any deposit of sulphur; With the hyposulphites, and with the mono, bi, and trisulphu- retted hyposulphates, sulphurous acid is disengaged, and a more or less copious deposit of sulphur formed. This reaction frequently does not follow unless the temperature be elevated. §416. Phosphates.—The alkaline phosphates alone are soluble in water: all the others are insoluble in it, but readily dissolve in an acid liquid. The soluble phosphates afford a precipitate with salts of baryta; but it is dissolved if the liquid be acidified with nitric or chlorohydric acid. The phosphates evince no apparent reaction with oil of vitriol, and are thus instantly distinguished from all salts, which disen- gage acid vapours under the same circumstances. All the phosphates, heated to a high temperature with a mixture of carbon and boracic or silicic acid, give off free phosphorus. A dry phosphate, heated with potassium, gives off phosphorus, which, by contact with water, disengages phosphuretted hydrogen. These two reactions are equally manifest with the salts formed by the other oxacids of phosphorus. BEHAVIOUR OP SALTS. 429 An insoluble phosphate may be readily converted into a soluble alkaline phosphate, by simply boiling it with a solution of an alkaline carbonate. The presence of phosphoric acid may be, subsequently, recognised in the liquid, by supersaturating it with chlorohydric acid, and ascertaining that it is not precipitated by the salts of baryta. But, if the acid be neutralized by ammonia, a precipitate of phosphate of baryta is immediately formed. The neutral liquid also affords a white precipitate with salts of lead; and phosphate of lead is easily known, because it is fused by the blowpipe into a globule which, on becoming solid, assumes crys- talline facets. § 417. Phosphites.—The alkaline phosphites alone are soluble. All phosphites are decomposed by heat, giving a residue of phos- phate, and disengaging a mixture of hydrogen and phosphuretted hydrogen. Nitric acid and chlorine transform them into phos- phates. The phosphites reduce a certain number of metallic oxides, —among others, those of silver and mercury, and the reaction is more rapid if the liquid be acidified. The red oxide of mercury, heated with the solution of a phosphite, to which a small quantity of chlorohydric acid has been added, is converted into a black powder of metallic mercury. § 418. Hypophosphites.—The reactions of the hypophosphites closely resemble those of the phosphites. They are decomposed by heat, affording phosphates, and evolving phosphuretted hydro- gen. Nitric acid and chlorine transform them into phosphates. They are distinguished from the phosphites, because they never precipitate the salts of baryta, while the phosphites do precipitate them when perfectly neutral. §419. Arseniates.—The alkaline arseniates alone are soluble; those of all the other metallic oxides are insoluble, but they readily dissolve in an excess of acid. Any arseniate, heated with boracic acid and charcoal in a small tube, closed at one end, gives a sublimate of arsenic, which forms a metallic ring in the upper part of the tube. The solutions of the arseniates, treated in Marsh’s apparatus (§ 236), afford copious arsenical spots. With the nitrate of silver, they give a brick-red precipitate, which dissolves readily in an excess of acid; so that the precipitate is only formed when the liquids are perfectly neutral. The soluble arseniates give a yellow precipitate with sulphuretted hydrogen, but a long time is frequently required for its appear- ance. § 420. Arsenites.—The arsenites, heated with charcoal and boracic acid, give a sublimate of arsenic. In Marsh’s apparatus they produce arsenical spots. If an acid be poured into the concentrated solution of an alkaline 430 THE METALS. arsenite, a crystalline precipitate of arsenious acid is formed. The arsenites in solution precipitate the salts of silver yellow, and those of copper, green; but the liquids must be perfectly neutral, for the insoluble arsenites are readily dissolved in an excess of acid. Sulphuretted hydrogen affords, with the arsenites in solution, a copious yellow precipitate, insoluble in an excess of acid, but which readily dissolves in ammonia. This precipitate is formed immedi- ately, while, with the arseniates, some lapse of time is necessary. The arsenites, heated with nitric acid, are converted into arseni- ates, with the evolution of reddish vapours. The arseniates have no similar properties, not being altered by oxidizing substances. § 421. Carbonates.—The alkaline carbonates are the only soluble carbonates. They are also the only ones which cannot be decom- posed by heat. All the other carbonates part with all their car- bonic acid at a higher or lower temperature. All the carbonates, without exception, are decomposed when heated to a very high temperature, with charcoal, carbonic oxide being disengaged. When vapour of phosphorus is passed over an alkaline carbonate heated to redness, the carbonic acid is completely decomposed, and carbon separated, colouring the substance black. The carbonates, treated with an acid, produce a lively efferves- cence, owing to the evolution of carbonic acid, and this reaction characterizes them; for carbonic acid is easily recognised by being inodorous and tasteless, and precipitating limewater. This re- action alone suffices to distinguish the carbonates from all other salts. § 422. Borates.—The alkaline borates alone are soluble; all the others are insoluble. At a high temperature, they fuse and form colourless glass, when the metallic oxides combined with the boracic acid are themselves colourless; otherwise they form coloured glass. Charcoal acts with difficulty on the borates; only a few of which are decomposed by it at a very high temperature, and pro- duce metallic borides. Sulphuric, nitric, and chlorohydric acids decompose the borates in the wet way, liberating boracic acid. If the solution of borate be concentrated, the boracic acid is precipitated in the form of small crystalline scales, in which the characteristic properties of the acid are easily detected. Boracic acid, on the contrary, expels these acids in the dry way. If a mixture of any borate and fluor-spar be heated with oil of vitriol, fluoride of boron is disengaged, recognised by the dense white fumes it gives off in the air, and its mode of decomposition by contact with water (§ 241). § 423. Silicates.—The majority of the silicates are insoluble, the alkaline, with a great excess of base, being alone soluble in water. The silicates, decomposable by sulphuric and chlorohydric BEHAVIOUR OP SALTS. 431 acids, are easily recognised; for when heated with the acid, silicic acid separates in the state of a colourless, transparent jelly, which aggregates into an insoluble white powder, and when collected on a filter, its characteristic properties are easily shown. The sili- cates which are not decomposed by the acids, may be readily con- verted into the former by fusing them in a platinum crucible with three or four times their weight of carbonate of soda. A more basic silicate is thus obtained, containing a large quantity of alkali, and it is easily and entirely decomposed by the acids, leaving a residue of gelatinous silica. The silicates generally fuse when subjected to heat; but some, as the silicates of alumina and lime, require the very highest tem- peratures. Charcoal reduces some of the silicates at a high tem- perature, only a portion of the metal separating, and the remaining silicate containing a large excess of acid. Those partially decom- posed by charcoal are such as contain easily reducible metallic oxides. The silicates, heated in a vessel of lead or platina with fluor- spar and oil of vitriol, disengage gaseous fluoride of silicium, which fumes in the air, and is decomposed by contact with water, preci- pitating gelatinous silica. § 424. Sulphosalts.—The sulphosalts, treated with powerful hut not oxidizing acids, as dilute sulphuric or chlorohydric, disengage sulphuretted hydrogen, and the sulphacid separates. The majority of sulphacids being insoluble in water, the properties characterizing them may be recognised in their precipitates. Thus, with the sulphocarbonate of the monosulphide of potassium, sulfhydric acid is disengaged, and liquid sulphide of carbon is pre- cipitated : KS,CS2+HC1=KC1+HS+CS3. With the sulpharseniate of the monosulphide of potassium, sulf- hydric acid is disengaged, and sulphide of arsenic is precipitated in the form of a yellow powder : KS,AsS5+HC1=KC1+HS+AsS5. With the sulfhydrate of the monosulphide of potassium, there is an analogous reaction, but sulfhydric acid only is disengaged, one-half of which proceeds from the monosulphide of potassium, and the other half from the sulphacid which separates. Since the reaction does not distinguish this sulphosalt from monosulphide of potassium, the following process is adopted:—The monosulphides of the alkaline and alkalino-earthy metals are the only ones which act the part of bases with sulfhydric acid ; if, therefore, a metallic salt, such as the sulphate of copper, be poured into a solution of sulfhydrate of monosulphide of potassium, a double decomposition ensues, sulphate of potassa and monosulphide of copper being 432 THE METALS. formed. But, as the latter sulphide does not play the part of a base with the sulphacid HS, this sulphide becomes free, and, con- sequently, sulfhydric acid is disengaged : KS,HS+Cu0,S08=K0,S0s+CuS+HS. If, on the contrary, a solution of sulphate of copper he poured into a solution of a monosulphide, a precipitate of metallic sulphide is formed, hut no sulphuretted hydrogen is disengaged: KS+Cu0,S03=K0,S03+CuS. In order that this last proposition may be true, the metallic solution should not contain an excess of acid, which would decom- pose a portion of the monosulphide and disengage sulphuretted hydrogen. The monosulphides and sulfhydrates of sulphides are, moreover, distinguished from the polysulphides, inasmuch as they do not afford, like the latter, a deposit of sulphur when decomposed by the acids. 433 OF THE METALS INDIVIDUALLY. § 425. In the following investigation of the most important me- tals, we shall preserve the classification indicated in § 276, viz. metals which are too oxidizable to be used in the metallic state, and those which remain unchanged in the air for so long a time that their alterability is no obstacle to their use. The first class will be thrown into three subdivisions: 1. The alkaline metals, comprising Potassium, Sodium, Lithium. The term alkaline metals has been given to them, because their oxides have, for a long time, borne the name of alkalies. 2. The alkalino-earthy metals, whose oxides partake, at once, of the properties of the alkalies and of the earths. They are, Barium, Strontium, Calcium, Magnesium, Glucinum. 3. The earthy metals, so called, because their oxides have for a long time borne the name of earths. They are, Aluminum, Zirconium, Thorium, Yttrium, Terbium, Cerium, Lanthanum, Didymium, Erbium. 434 I. ALKALINE METALS. POTASSIUM. • § 426. Potassium is a metal pretty extensively spread over the earth, but it exists only in combination with other bodies. A great majority of the minerals which compose the crystalline rocks, as the feldspars, micas, etc., contain silicate of potassa. The debris of these rocks, altered by water, constitute the sedimentary have lost a large portion of their potassa, but still retain a sufficient quantity to be found by chemical analysis. The salts of potassa are indispensable to the growth of plants, which gradu- ally abstract them from the soil and manure; and their ashes furnish the greater portion of the salts of potassa used in the arts. The consistence of potassium varies with the temperature. Below 32° it is slightly friable, and its fracture presents indi- cations of crystallization. At 59° it is soft, and may be kneaded, and easily cut with a knife. When recently divided, it affects the colour and lustre of silver, but the lustre is evanescent, for as the metal rapidly combines wTith the oxygen of the air, its surface becomes tarnished. At 131° it becomes perfectly liquid, and then resembles mercury. Lastly, it distils at a red-heat, as a beautiful emerald-green vapour. Its density has been found to be 0.865 at about 59°, and is consequently lighter than water. It oxidizes rapidly in the air, even at ordinary temperatures, its surface becoming covered with the hydrated oxide of potassium or potassa ; but some time is necessary, for the change to penetrate the centre of a globule of any considerable size. If it be heated in the air, it takes fire and burns with a violet flame. Potassium decomposes water at ordinary temperatures, disen- gaging hydrogen. If a fragment of it be thrown on water, it is observed to glide over its surface in the form of a brilliant little sphere, the size of which rapidly diminishes, and to be accom- panied by a violet-coloured flame. When the combustion ceases, the little globule bursts, and its fragments are thrown in every direction. In making this experiment, care must be taken to use a deep bell-glass (fig. 316), lest the eyes or person of the operator be injured by the explosion. After the experiment, the water in the bell-glass will be found to be alkaline, and to blue the reddened tincture of litmus. The various circumstances of this phenomenon are Equivalent =490.0. Fig. 316. POTASSIUM. 435 easily explained. The fragment of potassium swims on the water, because of its greater levity. Water being decomposed, the heat developed fuses the metal, which takes the form of a glittering globule; the hydrogen gas evolved raises up the metal, prevent- ing it from remaining constantly in contact with the water, and drives it over the surface. The temperature of the globule of potassium being sufficiently high to inflame the hydrogen, it burns, as fast as it is formed, with a violet flame, the colour of which is due to the admixture of a small quantity of vapour of potassium arising from the heated metal. Whenever the globule falls back on the surface of the liquid, the small quantity of oxide of potassium formed is dissolved in the water. Lastly, when the combustion ceases, there remains a small globule of very hot potassa, which falls on the liquid, where it bursts, in consequence of sudden cool- ing ; and as a large quantity of steam is instantly developed at this spot, its expansive force throws small fragments of potassa in every direction. The great liability of potassium to alteration, requiring peculiar care in its preservation, it is generally kept in ground-stoppered bottles, nearly filled with naphtha, which is a compound of carbon and hydrogen unalterable by the metal. Potassium, being one of the substances possessing the greatest affinity for oxygen, is constantly used to abstract the oxygen from oxidized bodies. Boron was prepared (§ 238) by decomposing boracic acid by potassium. The protoxide and deutoxide of ni- trogen (§ 111 and 115) were analyzed by decomposing them by potassium. Some bodies can, however, remove oxygen from the oxide of potassium at a high temperature, and set potassium free, such as iron, at a white-heat. At a dull red-heat, potassium de- composes carbonic acid; but, at a w7hite-heat, carbon deprives the potassa of its oxygen. Advantage is taken of this property in the preparation of potassium. § 427. Potassium was at first isolated by decomposing the hydrate of potassa by a powerful voltaic pile. To effect it, a cer- tain quantity of mercury was placed in a platinum crucible, and, above it, a concentrated solution of potassa, containing fragments of solid potassa. The negative pole of the pile being brought in contact with the platinum crucible, and the positive pole, termi- nating in a strong platinum wire, being plunged into the solution of potassa, the decomposition of the hydrated oxide of potassium commenced immediately. Water and oxide of potassium being decomposed at the same time, hydrogen and potassium were found at the negative, and oxygen at the positive pole. The hydrogen and oxygen were evolved in the gaseous state, the potassium was dissolved in the mercury, which assumed, after some time, a pasty consistence. The pasty metal being quickly introduced into a small glass retort, heated by an alcohol lamp, the mercury was 436 ALKALINE METALS. driven off, and a globule of potassium remained in the retort. Very small quantities of potassium were obtained in this way, sufficient, however, to verify its principal properties.* § 428. Soon after, larger quantities of potassium were obtained by decomposing potassa in vapour by iron at a white-heat. Fig. 317 represents the apparatus employed for the operation.f A gun-barrel abc is bent at b and i, so as to give it the shape represented in fig. 317; and as this portion bi is to be intensely ignited, its surface would soon oxidize and the barrel be rendered useless, if its surface were not protected by an unalterable lute which covered it completely. This lute is composed of 4 or 5 parts of sand and 1 part of potter’s clay, and being spread to a thickness of 1 or 2 centim. (J - f in.), is first dried slowly in the air, and then before the fire. The cracks made in drying are filled with clay. The gun-barrel being filled with bright iron turnings, or small bundles of clean iron ivire, and the part ah with pieces of potassa, it is placed in a reverberatory furnace (fig. 318). The end a of the iron tube is closed with a cork, fur- Fig. 317. Fig. 318. * Davy, in 1807, first isolated potassium in this way. f The credit of this process is due to Messrs. Gay-Lussac and Thenard. POTASSIUM. 437 nished with a tube, entering the test-glass E, filled with mercury. A chaffer GG' of wire or sheet-iron is suspended below the part ab. The extremity c is passed into a copper receiver deg, made of two pieces de and fg (fig. 317 and 318), fitted together by grind- ing, and the naphtha is introduced into the lower part ge, in order to collect the potassium. A tube t allows the escape of the gas formed during the experiment. The apparatus being arranged, the furnace is filled with char- coal, and as the natural draught would not afford sufficient heat, the combustion is assisted by a large bellows, the nozzle of which enters the door of the furnace, the surrounding apertures being closed with pieces of brick and clay. When the tube be has reached a strong white-heat, hot coals are introduced into the chaffer GG', so as to slowly fuse the fragments of potassa contained in the tube ab. The fused potassa flowing into the heated tube be, where it meets the iron intensely ignited, the decomposition of water and oxide of potassium takes place at the same time; the iron is converted into oxide of iron; the potassium in vapour is carried forward by the current of hydrogen gas, and condenses in the receiver ge. As it sometimes happens that the end c becomes closed during the experiment, so that the gases cannot readily escape, they would issue through the joints of the various parts of the apparatus, and render it useless. The disengagement tube aE remedies the in- convenience, and immediately indicates when the aperture c is obstructed, by gases escaping through the mercury in the test- glass E. § 429. Potassium is now* prepared by decomposing carbonate of potassa by charcoal at intense ignition, whereby much larger quantities of potassium can be procured than by the older pro- cesses. It is essential that the carbonate of potassa be intimately mixed with the charcoal. Only an imperfect mixture is obtained by mechanically mixing the carbonate with charcoal; and as this carbonate melts long before its decomposition by the charcoal can take place, the latter, being lighter, floats on the surface, and the mixture is destroyed. On the other hand, a very intimate mixture of potassa and charcoal can be obtained by decomposing certain salts of potassa with organic acids by heat. The bitartrate of potassa is well adapted to the purpose, as it leaves a great deal of charcoal, and is not expensive, if procured in the state of impure bitartrate or crude tartar or argol. The crude tartar being placed in a large clay crucible, closed by a cover, and luted to prevent the admission of air, is heated to redness in a furnace, until no more gas is disengaged. When the crucible is cooled, the black substance is pulverized in a mortar, * This process was contrived by Brunner. 438 ALKALINE METALS. mixed with coarsely broken charcoal, and introduced into a wrought-iron bottle. The iron flasks ordinarily used in commerce for mercury are well adapted to the object. They have only one opening at o (fig. 319), Avhich is closed with an iron screw for the transportation of mercury; but, for our purpose, a thread is cut on an iron tube, so as to fit the aperture o. The joint is closed as tightly as practicable by clay. In order to prevent the altera- tion of the bottle during the operation, its surface is covered with an argillaceous luting, carefully applied. The bottle, three-fourths filled with the mixture, is arranged, as represented in fig. 319, in a furnace in which intense ignition can be obtained. Fig. 319. This furnace is built of a rectangular form, with its walls of fire-brick, for ordinary bricks would fuse at the high temperature necessarily required. It is generally open at the top, to facilitate the arrangement of the iron bottle, or of crucibles when the fur- nace is used for other purposes, as well as to supply the fuel. The opening is closed with a cover M, made of bricks, bound in an iron frame. The furnace communicates with a high chimney U, by the fine 0, and a damper R serves to regulate the draught. The ash-pit C has an aperture in front, by which air enters the fur- nace. One of the side-walls of the furnace has a rectangular opening, which is closed with fire-brick when the furnace is used for heating crucibles ; but when employed for the preparation of potassium, it is closed by a cast-iron door m, having a hole through which an iron tube uo passes. POTASSIUM. 439 The bottle Y is placed in the furnace on two stout iron bars, or, better still, on two fire-bricks projecting from the sides, and the iron tube uo enters a copper receiver A, of peculiar construction. It is composed of two parts B and C, which fit into each other, represented in section in fig. 320, where they are separated. The lower part is a cylindrical copper vessel, with an oval base. The upper part, which serves as a cover, enters the former as far as the height mn, and is di- vided into two compartments by a vertical partition cd, which descends to within a short distance of the bot- tom of the vessel C, when the two parts are together. Two tubulures a, b are placed exactly opposite to each other, and the vertical wall cd has an opening in the direction ah. A third tubulure/is placed on the anterior face of the cover, as seen in fig. 319. Naphtha is poured into the vessel C to a depth of 5 or 6 centi- metres (2-2J in.), the two pieces fitted together, and the tube uo adjusted in the tubulure a, by closing the interstices tightly with an argillaceous lute. Into the tubulure / a glass tube g is fitted, which gives exit to the gas; and, lastly, the tubulure b is closed with a cork. The receiver rests on a support S (fig. 319), covered with a sheet-iron plate, having a drain at T. The apparatus being arranged, live coals are first introduced into the furnace, then common charcoal, and when the fire is well kindled, it is fed with a mixture of equal parts of charcoal and coke. At each time of charging with fuel, a poker should be passed into the furnace to prevent cavities from forming under the retort. The reaction of charcoal on the carbonate of potassa soon com- mencing, carbonic oxide gas is copiously disengaged from the tube fg; and the potassium set free volatilizes, condenses in the re- ceiver, and sinks under the naphtha. As the receiver would soon become heated by radiation from the furnace and the passage of heated gases, it is kept cool by allowing a constant current of cold water to flow over the top. The ledge mn prevents the water from entering into the lower compartment, and it finally runs off by the drain T. It frequently happens in the operation that the iron tube uo is obstructed by substances carried over mechanically, or by those arising from a peculiar reaction which will soon be explained. This is known by the cessation of the current of gas in the tube fg, and is remedied by introducing through the tubu- lure b, an iron wire (fig. 321) fastened to a wooden handle, and turning it around until it has pierced the deposit formed in the tube uo, and made a free passage for the gas. Fig. 320. Fig. 321. 440 ALKALINE METALS. The operation is terminated, when no more gas passes through the tube fg, although the tube uo be not obstructed. The receiver being removed, the potassium is found in the form of irregular globules, mixed with various accidental substances, which are se- parated by filtering through cloth. The impure potassium, being placed in a cloth tied like a bag, is plunged into a cup filled with naphtha at 120° to 140°. The bag being compressed with pincers, the potassium filters in the form of metallic globules through the cloth, and falls to the bottom of the capsule, where it collects into larger globules. The foreign matter remains in the bag. It has been stated that potassium decomposes carbonic oxide at a dull red-heat; and hence, it is difficult to prevent the occurrence of an inverse reaction in our apparatus, and the loss of a portion of the isolated potassium during the first reaction. The carbonic oxide gas and vapour of potassium, when leaving the retort, enter into the iron tube uo, where they meet a much lower temperature, and the inverse reaction ensues. A portion of the potassium de- composes carbonic oxide, forming peculiar products, to which the names of croeonate and rhodizonate of potassa have been assigned. They are deposited in the tube uo with free carbon, and tend to obstruct it. A portion of them is carried as far as the receiver, in the form of black flakes, which may be used to extract the two salts of potassa just named. To diminish, as much as possible, the loss of potassium occasioned by this inverse reaction, it is neces- sary to shorten the tube uo as far as practicable. § 430. To obtain absolutely pure potassium, it is redistilled in a wrought iron vessel A (fig. 322), to which is screwed the curved iron tube abc. Po- tassium, and a portion of naphtha are intro- duced into the vessel, which is then heated in a furnace, and the extremity of the tube abc plunged into a bottle containing naphtha. From time to time, a gentle blow should be given to the tube abc, in order to cause the melted potassium to flow into the receiver. Fig. 322. COMBINATIONS OF POTASSIUM WITH OXYGEN. § 431. Potassium forms two compounds with oxygen; a pro- toxide to which has been assigned the formula KO, and a peroxide which contains three times as much oxygen, and, consequently, takes the formula K03. When a globule of potassium is heated in a small silver tray put in a glass tube, and traversed by a current of dry oxygen, the metal takes fire, and is changed into a yellow fusible substance, which is peroxide of potassium. It readily dissolves in water, but is decomposed, two-thirds of its oxygen being set free, and pro- POTASSIUM. 441 toxide of potassium dissolved. If the solution be evaporated to dryness, the hydrated protoxide of potassium is obtained, "which fuses at a dull red-heat, but cannot be deprived of its water of combination. The preparation of the protoxide is accompanied by great diffi- culties. In order to obtain it, a known weight of potassium is converted into peroxide by heating it in a silver tray, in a current of oxygen; twice the weight of potassium in the peroxide is put with the latter in the same tray, which is heated in the same tube, in a current of nitrogen gas: K034-2K=3K0. It may also be obtained by heating a known weight of the hy- drated protoxide of potassium KO-f-IIO, or potassa, with a weight of potassium equal to that which exists in the potassa, in which case the hydrogen of the water is set free, and 2 equivalents of protoxide formed: K0,H0+K=2K0+H. The protoxi e cannot be obtained by decomposing nitrate of potassa by heat, a process by which many anhydrous protoxides are prepared, as those of barium, strontium, calcium, etc. Nitrate of potassa, heated in a glass or porcelain retort, decomposes at a dull red-heat into oxygen, which is set free, and a nitrite, which remains in the retort: K0,N05=K0,N03+20. If the heat he raised still higher, the nitrite itself is decomposed, and oxygen and nitrogen evolved; but the protoxide of potassium seizes on a portion of the oxygen, and passes partly into the state of peroxide. A complete decomposition of the nitrite cannot be effected in glass or earthenware vessels; for the silicates consti- tuting the body of these vessels are attacked by the oxides of potassium, and the vessels soon destroyed. There is no better success in a vessel of platinum ; for it is soon corroded by the oxides of potassium, especially in the presence of oxygen. Silver resists the action of these oxides much better, but it is too fusible to allow the complete decomposition of the nitrite. It will soon be shown that the hydrate of potassa may be readily obtained in large quantities, and that it is one of the most useful substances in the laboratory. Of the two compounds which potassium forms with oxygen, only the protoxide plays the part of a base, and it is the most powerful base of our laboratories. No compound formed by the peroxide being yet known, it possesses but little interest. It is immediately decomposed by contact with water and the acids, disengaging oxygen, and forming a salt of the protoxide of potassium. 442 ALKALINE METALS. SALTS FORMED BY THE PROTOXIDE OF POTASSIUM, OR POTASSA. Combinations of the Protoxide of Potassium with Water. § 432. The protoxide of potassium, or potassa, forms, with water, two definite compounds or hydrates, a monohydrate KO+HO or KO,HO, and a pentahydrate KO + 5IIO. When potassium decomposes water, the hydrate of potassa KO,HO is formed, and remains in solution in the water. The same hydrate is produced by decomposing a salt of potassa by a base which forms an insoluble compound with the acid of the salt. The last process is the one always used in the labora- tory for the preparation of the hydrates of potassa, which are very important reagents- For this purpose, carbonate of potassa is decomposed by lime, an insoluble carbonate of lime being formed, and the potassa remaining in the liquid in the state of hydrate. One part of carbonate of potassa is dissolved in 10 of water. If the carbonate does not dissolve without residue, on account of its impurity, the liquid is allowed to stand, and then decanted into a clean cast-iron kettle, in which it is boiled. Slacked lime diffused in water is then added by small quantities to the boiling liquid. A small quantity of the liquid being taken up with a pipette and poured into a glass, is allowed to repose for a few moments, until the suspended matter is deposited, when a portion of the clear liquid is transferred to a test-glass, and an excess of chlorohydric acid added. If all the carbonate of potassa has been converted into hydrate, no effervescence will ensue. If effervescence takes place, the ebullition is continued for some time, small quantities of lime being added, if necessary, until no effervescence occur in another experiment performed in the same way. The kettle being removed from the fire, the liquid is allowed to clarify by repose, keeping the kettle covered, to prevent the potassa from absorbing carbonic acid from the air. If the potassa is to be preserved in solution, the liquid is drawn off with a siphon, and collected in a ground-stoppered bottle. Bottles made of hard green glass are most suitable, inasmuch as those of flint-glass contain more or less oxide of lead, which is attacked by the solution of potassa, so that the latter will be impregnated, after a time, with an appreciable quantity of the oxide, and its efficiency as a test injured. If solid potassa is to be made, the solution is evaporated rapidly in a copper, or still better, in a silver vessel. The ebullition should be very active, in order that the constant evolution of vapour may prevent the contact of the air with the potassa, and the con- sequent abstraction of carbonic acid. The temperature being at length elevated to dull redness, the hydrate of potassa KO,HO, which alone remains, fuses into a liquid of an oily consistence. If a small quantity of carbonate of potassa has been formed during POTASSIUM. 443 the operation, as it fuses only at a much higher temperature, it swims on the surface of the hydrate, and may be skimmed off. The melted hydrate is then poured upon a copper plate, on which it instantly congeals. The potassa is broken into pieces, and pre- served in well-closed bottles. The hydrate of potassa, thus prepared, is the caustic potash of commerce. When purified carbonate of potassa has been used, and the operation carefully conducted, the hydrate of potassa is nearly pure. This is, however, rarely the case with the caustic potash of commerce, for beside the carbonate of potassa used in its manufacture generally containing sulphate and silicate of potassa and chloride of potassium, the decomposition of the car- bonate is rarely complete. § 433. To purify crude caustic potash, it is introduced, broken in small pieces, into a large flask filled with very strong alcohol. The liquid is frequently shaken, and even moderately warmed, to hasten the solution, and then allowed to repose. A crystalline deposit, chiefly composed of sulphate of potassa and chloride of potassium, is formed at the bottom of the flask, above which is a sirupy liquid, formed chiefly of a solution of carbonate of potassa in the water, abstracted from the alcohol. The rest of the liquid is a solution of the monohydrate of potassa in nearly absolute alcohol. The supernatant liquid is drawn off by a syphon, poured into a retort or other suitable apparatus for distilling, and after distilling off about two-thirds of the alcohol, which is absolute, the remaining liquid is poured into a silver dish and evaporated as rapidly as possible. It is lastly heated to dull redness, in order to fuse the hydrate of potassa, which is then formed on the silver plate. The alcoholic solution is generally coloured brown during the evaporation, owing to the alteration of a small portion of alcohol by potassa and the oxygen of the air, forming a brown organic acid, which remains combined with the potassa. But, when the potassa melts, the substance loses its colour entirely, the organic acid being destroyed, and affording carbonic acid, which remains combined with the potassa. The potassa, thus purified, and called alcoholic potassa, always contains a certain quantity of carbonate, but is entirely freed from chlorides and sulphates. If we wish to deprive it entirely of carbonic acid, it is redissolved and boiled with a small quantity of milk of lime, allowed to cool, and kept in a well-closed bottle. The liquid then contains a small quantity of lime in solution, which may be precipitated by the addition of a few drops of car- bonate of potassa. § 434. The decomposition of carbonate of potassa by lime is only effected with ease when the liquid is diluted, so that a very weak solution of potassa is the necessary result, and a great deal of water must be evaporated in order to obtain solid potassa. 444 ALKALINE METALS. When the carbonate is dissolved in a small quantity of water, it cannot be reduced to the caustic state, even by prolonged ebulli- tion with a great excess of lime. Moreover, when a concentrated solution of caustic potassa is boiled with carbonate of lime, the potassa abstracts nearly all the carbonic acid from the lime. It is evident, therefore, that with a solution of carbonate in a certain state of concentration, its decomposition by lime must stop at a given point, which cannot be exceeded by prolonging the opera- tion. We may even retrograde, that is to say, may form again a new quantity of carbonate of potassa, if the liquid becomes too concentrated by boiling. Theoretically, 1 equivalent of lime CaO=28 will decompose 1 equivalent of carbonate of potassa K0,C03=69.2; but experi- ence shows that, in order to obtain a rapid decomposition, at least double the quantity of lime just mentioned should be employed. And so much the more lime must be used, as the solution of potash is concentrated. Caustic potassa (hydrated potassa) presents the form of opaqu<» white masses, with a crystalline fracture. Its density is about 2.1. It melts at a dull red-heat, and volatilizes without alteration at a white-heat. It parts with its water only when in contact with a more powerful acid with which the oxide of potassium can combine. § 435. In order to determine experimentally the quantity of water contained in the hydrate of potassa, a certain quan- tity of the hydrate is weighed rapidly in a platinum crucible, covered by its lid to prevent its absorbing water during weighing. A small quantity of water is added to dissolve the potassa, and a slight excess of sulphuric acid carefully poured in to form sulphate of potassa. It is then evaporated to dryness with care, to avoid loss by projection during the evaporation. The dried substance is calcined at a strong red-heat, to drive off the excess of sulphuric acid and reduce the residue to the state of neutral sulphate of potassa K0,S03. The crucible is again weighed, and gives a weight P of sulphate of potassa, produced by the weight p of hydrate of potassa. If we knew the composition of the sulphate of potassa, we should immediately know the weight q of anhydrous potassa KO contained in the weight P of sulphate of potassa, and should thence conclude that a weight p of hydrate of potassa con- tains a weight q of anhydrous potassa, and consequently a weight {p—q) of water. Let us suppose that we did not know the composition of the sulphate of potassa. The weight P of sulphate of potassa is dis- solved in distilled water, washing the crucible several times, so as not to lose the smallest portion of the substance. An excess of a solution of chloride of barium is poured into the collected waters, slightly acidulated by the addition of a few drops of chlorohydric acid, and heated to the boiling point. The sulphuric acid will be POTASSIUM. 445 completely precipitated in the state of insoluble sulphate of baryta, which is readily deposited in hot liquids. The precipitate being collected on a small filter, washed with distilled water, and then dried, is calcined in contact with the air to burn off the filter, when a weight Q of sulphate of baryta is obtained. Now, we may grant that the composition of this sulphate is known; for, if it were not, it might be determined by the experiment described § 135. It will thus be found that 100 parts of hydrate of potassa KO,HO contain 16 parts of water, corresponding to the following composition in equivalents: 1 eq. of protoxide of potassium 47.2 83.99 1 “ “ water 9.0 16.01 1 “ jn q2 7 « “ 122 “ 16.8 “ “ 212 “ 25.3 “ From this it will be seen that the solubility of sulphate of po- tassa increases in proportion to the temperature ; or, in other words, it is represented by a right line. (See plate at page 40T.) It is in- soluble in absolute alcohol. If the preceding sulphate be dissolved in an excess of sulphuric acid, a liquid is obtained which gives, on evaporation, another crystallizable sulphate, termed the bisulphate of potassa, but is more correctly called a double sulphate of potassa and water. Its for- mula is K0,S03+H0,S03. Heated to 392°, it fuses Avithout de- composition and Avithout parting Avith its water. At a higher tem- perature, it gives off monohydrated sulphuric acid, and the simple sulphate K0,S03 remains. Concentrated alcohol also removes the sulphate of Avater, and leaves the sulphate K0,S03. The double sulphate of potassa and water may also combine in other proportions. If to the simple sulphate of potassa K0,S03, a quantity of sulphuric acid be added equal to one-half of that which the salt contains, a crystallized salt is obtained with the for- mula 4(K0,S08)+H0,S03. By treating this salt with a small quantity of water, it is decomposed into the simple sulphate, Avhich remains, and the double sulphate K0,S03-f H0,S03, which is dis- solved : 4(K0,S03)+H0jS03=3(K0,S03)+(K0,S03+H0,S03). Chlorate of Potassa. §454. Chlorate of potassa K0,C105 is an anhydrous salt, which crystallizes in the form of small spangles. They are, how- POTASSIUM. 465 ever, larger when crystallization is effected slowly. It is much more soluble in hot than in cold water: 100 parts of water at 32° dissolve 3.33 pts. of chlorate of potassa. “ 56° “ 5.60 . “ “ 27f° “ 6.03 “ “ 76° “ 8.44 “ “ 95° “ 12.05 “ “ 120J° “ 18.96 “ “ 166|° “ 35.40 “ “ 220|° “ 60.24 “ Its solubility, therefore, increases rapidly with the temperature, and is represented by a curve, the convexity of which is turned to- ward the axis of temperature. (See plate at page 407.) Alcohol does not appreciably dissolve it. Chlorate of potassa fuses at about 750°. At a higher heat, it parts with its oxygen, and is ultimately reduced to chloride of potassium. It deflagrates vividly on ignited coals. It is one of the most active oxidizing agents, and forms explosive mixtures with the majority of combustible substances. Thus, an intimate mixture of it and sulphur produces a violent detonation when placed on an anvil and struck with a hammer. These mixtures should be made with great care and in small quantities, to avoid accidents. The detonating mixtures formed of chlorate of potassa are much more powerful than the corresponding mixtures made with nitre. Gunpowder, much superior to that in ordinary use, has been made with chlorate of potassa; but it was excessively explosive, and burst the firearms in which it was used. Its preparation and preservation being very dangerous, its manufacture has been abandoned. A mixture of chlorate of potassa and sulphur has also been used in the fabrication of percussion caps for guns, but fulminating mercury is now preferred. If a drop of concentrated sulphuric acid be thrown on a mix- ture of sulphur and chlorate of potassa, the sulphur takes fire. Advantage was taken of this property for producing fire, and the process was generally followed until it was superseded by the phos- phoric matches described § 208. A paste was made with 30 parts of chlorate' of potassa and gum- water, and 10 parts of flowers of sulphur added, with a little cin- nabar to colour it. The end of each match, previously covered with sulphur, was dipped into the mixture, and allowed to dry. On the other hand, a small glass bottle was prepared, containing as- bestus imbued with oil of vitriol. On plunging the match into this bottle, the paste of sulphur and chlorate became moistened by the sulphuric acid; it took fire, and the combustion extended to the sulphur, and thence to the match. The bottles were kept well 466 ALKALINE METALS. stopped ; otherwise the sulphuric acid would have attracted moisture from the air, and its action on the mixture of chlorate and sulphur ■would not have been sufficiently powerful to excite combustion.* § 455. Chlorate of potassa is obtained by acting with chlorine upon a concentrated solution of potassa, the reaction taking place between 6 equivs. of chlorine and 6 equivs. of potassa: 6K0+6C1=5KC1+K0,C10S. Chlorate of potassa, being much less soluble in cold water than chloride of potassium, separates in the form of crystalline spangles, while the chloride remains in solution. To prepare a large quan- tity of the chlorate, the apparatus is arranged in a peculiar way. As the tube, conveying chlorine into the solution of potassa, might be obstructed by crystals of chlorate, it is better to select a large tube, or better still, to arrange the apparatus as in fig. 338. Fig. 338. Chlorine gas is generated in the flask A, and washed in the bottle B, containing water. The bottle C contains the solution of po- tassa, or, preferably, of carbonate of potassa, as this salt is less ex- pensive. Two tubes pass through the cork of the bottle C ; a nar- row one cd, which allows the escape of gas, and a straight tube ab, 15 mm. (J inch) in diameter, open at both ends, and descending nearly to the bottom of the bottle. The tube ef of the washing-bot- tle, is inserted into the tube ab by means of a cork a. If the end b of the large tube be closed by the deposition of crystals, it can readily be cleared by a glass rod introduced through the open- ing a. During the first stage of the operation, bicarbonate of potassa, * The paste for these matches was generally made of a mixture of the above substances, with a quantity of sugar. They were superseded by the lucifer match, which was a flat splint of wood tipped by a mixture of chlorate, sulphur, and gum, and was drawn rapidly between sand-paper to inflame it. These again gave place to the locofoco or phosphoric match.—J. C. B. POTASSIUM. 467 chloride of potassium, and hypochlorite of potassa are formed, and very little chlorate. The greater part of the chloride being de- posited in crystals, it is better, at this period, to interrupt the pro- cess, in order to allow as much as possible of the chloride to de- posit. The supernatant liquid is decanted, and again subjected to the action of the chlorine until it is supersaturated, when a crystalline deposit of chlorate is formed, which is more copious if the liquid be allowed to cool completely. The mother liquor affords, by evaporation, an additional quantity of chlorate. The chlorate thus obtained always contains some chloride, to remove which it is first treated with a small quantity of cold water, which dissolves the greater part of the chloride, and the remaining crys- tals are dissolved in boiling water. The liquid, on cooling, deposits chlorate of potassa nearly pure. Hypochlorite of Potassa. § 456. By passing chlorine through a cold and dilute solution of carbonate of potassa, a liquid is obtained which contains chlo- ride of potassium and hypochlorite of potassa. 2K0+2C1=KC1+K0,C10. This solution destroys vegetable colours rapidly, and is there- fore used in bleaching; but, on a large scale, the cheaper hypo- chlorite of lime is preferred. In commerce, this bleaching liquid is called Javelle water, because it was first prepared at Javelle, near Paris.* Oxalates of Potassa. §457. Oxalic acid forms three compounds with potassa. By neutralizing a solution of potassa with oxalic acid, a solution is obtained which crystallizes on evaporation, in the state of a neu- tral oxalate, with the formula K0,C303+H0. It is soluble in 3 parts of cold water. If oxalic acid be added to a solution of the neutral oxalate, a second crystallizable oxalate is obtained, the binoxalate, with the formula K0,C303+H0,C303-f-2H0. It may be regarded as a double oxalate, composed of the neutral oxalate of potassa and monohydrated oxalic acid. It requires 6 parts of boiling and 40 parts of cold water for its solution, so that it is easily separated from the neutral oxalate by crystallization. The binoxalate exists in the juice of many plants, and the acidity of sorrel is chiefly due to it. A large quantity is extracted from these juices, and is com- mercially known as salt of sorrel. If to the binoxalate of potassa a quantity of oxalic acid be added * It is usually prepared from carbonate of soda in England and the U. S., and termed bleaching soda.—J. C. B. 468 ALKALINE METALS. equal to that which it already contains, and the whole be dissolved in a small quantity of boiling water, there is deposited, on cooling, a quadroxalate with the formula K0,4C303+7H0. It should probably be regarded as a double oxalate, formed by the combina- tion of 1 equiv. of the simple oxalate K0,Ca03 and 3 equivs. of monohydrated oxalic acid H0,C303, and its formula should be written K0,Cs03-f 3(H0,C303)+4H0. COMPOUNDS OF POTASSIUM AND SULPHUR. § 458. Many compounds of sulphur and potassium are known, of which chemists admit five, viz : Monosulphide of potassium KS, corresponding to the protoxide KO, Bisulphide “ KS2, Trisulphide “ KS3, corresponding to the peroxide KOa, Quadrisulphide “ KS4, Pentasulphide “ KSs. The monosulphide is obtained by heating in a crucible a mixture of sulphate of potassa and charcoal: K0,S08+4C=KS+4C0. The sulphide fuses into a reddish mass. If a mixture of 2 parts of sulphate of potassa and 1 part of lampblack be heated, a very divided sulphide is obtained, the particles of which cannot unite together, on account of the particles of carbon with which they are intimately mixed. This sulphide is so inflammable, that it takes fire on being projected into the air, and is hence calledpyrophorus. The monosulphide, as prepared above, is never pure, but always contains a small quantity of poly sulphide, which is readily detected by pouring into its solution an excess of acid, when a slight deposit of sulphur always ensues: it would not occur if the solution con- tained only monosulphide (§ 394). The best mode of preparing the monosulphide consists in divid- ing a solution of potassa into two equal parts, saturating one of them with sulphuretted hydrogen, and then mixing it with the other, which is in the state of caustic potassa. The solution of po- tassa, saturated with sulphydric acid, is converted into a sulpho- salt, the sulphydrate of monosulphide of potassium KS,HS. When this salt is mixed with a quantity of potassa equal to that which produced it, it gives a simple monosulphide: KS,HS+KO,HO=2KS+2HO. By evaporating the liquid, the monosulphide is deposited as a colourless, crystalline mass. The other sulphides are readily prepared from the monosulphide POTASSIUM. 469 by heating 1 equiv. of it with 1, 2, 3, or 4 equivs. of sulphur. The pentasulphide KS is most easily obtained, it being sufficient to heat the monosulphide with an excess of sulphur, and raise the temperature so high that the sulphur, which cannot enter into com- bination, is volatilized. The temperature should not, however, be raised to bright redness, for the pentasulphide would then give off a portion of its sulphur, and pass into the state of trisulphide. Pentasulphide is produced under many other circumstances. If a mixture of carbonate of potassa and sulphur be heated, the reaction begins at the fusing point of sulphur, and carbonic acid is disengaged. If the sulphur be in excess, and the temperature be not elevated above 480°, pentasulphide of potassium and hypo- sulphite of potassa are formed, and remain mixed with the excess of sulphur: 3KO+12S=2KS5+KO,SA- But, if the mixture be heated to redness, the hyposulphite is de- stroyed, and the excess of sulphur distils over, so that the pentasul- phide of potassium and sulphate of potassa are obtained. In fact, at a red-heat, the hyposulphite is converted into pentasulphide and sulphate: 12(K0,Sa03)=9(K0,S03)+3KSs. Pentasulphide may be separated from sulphate by treating the mixture with alcohol, which dissolves only the sulphide. If charcoal be added to the mixture of carbonate of potassa and sulphur, at a dull red-heat, pentasulphide of potassium alone is formed. This compound may also be obtained in the humid way, by boil- ing a solution of caustic potassa with an excess of sulphur. A large proportion of the sulphur is dissolved, and a liquid of a deep orange-yellow colour is formed, containing pentasulphide and hy- posulphite. The pentasulphide obtained by any of these processes is called liver of sulphur, and is used in cutaneous diseases. SULPHOSALTS FORMED BY THE MONOSULPHIDE OF POTASSIUM. § 459. Monosulphide of potassium combines with a great number of electronegative sulphides, with which it forms true salts; but the majority of them have not yet been studied with sufficient care. The most important are the sulfhydrate and sulphocarbonate of sulphide of potassium. Sulfhydrate of potassium* is obtained by passing, to saturation, * As we say, in English, arseniate of iron, meaning thereby oxyarseniate of ox- ide of iron, so we may say, for brevity, sulpharseniate of iron, instead of the long name sulpharseniate of sulphide of iron. In like manner, we prefer the abbre- viated expression sulfhydrate of potassium, for sulfhydrate of sulphide of po- tassium.—J. C. B. 470 ALKALINE METALS. a current of sulfhydric acid through a solution of potassa. The concentrated liquid deposits crystals, of which the formula is KS + IIS or KS,HS. It will be seen to correspond exactly to the hydrate of potassa KO + HO, in which an equivalent quantity of oxygen takes the place of sulphur. The sulphocarbonate of potassium is obtained by pouring bisul- phide of carbon into an alcoholic solution of monosulphide of po- tassium. An orange-coloured crystalline deposit is formed, the sulphocarbonate of sulphide of potassium KS,CS2, which may be redissolved in water or boiling alcohol and crystallized. COMPOUND OF POTASSIUM AND CHLORINE. § 460. Only one compound of potassium and chlorine is known, and is obtained by neutralizing a solution of potassa or carbonate of potassa with chlorohydric acid. The liquid on evaporation de- posits cubic anhydrous crystals of chloride of potassium, KC1. The density of this chloride is about 1.84; it fuses at a red-heat, without decomposition, and, at a higher temperature, gives off a considerable quantity of vapour. In France, chloride of potassium is extracted from varec, derived from certain plants growing on the rocks covered by the waters of the ocean. After being gathered on the shore, where they are thrown by the waves, they are dried, and reduced to ashes in small pits made in the ground. A half-melted ash remains, which is termed varec-soda. It is lixiviated hot, and the various salts it contains extracted by successive crystallizations. It generally fur- nishes 80 per cent, of chloride of potassium. Chloride of potassium is obtained, as a secondary product, in many of the arts. A considerable quantity is formed in refining crude potash furnished by the lixiviation of ashes. It is also ex- tracted from the mother liquor remaining from the refining of nitre. We have seen (§ 455) that it is obtained in the manufacture of chlorate of potassa. Lastly, the ashes of the leaves and stems of tobacco contain a considerable quantity of it. It is a valuable salt, because it is readily converted into other salts of potassa, by double decomposition. It may be used to convert the nitrate of lime into nitrate of potassa, in the extraction of nitre from nitrified substances. It is also used in the preparation of alum. As chloride of potassium produces a considerable depression of temperature by solution in water, advantage is taken of this pro- perty, as will be shown hereafter, to ascertain the proportions of chlorides of potassium and sodium contained in a mixture of the two salts. COMPOUND OF POTASSIUM AND IODINE. § 461. Iodide of potassium is obtained bj dissolving iodine in a concentrated solution of potassa until the liquid is coloured by an POTASSIUM. 471 excess of iodine. A crystalline deposit of iodate of potassa is formed, and the liquid contains, at the same time, iodide of potas- sium and some iodate. If the iodide only is to be prepared, the liquid is evaporated to dryness, and the residue calcined in a pla- tinum crucible. The iodate is decomposed, and nothing remains but the iodide, which is redissolved in wrater and crystallized. Iodide of potassium forms anhydrous cubic crystals. A considerable quantity of iodide of potassium may be extracted from the mother waters of varec by crystallization, when they have deposited chlorides of potassium and sodium, as well as the sul- phates which they hold in solution. COMPOUND OF POTASSIUM AND CYANOGEN. § 462. Cyanide of potassium is most readily obtained by decom- posing, at a red-heat, the double cyanide of potassium and iron 2KCy+FeCy, commonly called prussiate of potash. The cyanide of iron alone decomposes, forming an insoluble compound of iron and carbon. The residue, treated with water, yields cyanide of potassium, which is crystallized by evaporation in anhydrous cubes. The manufacture of the impure cyanide of potassium, from which the double cyanide of iron and potassium is made, will be described under the compounds of iron. DISTINCTIVE CHARACTERS OP THE SALTS OF POTASSA. § 468. Alkaline salts are distinguished from all other metallic salts, by affording no precipitate with a solution of alkaline car- bonates. The salts of potassa are recognised by the following properties: 1. By their physical properties, chiefly by those of the sulphate of potassa, an anhydrous, easily crystallizable salt, having a cer- tain degree of hardness. 2. By forming with the sulphate of alumina a double salt, alum, which crystallizes in regular octahedrons. It is sufficient to pour into a concentrated solution of a salt of potassa, a concentrated solution of sulphate of alumina, and shake the liquid, to have a precipitate of alum, composed of small regular octahedrons, easily recognised by the microscope. 3. By forming with tartaric acid a bitartrate of potassa, slightly soluble in water, so that if a solution of tartaric acid be poured in excess into a slightly concentrated solution of a salt of potassa, a precipitate is formed. 4. By affording, with the bichloride of platinum, a yellow pre- cipitate of the double chloride of potassium and platinum, when the solution is not very dilute; and which is more copious if a cer- tain quantity of alcohol be added to the solution. The precipitate 472 ALKALINE METALS. of this double salt is destroyed by a red-heat; bichloride of plati- num being decomposed and metallic platinum remaining, and the chloride of potassium being set free. By treating the residue with water, chloride of potassium alone is dissolved. 5. The salts of potassa give, with a solution of silicofluohydric acid, a translucent, gelatinous precipitate of the double fluoride of potassium and silicium, at first scarcely visible in the liquid, but depositing, after some time, in the form of a colourless, nearly transparent jelly. 473 SODIUM. § 464. Sodium exists, in the state of silicate of soda, in certain minerals constituting primary rocks. Combined with chlorine, it forms chloride of sodium, or sea-salt, which is in solution in sea- water, and forms, in many countries, considerable masses among strata of the trias group. Plants growing on the borders of the ocean absorb a considerable quantity of the salts of soda, which are found in their ashes. Sodium closely resembles potassium in its physical properties. It is brittle at a low temperature, and then presents a crystalline fracture ; at the ordinary temperatures of 60° or 68°, it is so soft as to be easily cut with a knife; about 140°, it may be moulded like wax, and at about 194° becomes liquid. It boils at a red-heat, and distils at a lower temperature than potassium. Sodium, when freshly cut, has a brilliant metallic lustre re- sembling silver; but the brilliancy is of short duration in the air, as the metal rapidly combines with oxygen. It should therefore be preserved in naphtha. Its density is greater than that of po- tassium, being about 0.97 at ordinary temperatures. Sodium decomposes water at the lowest temperatures. A piece of it thrown on water, fuses into a brilliant globule, from the heat disengaged by its oxidation, and runs over the surface of the liquid, but without the inflammation exhibited by potassium. The gas, however, may be inflamed by keeping the globule stationary on the surface of the liquid, whereby there is less loss of heat, and the temperature rises sufficiently high to inflame the hydrogen. Com- bustion likewise ensues if the metal be thrown on water thickened by gum or starch. The liquid becomes strongly alkaline from the hydrate of protoxide of sodium or soda, which is dissolved. § 465. Sodium is obtained by the same processes as potassium. It was first procured, in small quantities, by the decomposition of soda by the voltaic pile (§ 427), and subsequently by decomposing the hydrate of soda by incandescent iron (§ 428). It is now pre- pared by decomposing carbonate of soda by charcoal at a high temperature, in a wrought-iron vessel, in the same manner as pre- scribed for potassium (§ 429). To obtain an intimate mixture of carbonate of soda and charcoal, a given amount of pure carbonate of soda is dissolved in the smallest quantity of hot water possible, and about one-third as much powdered charcoal added to the liquid. A homogeneous paste is made, which is completely dried, and a certain quantity of charcoal, in small pieces, is added, to render the mass more porous. This mixture is placed in the iron Equivalent =23. 474 ALKALINE METALS. bottle. The preparation of sodium is easier than that of potas- sium, in consequence of the lower temperature at which it distils. COMPOUNDS OF SODIUM AND OXYGEN. § 466. Sodium forms two compounds with oxygen, which resem- ble those of potassium, and are prepared in the same manner. Sodium heated in dry oxygen burns, and is converted into a peroxide Na03. When the peroxide is heated with twice as much sodium as it already contains, it gives the anhydrous protoxide of sodium. When sodium is oxidized by decomposing water, the protoxide is still formed, but it combines immediately with the water, producing a hydrate which cannot be decomposed by heat. The composition of protoxide of sodium is deduced from the analysis of chloride of sodium, like that of protoxide of potassium from its chloride (§ 435). It has been found to contain Sodium — 74.19 Oxygen 25.81 100.00 The equivalent of sodium is then given by the proportion: 25.81 : 74.19 : : 8 : x; whence a;=23. SALTS FORMED BY THE PROTOXIDE OF SODIUM OR SODA. § 467. The protoxide is the only oxide of sodium which plays the part of a salifiable base, and affords a great number of salts, of the highest importance. Hydrate of soda. § 468. This compound is formed when sodium is oxidized by contact with water. It is prepared in the laboratory by decom- posing carbonate of soda in solution by hydrate of lime, precisely according to the process for preparing the hydrate of potassa (§ 432), and demanding like precautions. To decompose 1 eq. of dry carbonate of soda... 53 1 “ of anhydrous lime 28 is required; but in order to make the reaction complete and rapid, it is better to use twice as much lime, so that nearly equal weights of carbonate of soda and lime are employed. The caustic solu- tion, separated from the sediment by decantation, is rapidly eva- porated to dryness, and the fused caustic soda poured upon a copper plate, where it solidifies on cooling. It may be purified by dissolving it in alcohol, as was done with potassa (§ 433). Hydrate of soda resembles perfectly, in appearance, hydrate of potassa, and, like the latter, contains 1 equiv. of water: SODIUM. 475 1 eq. of protoxide of sodium 31 77.5 1 “ water... 9 22.5 1 “ hydrate of soda. 40 100.0 It does not part with its water at any temperature, and distils, without change, at a strong red-heat. It may be used for the same purposes as caustic potassa, and is preferable to the latter, being cheaper and more readily obtained in a state of purity; because very pure carbonate of soda is found in commerce. A solution of hydrate of soda, concentrated when hot, deposits, on cooling, crystals more hydrated, but their exact composition has not yet been ascertained. The solid hydrate, exposed to the air, soon deliquesces by combining with the water of the atmosphere ; but if the sirupy liquid be left indefinitely exposed, crystals of carbonate of soda form, presenting an appearance of efflorescence. Hydrate of potassa does not exhibit similar phenomena, because its carbonate is deliquescent. Sulphate of Soda. § 469. Soda and sulphuric acid combine in several proportions; the most important compounds being the neutral sulphate and the bisulphate. Neutral sulphate of soda was formerly called Glauber's salt, and is found, in commerce, in large crystals, containing more than one-half of their weight of water, and with the formula Na.0,S03 +10HO. 1 eq. of anhydrous sulphate of soda 71 44.1 10 “ water 90 55.9 1 “ crystallized sulphate of soda 161 100.0 The crystallized salt fuses in its own water of crystallization, at a slightly elevated temperature. If the heat be continued, a portion of the water is driven off in vapour, and a crystallized de- posit is formed of anhydrous sulphate of soda. The same anhy- drous sulphate is deposited when the aqueous solutions are crystal- lized, at a temperature above 91 J°. The sulphate with 10 equivs. of water, is only formed when crystallization takes place at a tem- perature below 68°. It effloresces in the air, and its crystals soon fall to powder. If the crystallization takes place between 68° and 86°, there is still formed a hydrated sulphate, which contains, how- ever, less water than the preceding. The crystals of this hydrate are unalterable in the air. The crystals of anhydrous sulphate likewise fall to powder in the air, but this efflorescence is due to an opposite cause from that of the sulphate with 10 equivs. of water, and is owing to the fact that the anhydrous sulphate takes water from the atmosphere, and dis- aggregates by changing into the second hydrate. 476 ALKALINE METALS. The solubility of this salt presents a remarkable anomaly, of which we have already spoken (§ 369). Below 32° its solubility is hut feeble, for 100 parts of water at 32° only dissolve 5; but it increases rapidly with the temperature to 91.4, which is its maxi- mum. 100 parts of water then dissolve 322 parts of the sulphate with 10 equivs. of water. Its solubility after this diminishes with the temperature. We have graphically represented, on the plate an- nexed to page 407, the curve of solubility of the sulphate, supposed to be anhydrous. We have also described (§ 363) a remarkable peculiarity in the crystallization of this salt. Sulphate of soda exists in small quantities in the waters of the ocean and of many saline springs ; and a large quantity may be extracted from the mother liquors after making salt, by merely re- ducing them to a very low temperature, in order to diminish as much as possible the solubility of the sulphate of soda. Large quantities are thus collected in the vicinity of Montpellier, during winter, from the mother waters of the salines. We shall hereafter treat of this manufacture. § 470. The greater portion of the sulphate of soda consumed in France is prepared by decomposing sea-salt by sulphuric acid. This process was explained in § 184, where chlorohydric acid is obtained at the same time. In localities where this acid is not sold, the decomposition is effected in a reverberatory furnace ; but as the acid fumes cannot be allowed to escape into the air, to the injury of the surrounding vegetation, the manufacturers are obliged to condense them by passing the gas from the furnace through brick flues, in which water flows constantly, and which communi- cate with a chimney the draught of which is increased by a special fire. The acid waters are carried off by wells, as much as prac- ticable. § 471. Sulphate of soda may be fused at a red-heat, without de- composition. By adding to a solution of it a quantity of sul- phuric acid equal to that which it already contains, an acid liquid is obtained, which gives, after proper evaporation, crystals of the bisulphate of soda, with the formula NaO,2SOs+3HO. The for- mula may also be written NaO,S03-f H0,S03+2II0; in which case it is regarded as a double sulphate of soda and water. This salt, heated with caution, first fuses in its crystal water, and then readily parts with the two equivalents; but if the heat be con- tinued, its third equivalent of water may be driven off, and an an- hydrous salt obtained which is a true bisulphate. The anhydrous bisulphate, heated in a retort, gives off one-half of its acid in the state of anhydrous sulphuric acid (§ 138). By adding to the solution of the neutral sulphate smaller quan- tities of sulphuric acid, another double sulphate of soda and water may be obtained, with the formula 3(Na0,S03)+H0,S03+2H0. SODIUM. 477 Carbonates of Soda. § 472. Neutral carbonate of soda is a salt of vast importance, on account of its uses in the arts, and may be prepared in various ways. For a long time it was only obtained from the lixiviation of the ashes arising from the combustion of plants growing on the shores of the sea. Plants growing inland alford ashes highly charged with the salts of potassa, while marine plants contain prin- cipally salts of soda. The various marine plants furnish very dif- ferent proportions of soda; those which are richest being the salsola soda and the salicornia europoea. The sea-wrack, or varec, contains but little carbonate of soda, but a considerable quantity of sulphate and chloride. Spain formerly furnished the greater part of the carbonate used in Europe, and it was called Alicant, or Malaga soda, or barilla.* A certain quantity was also collected on the shores of the Mediterranean in France, and called Nar- bonne soda. During the wars of the Revolution and the Empire, as the supply of Spanish soda was very limited, it rose to a very high price. Chemists, encouraged by the government, made many attempts to manufacture the article, from materials indigenous to the country. After many experiments, which answered more or less perfectly, one was found which furnished soda in any quantity and at a reasonable price. This process, still in use, is called the process of Leblanc, from the French physician who discovered it. The process of Leblanc consists first in converting chloride of sodium into sulphate of soda by sulphuric acid; then in decom- posing the sulphate by heat, with a mixture of carbonate of lime and charcoal. Carbonate of soda and oxysulphide of calcium are the result, and are easily separated. If a mixture of 1 equiv. of sulphate of soda and 1 equiv. of car- bonate of lime be fused together, there is a double decomposition, sulphate of lime and carbonate of soda being formed; but, if it be now treated with water, the greater part of the carbonate of soda passes again into the state of sulphate. This inverse reaction, ef- fected by water, is owing to the fact that carbonate of lime is much more insoluble than the sulphate. If 4 equivs. of charcoal be added to the mixture of 1 equiv. of sulphate of lime, and 1 equiv. of carbonate of lime, the sulphate of lime is changed into a sulphide by the action of heat; and, as this sulphide is but slightly soluble in water, we may hope to separate carbonate of soda by water. The separation is, however, imperfect, for a certain quan- tity of carbonate of lime is always formed, and sulphide of sodium dissolves at the same time with carbonate of soda. But if we heat together * It was also largely prepared on the coast of Scotland, and termed kelp. 478 ALKALINE METALS. 2 eq. of sulphate of soda 142 3 “ carbonate of lime 150 9 “ carbon 54 ~346 the reaction takes place in the following manner :— 2(Na0,S03)+3(Ca0,C03)+9C=2(Na05C02)+2(CaS,Ca0) + 10CO, the 2 equivs. of sulphide of calcium combining with 1 equiv. of lime to form an oxysulphide of calcium 2CaS,CaO, perfectly insoluble in water. Water dissolves out from this substance only the carbonate of soda. The best proportions which have been found by experience, and which do not differ essentially from those found by theory, are 1000 of anhydrous sulphate of soda, 1040 of carbonate of lime, 530 of charcoal. The reaction takes place on the brick hearth of a reverberatory furnace, where the mixture is heated to the point of fusion and constantly stirred. Carbonic oxide gas is evolved, and burns with small jets of bluish flame. When the evolution of this gas has ceased, the workman withdraws a small quantity of the mass from the furnace, in order to judge, by its homogeneous appearance, if the reaction is perfected. He then rakes out the doughy sub- stance, which, being cooled, is pulverized in a vertical mill, and then lixiviated in boxes, as described §447. The liquids are evaporated in kettles to the crystallizing point, and then run into the crystallizers, where the salt is deposited. To obtain pure car- bonate of soda, the crystallization must be disturbed by continually stirring the liquid, whereby small granular crystals are deposited, which are gradually removed, and then washed with pure water. Fig. 339. Fig. 339 represents a vertical section of a soda-furnace of re- cent construction, in which the heat of the fuel is greatly econo- mized. The flame of the fuel, which burns on the grate F, tra- verses the compartment A, where the highest temperature pre- SODIUM. 479 vails; passes into the compartment B, which is only separated from A by a low brick-wall; traverses the compartment C, and escapes by the chimney 0. The hearth of the compartments A and B is of brick ; that of C is made like a water-tight kettle. The roof of C is formed by a sheet-iron kettle D. The mixture of sulphate of soda, carbonate of lime, and charcoal, is first heated in the compartment B, and then conveyed into A, where the chemical reaction takes place. The liquors from lixiviation of the crude soda being concentrated by heat in the kettle D, are con- veyed into the kettle C, where the evaporation is finished. The carbonate of soda is withdrawn perfectly dry from the furnace. The soda furnaces at Marseilles are usually composed only of the compartments A and B. The former, A, is destined for the fabrication of carbonate of soda, while in B the sulphate is pre- pared by the reaction of sulphuric acid on chloride of sodium. The chlorohydric acid gas which is evolved traverses the condens- ing tubes, where the greater portion of it is dissolved. §473. Carbonate of soda crystallizes when cold, in large crystals, containing 62.9 per cent, of water, with the formula Na0,C03+ 10HO. They soon effloresce in the air, and are more soluble in hot water than in cold. Boiling water dissolves nearly its own weight of them, while cold water only dissolves one-half. It may also crys- tallize with a smaller proportion of water when formed in a hot solution; for the small granular crystals deposited from a liquid concentrated by boiling contain only about 18 per cent, of water. Carbonate of soda, when heated, soon parts with its water, and, at a red-heat, fuses into a very fluid liquid, which crystallizes on cool- ing- §474. Bicarbonate of Soda.—The bicarbonate is obtained by passing a current of carbonic acid gas through a concentrated solu- tion of the neutral carbonate, and, not being very soluble in water, is deposited in the form of crystals. Its composition is NaO,2COa+HO or Na0,C03+H0,C02. 100 parts of water, at ordinary temperatures, dissolve about 8 parts of this salt. The bicarbonate is easily decomposed by the action of heat, and parts with one-half of its carbonic acid, leaving a neutral carbo- nate. Its solution is also decomposed by heat, and prolonged ebullition changes it into the neutral carbonate. The bicarbonate is extensively used in medicine as an antacid. Small pastilles, made by mixing it with some sweetened vehicle, are called Darcet’s digestive pastilles. Bicarbonate of soda is manufactured by exposing, in wooden boxes, the crystallized neutral carbonate to a current of carbonic acid gas, which changes it entirely into bicarbonate. The carbonic acid employed is sometimes derived from natural sources. 480 ALKALINE METALS. §475. Sesquicarbonate of Soda.—A crystallized sesquicarbonate, with the formula 2Na0,3C03+4H0, is found in nature, and known by the name of natron, or sal trona. In certain hot countries, as Egypt, Mexico, and India, during the rainy season, small lakes or ponds are formed in low spots, the waters of which evaporate during the hot season, and develop efflorescences or crystalline masses of natron. This salt does not effloresce in the air, and often presents a considerable degree of hardness. It appears to be formed by the reaction of carbonate of lime on sea-salt; at least, a considerable quantity of sesquicarbonate of soda SWa,3Na0,P05-f-24H0.—If an excess of carbonate of soda be added to a solution of the preceding phos- phate, and the concentrated liquid be crystallized, a salt is obtained with a strong alkaline reaction, and composed according to the for- 482 ALKALINE METALS. mula 3NaO,POs+24IIO. It readily parts with its water by heat, and then presents the composition 3NaO,POs. It may be fused without changing its state of tribasic phosphate ; and by resolution in water, it crystallizes again as the original phosphate 3NaO,POs +24110. A neutral solution of nitrate of silver, poured into the solution of this phosphate, with its strong alkaline reaction, gives a yellow precipitate of phosphate of silver 3AgO,POs, and the liquid, after precipitation, is neutral to coloured reagents, as represented by the following equation: 3NaO,POs-f 3(Ag0,N05)=3Ag0,P0s-{-3(Na0,N05). The liquid therefore contains, after precipitation, 3 equivs. of nitrate of soda, and must be neutral to coloured tests. 3. Tribasic Phosphate of Soda, (NaO+2HO)POs+2HO.—If an excess of phosphoric acid be added to the first tribasic phos- phate (2NaO + II0)P05+ 24IIO, and the liquid concentrated, a third phosphate NaO,POs+4IIO is obtained, and should be written (NaO-f 2HO)POS+2HO, for the phosphoric acid in it is always saturated by 3 equivs. of base ; but only one of them is soda—the two others are water. This salt has a strong acid reaction. By heating it moderately, its 2 equivs. of water of crystallization can be driven olf, and the salt reduced to the state (NaO + 2HO) P05. The salt still preserves its 3 equivs. of base, for if it be dis- solved in water, and again crystallized, the original salt (NaO-f- 2H0)P05-}-2II0 is obtained. But if the temperature be raised still higher, we can abstract from it successively, first one, and then two equivalents of basic water. In the first case, the phos- phoric acid is only saturated by 2 equivs. of base (Na0+II0)P05, and the salt has become bibasie phosphate, or a pyrophosphate. In the second case, it contains only 1 equiv. of base NaO,POs, and has become a monobasic phosphate or metaphosphate. By redis- solving these new salts in water, they preserve the modification imparted to them by the heat, and no longer reproduce crystals of the original salt (Na0 + 2II0)P05+2H0. A neutral solution of nitrate of silver, poured into the solution of the third tribasic phosphate (NaO + 21IO)POs+ 2HO, which has a strong acid reaction, gives a yellow precipitate of phosphate of silver 3AgO,POs, and the liquid then exhibits an acid reaction, as may thus be shown: (Na0+2H0)P05+3(Ag0,N05)=3Ag0,P05+Na0,N05 +2(H0,N0S). Thus, the liquid contains, after precipitation, 1 equiv. of nitrate of soda, and 2 equivs. of nitrate of water, so that its reaction must be strongly acid. SODIUM. 483 Bibasic Phosphates of Soda, or Pyrophosphates. §480. It was shown (§479) that by calcining the first tribasic phosphate of soda (2NaO-f HO)POs-f24HO, so as to expel, not only its water of crystallization, but also its basic water, a salt is obtained which, when redissolved in water, no longer reproduced the original salt. In fact, if the ordinary phosphate be subjected to igneous fusion, and crystallized a second time in water, a new salt is obtained, the formula of which is 2NaO,POs-f 10HO, and its properties are very different from those of the phosphate from which it originated. It is generally called the pyrophosphate of soda, and does not effloresce in the air. A neutral solution of nitrate of silver, poured into the solution of this salt, which has an alkaline reaction, gives a white precipi- tate of phosphate of silver 2AgO,POs, which presents, conse- quently, a different composition from the yellow precipitate afforded by the tribasic phosphates; moreover, the liquid is neutral after precipitation. The reaction now takes place between 1 equiv. of bi- basic phosphate and 2 equivs. of nitrate of silver, so that the liquid contains, after precipitation, only 2 equivs. of ni- trate of soda, and is consequently neutral to coloured reagents. Second Bibasic Phosphate of Soda.—If an excess of phosphoric acid be added to a solution of the preceding salt, a crystallized salt NaO,POs+HO is obtained ; but its formula should be written (NaO + HO)POs, for the phosphoric acid is always -combined in it with 2 equivs. of base, 1 equiv. of soda, and 1 equiv. of water. This salt has a strong acid reaction. If it be heated only so far as to drive off its crystal-water, and not its basic water, (which last requires a higher temperature,) the salt is not changed ; for by resolution in water the original salt is reproduced. But if it be heated so as to drive off its basic water, the phosphoric acid remains combined with only 1 equiv. of base, and the salt is entirely changed, for it no longer reproduces the original compound when redissolved in water. It has then become a monobasic phosphate or a metaphosphate. The bibasic phosphate of soda (NaQ-f HO)POs gives, with ni- trate of silver, the white precipitate 2AgO,POs of phosphate of silver, and the liquid becomes acid, as shown by the equation, (Na0+H0)P0s+2(Ag0,N05)=2Ag0,P05+Na0,N05 +H0,N05. The same bibasic phosphate (NaO-f HO)POsis obtained by heat- ing the third tribasic phosphate (NaO+2HO)POs-f-2HO, so as to expel, not only its water of crystallization, but also 1 equiv. of its basic water. 2Na0,P05+2(Ag0,N05)=2AgO,P05+2(Na0,N05); § 481. If the third tribasic phosphate (Na0+2H0)P05-f 2110, Monobasic Phosphate of Soda, or Metaphosphate. 484 ALKALINE METALS. or the second bibasic phosphate (NaO + IIO)POs, be heated to fu- sion, both the basic and crystal-water are entirely driven off, the phosphoric acid remains combined with a single equivalent of base, and a salt is obtained the properties of which are very different from those of the substances from which it originated. In fact, this monobasic 'phosphate, also called metaphosphate, is a deliquescent salt, which does not crystallize when dissolved in watei\ With nitrate of silver it gives a white precipitate, but which differs greatly from the white precipitate of the bibasic phosphates ; for its composition is AgO,POs. Double decomposition takes place between 1 equiv. of metaphosphate of soda and 1 equiv. of nitrate of silver, and the liquid is neutral after precipitation: Na0,P0s+Ag0,N05=Ag0,P05+Na0,N05. The metaphosphates are also distinguished from the ordinary phosphates and pyrophosphates, by coagulating the white of egg, while the latter do not produce this effect. § 482. These three series of phosphates, which we have just described under the phosphates of soda, are also formed with the majority of other bases ; but they have not hitherto been suffi- ciently studied. They are even found in the phosphates of water, or hydrated phosphoric acids. It was stated (§ 211) that phosphorio acid afforded three definite hydrates : the hydrates 3IiO,POs and 2HO,POs, which have been obtained in a crystalline form, and the hydrate II0,P05, which remains when either of the preceding hy- drates is heated to redness. The hydrate 3IIO,P05, or tribasio phosphate of water, is most easily prepared, being formed in solu- tion whenever phosphorus is oxidized in the humid way, such as by treating it with nitric acid. This hydrate, saturated with bases, affords only tribasic phosphates. The monohydrate HO,POs or monobasic phosphate of water, gives with bases principally mono- basic phosphates or metaphosphates; but a certain quantity of bibasic, and even of tribasic phosphate, is nearly always found in the liquid. The crystallized bihydrate 2HO,POs or bibasic phosphate of water, furnishes a mixture of bibasic and tribasic phosphate. This circumstance is owing to the fact that the monobasic and bi- basic phosphates of water are much less fixed than the correspond- ing phosphates of soda, so that when dissolved in water, they are readily converted into the tribasic acid. The majority of chemists* consider these three series of phos- phates as formed by three different modifications, or three isomeric states of phosphoric acid, in which the acid presents the same elementary composition, but has different properties. The most striking differences, or rather those which are best ascertained, are * To Mr. Graham, an English chemist, is due the thorough investigation of the various phosphates ; previous to which it was impossible to explain the various and anomalous properties observed in the phosphates of soda. SODIUM. 485 that in the three states the power of saturation of phosphoric acid differs as regards the bases. Thus, the tribasic phosphates contain an acid, called ordinary phosphoric acid, possessing the property of combining with 3 equivs. of base. The bibasic phosphates, or py- rophosphates, contain an acid isomeric with the former, and called pyrophosphoric, but it combines with only 2 equivs. of base. Lastly, in the monobasic or metaphosphates, a third isomeric state of phosphoric acid is found, metaphosphoric acid, which combines with only 1 equiv. of base. § 483. These peculiarities of phosphoric acid can, however, be explained, without supposing the acid to exist in three different isomeric states, by admitting a principle in perfect conformity with the laws of mechanical equilibrium, and which appears to be established by a great number of experiments. I shall call it the principle of preservation of molecular grouping. The molecules of ordinary phosphoric acid and of the tribasic phosphates consist of groups formed of 1 equiv. of phosphoric acid and 3 equivs. of the base. These groups, or molecular systems, possess a certain degree of stability, and a great tendency to re- main permanent. In all double decompositions, the three equiva- lents of the original base are totally or partially replaced by equivalent quantities of other bases, but the tribasic grouping is always maintained. Nevertheless, if the tribasic phosphate con- tains volatile bases, as water or ammonia, heat may drive them off; the tribasic grouping will be destroyed, and a bibasic or monobasic grouping be formed. If only one equivalent of the base be disen- gaged, a bibasic grouping will be formed, and this grouping, once formed, will have a tendency to permanence equal to that of the tribasic grouping. In reactions effected in water, the grouping is only modified by double decomposition, by means of substitution. Its basic equivalents are replaced by equivalents of other bases, but the bibasic grouping is preserved. If the tribasic phosphate contain 2 equivs. of a volatile base, then 2 equivs. may be disengaged by heat, and a third grouping, the monobasic, be formed, which has an equal tendency to perma- nence, and is only modified by substitution when subjected to feeble reactions, like those of double decomposition effected in the humid way. Thus, while the three groupings will remain permanent in all fee- ble reactions, they may pass from one to the other in powerful reac- tions. As it was shown that heat could transform the tribasic into the bibasic and monobasic groups, so the inverse is practicable, and the latter groups may be transformed into the tribasic by merely fusing them with an excess of a powerful base, such as potassa or soda. Influenced by these bases, which have a great affinity for acids, phosphoric acid combines with the greatest quantity of base possible, and constitutes the tribasic group. The transformation 486 ALKALINE METALS. does not ensue, however, in the humid way, at ordinary tempera- tures, because it is opposed by the tendency of the molecular sys- tem to stability, which requires a high temperature to destroy it. This tendency of molecular grouping to permanence appears to play an important part in chemical phenomena, and explains at once the different actions frequently observed between anhydrous and hydrated acids. Hydrated acids are ready-formed saline groupings, which, when brought into contact with bases, merely substitute a more powerful base for their water. Anhydrous acids, on the contrary, present entirely different groups ; and frequently only combine with bases by transforming their original grouping into a saline grouping, through the intervention of a high tempera- ture, so that at ordinary temperatures they do not behave like true acids, which do not combine directly with bases. So also hydrated bases, such as the hydrate of potassa, present a saline group, in which the water plays the part of an acid. Such bases will immediately form salts, even with an anhydrous acid, because the saline grouping is already 'perfectly established, and the acid merely takes the place of the water. Chlorate of Soda. § 484. The chlorate of soda may be prepared, like the chlorate of potassa, by the reaction of chlorine on a concentrated solution of soda. The liquid contains, after the operation, chlorate of soda and chloride of sodium, but they are separated with difficulty from each other by crystallization. Chlorate of soda is easily obtained pure by decomposing a concentrated hot solution of chlorate of potassa by a solution of bitartrate of soda. The bitartrate of po- tassa, being slightly soluble, almost immediately separates from the liquid, and the solution contains chlorate of soda, which can be crystallized by evaporation. § 485. The neutral borate of soda is called, in the arts, borax. It is found, in commerce, in two states, common borax and octohe- dral borax, which differ from each other only in the proportion of their water of crystallization. The formula of ordinary borax is NaO,2B03-flOIIO, containing 42.7 per cent, of water; that of octohedral borax is Na0,2B03-f 5IIO, containing only 80.8 per cent, of water. Common borax effloresces slightly in the air; dissolves in 2 parts of boiling and 12 parts of cold water, and the solution has a strong alkaline reaction. When heated, it first fuses in its water of crys- tallization, then swells up, and becomes a spongy mass of anhydrous borax. At a more elevated temperature, the anhydrous salt un- dergoes igneous fusion, producing a viscous liquid, which does not Borates of Soda. 487 crystallize on solidifying, but presents a vitreous appearance. Fused borax is so tenacious that it may be drawn out into long delicate threads. When hot, the glass of borax dissolves the ma- jority of the metallic oxides, assuming peculiar colours, which ena- bles us to distinguish the various metals from each other. This property is very valuable in qualitative analysis, and may be proved on very small quantities of matter. For this purpose, a platinum wire (fig. 340) is generally employed, one end of it being twisted into a ring. The moistened ring being dipped into powdered anhydrous borax, a few particles of the metal to be examined are added to it. The mixture is then fused by the blowpipe (fig. 341) into a globule or bead, in which the metallic oxide is dissolved, and which, on cooling, presents the peculiar colour characteristic of the metallic oxide.* The flame used to effect the fusion may be that of an alcohol or oil lamp, or of a wax candle. When the metal has several degrees of oxidation, it frequently happens that it imparts two very dif- ferent colours to the borax; and as these colours may be produced at pleasure, they are both used to ascertain the nature of the metal. In the brilliant part b of the flame (fig. 342) which is imme- diately beyond the interior dark cone aa' the gas acts reducingly, because its combustible parts are not yet entirely burned, especially if the aperture of the blowpipe be very small. At whence PX 2.18 —px 3.869 X 0.405 Instead of weighing the precipitate of sulphate of baryta which the solution of the saline mixture yields with an excess of chloride of barium, the volume of a solution of chloride of barium may be determined, which exactly precipitates the sulphuric acid from the sulphates to be analyzed. To effect this, a solution of chloride of barium is prepared, so that a volume of 50 cubic centimetres will exactly precipitate 5 gm. of real sulphuric acid. It is evident that the number of cubic centimetres necessary to produce the complete precipitation will represent the number of decigrammes of sulphuric acid which existed in the solution subjected to analysis. The principal objection to this process is that the sulphate of baryta is not deposited rapidly in a cold liquid, and it is necessary to filter, from time to time, a small quantity of the liquid, in order to ascertain that no more sulphuric acid remains to be precipitated. § 526. The proportions of two chlorides of sodium and potassium mixed together may be determined in the same way. A given weight of the mixture is dissolved in water, and the chlorine pre- cipitated by nitrate of silver. The weight of chlorine combined with the two bases is calculated from that of the chloride of silver obtained, and the proportions of the two chlorides determined by a calculation similar to that just made of the sulphates. A standard solution of nitrate of silver can also be employed, and the exact volume measured which is required to precipitate the whole of the chlorine contained in a given weight of the substance. These methods of analysis afford considerable accuracy when the two bases have very different equivalents and exist in nearly equal proportions in the mixture. But the result would be very uncertain AMMONIA. 525 if the numerical value of the equivalents differed but slightly, or if one of the bases predominated much over the other. § 527. When the two alkaline metals are in the state of chlorides, a given weight of the mixture is dissolved in a small quantity of water, and a concentrated solution of perchloride of platinum poured into it, until the liquid assumes a very decided yellow colour. Chloride of potassium is precipitated in the state of a double chloride of potassium and platinum, but as a small quantity of it remains in solution, it may be easily separated by evaporating the liquid to dryness and treating it with alcohol, which dissolves the double chloride of platinum and sodium, and leaves all the double chloride of platinum and potassium. The precipitate is col- lected on a filter, washed with alcohol, and weighed after desicca- tion. The composition of the double chloride being known, the quantity of chloride of potassium it contains can be immediately deduced from it. The two processes just described are those generally used in scientific inquiries to determine the proportions of potassa and soda in a mixture. They are, however, too delicate to be applied to the arts. We shall mention some practical methods, which may be of service to manufacturers in special cases. § 528. Chloride of potassium is used by the makers of saltpetre; but the commercial chloride is always mixed with chloride of sodium, which is valueless in the fabrication of saltpetre. The proportions of the two chlorides may be ascertained in a very simple way, and with sufficient exactness for all commercial pur- poses. This method is founded on the very unequal decrease of tem- perature produced by the chlorides of potassium and sodium on the water in which they are dissolved. We have seen (§373) that 50 gm. of chloride of potassium, dissolving in 200 gm. of water, produce a decrease of temperature of 24.5°, while 50 gm. of chloride of sodium only produce a decrease of 3.4°. 50 gm. of the mixture is put into a bottle containing 200 cubic centimetres of water at the surrounding temperature, which is exactly measured by a delicate thermometer plunged into the liquid. In order to hasten the solution, it is stirred with the thermometer, and, after complete solution, the temperature is again observed. Let us sup- pose that it indicates a depression of temperature of t°, produced by the act of dissolving. This datum alone will allow us to calcu- late the proportion of the two chlorides, if no other salt be present in the mixture. If x be the number of grammes of chloride of po- tassium in the 50 gm. of the mixture, the decrease of temperature 9 produced by the x gm. of chloride of potassium, by dissolving in 200 cubic centimetres of water, will be given by the proportion 50 : 24.5° :: x : e, whence e= • x- 526 ALKALINE METALS. So again, the decrease of temperature 0' produced by the solu- tion of the (50—a;) gm. of chloride of sodium, by dissolving in 200 cubic centimetres of water, will be given by the proportion 50 : 3.4° : : 50—x : 0', whence (50 =2;). The decrease of temperature produced by the 50 gm. of the mixture will therefore be expressed by (50-*). But the decrease of temperature observed is t; we have, there- fore, z+^f.(50—x)=t, whence 50 (t—3.4) 21.1 » § 529. When the two salts are in the state of sulphates, we can determine pretty exactly their respective proportions, by a process founded on the increased density occasioned by sulphate of soda dissolving in a saturated solution of pure sulphate of potassa. The increase is moreover the more sensible, as the solubility of sulphate of potassa is remarkably increased by the presence of sulphate of soda. Let us take 50 gm. of a mixture of known proportions of sul- phate of soda and sulphate of potassa, and treat them with 300 cubic centimetres of a solution of sulphate of potassa saturated at a con- stant temperature of 68°. This quantity of water would be suffi- cient to dissolve entirely the sulphate of soda of the mixture, even if the latter were wholly composed of it. If no residue remains, which only happens in mixtures very poor in sulphate of potassa, we will add an excess of this salt, in order that the liquid may be saturated with it. A hydrometer is plunged into the liquid. It is evident that the instrument can be graduated so that its degrees shall mark precisely the percentage of soda existing in the mixture. Thus, when the hydrometer is plunged into a solution of pure sulphate of potassa, 0° is marked at its level. When plunged into a solution obtained by digesting a mixture of 50 gm. of dry sul- phate of soda and an excess of sulphate of potassa with a saturated solution of sulphate of potassa at 68°, the number of degrees is marked at this level equal to the percentage of soda existing in the dry sulphate of soda. Lastly, some intermediate points of the scale are determined in the same way. The instrument, thus graduated, is called a natro- meter, and may be used to determine the proportion of soda con- tained in any saline mixture composed of potassa and soda alone, provided that the two bases are combined with an acid which can be easily driven off by sulphuric acid. To do this, 50 gm. of the mixture are put into a porcelain capsule, decomposed by sulphuric acid, and evaporated to dryness to drive off the other volatile acids. AMMONIA. 527 It is treated with a small quantity of hot water, and the excess of acid neutralized with carbonate of potassa. The liquid being cooled to 68°, when a great deal of sulphate of potassa usually separates, is filtered into a test-glass which has been marked at a point cor- responding to 300 cubic centimetres by volume. The precipitated sulphate of potassa is washed with a solution of the same sulphate saturated at 68° until the level of the liquid reaches the mark on the test-glass. The natrometer being plunged into the liquid, the number of degrees indicated is observed, and is equal to the num- ber of hundredths (per cent.) of soda contained in the mixture. In the experiment just described, the same temperature is em- ployed : the instrument may be graduated so as to determine the proportion of soda at any temperature. In that case, two scales are marked on the stem of the hydrometer; one indicating, for each degree of the centigrade thermometer, the level of a saturated solu- tion of pure sulphate of potassa, which may be called the scale of temperature ; the divisions of the second, representing the hun- dredths of soda, may be termed the soda-scale. The zeros of the two scales coincide, so that, if operating at 32°, the soda will be directly determined by the soda-scale. But if operating at 77°, the instrument is plunged into a solution of pure sulphate of po- tassa saturated at that degree to a level that would indicate 8 hun- dredths of soda. At this point, therefore, the zero of the soda- scale should commence for this temperature. Experience shows that the divisions of the soda-scale are the smaller as they correspond to a greater proportion of soda; while the divisions of the scale of temperature which mark the densities of the solution of sulphate of potassa saturated at different tem- peratures are remarkably equal. When a test is made with the natrometer at a temperature t, this number t, representing the tem- perature of the liquid, is subtracted from the observed level m on the scale of temperature, and the number of divisions n noted on the soda-scale corresponding to the number (m—t) of divisions on the scale of temperature. This number expresses the hundredths of soda with sufficient accuracy for all commercial purposes. 528 II. ALKALINO-EARTHY METALS. BARIUM. Equivalent = 68.64 (858, 0 = 100). § 530. Barium* may be obtained by decomposing its protoxide by the galvanic battery. Some mercury being placed in a platinum capsule communicating with the negative pole of a battery, a solu- tion of baryta mixed with crystals of hydrated baryta is poured upon it, and the positive pole plunged into the paste. The decomposi- tion of the baryta and water are simultaneous; and as barium is set free, it combines with the mercury, which soon loses its fluidity. When the mercury is charged with an appreciable quantity of barium, it is removed, dried rapidly, and distilled in a glass retort, through which a current of nitrogen or hydrogen is passed to pre- vent all oxidizing action. The mercury volatilizes, leaving barium in the form of a metallic globule, if the heat has been carried to redness; but as barium attacks glass at this temperature, it is better not to raise the heat so high. Barium may also be obtained by decomposing anhydrous baryta by the vapour of potassium at a red-heat. For this purpose, an iron tube is used, open at both ends, in the middle of which is placed a platinum cup containing the anhydrous baryta, and, at a certain distance from one end of it, some pieces of potassium. A current of hydrogen gas being passed in through the same end, that part containing the baryta is highly heated, and communicates its heat to the potassium, which is vapourized. The vapour of potassium decomposes the baryta, oxide of potassium being formed and ba- rium set free. The tube is allowed to cool perfectly in the current of hydrogen gas, the cup removed, and the substance treated by mercury, which dissolves the barium. The amalgam, distilled in a current of hydrogen gas, leaves metallic barium. Barium exhibits the colour and lustre of silver; possesses a cer- tain degree of malleability; melts at a red-heat, but is not suffi- ciently volatile to be distilled; and is heavier than oil of vitriol, for a globule of the metal sinks to the bottom of the acid. Barium has a powerful affinity for oxygen, so that it oxidizes rapidly in the air and decomposes -water immediately in the cold. The great density of the compounds of barium distinguishes them from the compounds of the alkaline, alkalino-earthy, and earthy * Baryta was discovered in 1774, by Scheele. Davy isolated barium in 1807, by decomposing baryta by the galvanic battery. He obtained strontium and cal- cium in the same way. BARIUM. 529 metals ; and it was from this property the metal received its name, (from /3avps, heavy.) COMPOUNDS OF BARIUM AND OXYGEN § 531. Barium forms two compounds with oxygen : the protoxide BaO, or baryta, and the binoxide Ba02. Protoxide of Barium and Baryta.—Two insoluble salts of baryta are found in nature; the carbonate and the sulphate, from each of which baryta may be obtained. The carbonate calcined in a strong forge-fire loses all its carbonic acid, and baryta remains; but a lower temperature will suffice if the carbonate be previously mixed with charcoal, because the carbon has a tendency to abstract a portion of oxygen from the carbonic acid. Carbonic oxide is disengaged, and the baryta is mixed with charcoal; which is not objectionable, if the base is to be dissolved in water. Baryta is generally made by dissolving the carbonate in nitric acid, and evaporating the liquid to form the nitrate of baryta in anhydrous crystals. The nitrate is put into a porcelain retort (fig. 365), the mouth of which is closed with a bored cork, and the retort heated gradually in a reverberatory furnace until no more gas is evolved. Baryta remains in the form of a grayish-white porous mass which appears to have been fused; but baryta itself is infusible at a furnace-heat, and it is the nitrate which fused on the first im- pression of heat: as it decomposed, its fluidity diminished and the substance be- came doughy, until it at last remained puffed up by the bubbles of gas which traversed it. Anhydrous baryta fuses only at the highest temperatures, such as are produced by the oxy- hydrogen blowpipe. § 532. In order to obtain baryta from the natural sulphate, it is first transformed into a sulphide by calcination with charcoal. The sulphate, finely powdered, is mixed with one-tenth of its weight of charcoal, and sufficient oil added to form a consistent paste, which is heated to redness in a clay crucible. The oil is intended to bring every particle of sulphate in contact with the charcoal, and is de- composed by heat, leaving a residue of carbon intimately mixed with the substance. Organic matters may be substituted for the charcoal and oil, such as sugar, starch, and resin, for they leave a copious residue of carbon when decomposed by heat, and, moreover, melt before de- composition. The calcined matter is treated with boiling water, Fig. 365. 530 ALKALINO-EARTHY METALS. which dissolves the sulphide of barium. Nitric acid, gradually poured into the filtered liquid, converts the sulphide of barium into nitrate of baryta with disengagement of sulphohydric acid. The nitrate of baryta obtained by evaporating the liquid yields by cal- cination caustic anhydrous baryta, the density of which is about four times that of water. § 533. Baryta has a great affinity for water, for upon pouring a small quantity of water upon it, there is a considerable elevation of temperature, and a portion of the water is disengaged as steam. The baryta is converted into a hydrate, which falls to dust if the quantity of water added be not too great. When once combined with water, it can no longer be restored to the anhydrous state by heat alone. Hydrate of baryta is frequently used in the laboratory, and may be prepared by treating anhydrous baryta with water; or it may also be obtained immediately from the solution of sulphide of ba- rium above mentioned by merely boiling it with oxide of copper. Copper seizes on the sulphur, forming an insoluble sulphide of copper, and hydrate of baryta remains in the liquid: BaS + CuO=BaO + CuS. It is easy to ascertain the moment at which the sulphide of ba- rium is entirely changed into oxide, by pouring a small quantity of the liquid into a test-glass and adding a solution of acetate of lead. If no more sulphide remains, a white precipitate of the hy- drated protoxide of lead is formed ; but if any sulphide remain, the precipitate is more or less dark, from the admixture of black sul- phide of lead with the white hydrate. If the solution of barium subjected to the experiment be concentrated, it is sufficient to allow the liquid to cool, when a large portion of the hydrate of baryta crystallizes; but if it be dilute, it must be rapidly concentrated by heat. s Hydrate of baryta crystallizes in the form of laminae, or in large prismatic crystals if the crystallization is slow. It contains 10 equivalents of water, so that its formula is BaO + lOHO. The crystals, when heated, readily part with 9 equivalents of water, and are restored to the state of a monohydrate BaO + HO, which is no longer decomposed by heat. The monohydrate melts at a red-heat, and is not sensibly volatile. It dissolves in 2 parts of boiling, and 20 parts of cold water. Its solution is strongly alka- line ; and it quickly attracts the carbonic acid of the air, becoming cloudy from the formation of insoluble carbonate of baryta. The hydrate of baryta and all the soluble compounds of barium are energetic poisons. § 534. The composition of baryta, BaO, is deduced from the analysis of the chloride BaCl, which, when crystallized, contains water in combination, but soon loses it by the action of heat. Ten BARIUM, 531 grammes of anhydrous chloride being dissolved in water and boiled, nitrate of silver in excess is poured in, when an insoluble chloride of silver is precipitated, collected, and weighed after de- siccation. The weight of the chloride of silver will be 13.773 gm., containing 3.406 gm. of chlorine, so that in 10 gm. of chloride of barium, there are Chlorine 3.406 gm. Barium 6.594 10.000 In order to find the quantity of oxygen which forms baryta with the same quantity of barium, it is sufficient to make the proportion 35.5 : 8 :: 3.406 gm. : x, whence x =0.768. Thus, baryta is formed of Barium 6.594 gm. Oxygen 0.768 Baryta 7.362 The equivalent of barium will be given by the proportions, 0.768 : 6.594 :: 8 : x 1 , or 3.406 : 6.594 :: 35.5 : x }whence *-<38.64 Thus, the oxide of barium is composed of 1 eq. of barium 68.64 ... 89.57 1 “ oxygen 8.00 ... 10.43 1 “ baryta 76.64 ... 100.00 and the chloride of barium of 1 eq. of barium 68.64 ... 65.94 1 “ chlorine 35.50 ... 34.06 1 “ chloride of barium 104.14 ... 100.00 The quantity of water existing in the hydrate is ascertained by the process described for hydrate of potassa (§ 435). §535. Binoxide of Barium.—The protoxide is converted into the binoxide when heated in a current of oxygen at a temperature of 550° to 750°. The baryta, broken into fragments, being put into a green glass retort, to the bottom of which the current of oxygen is passed, absorbs the latter without changing its form, its colour only becoming slightly more gray. The binoxide readily combines with water, forming a white hydrate slightly soluble in water. Boiled with water, the hydrated binoxide is decomposed, oxygen being evolved, and baryta dissolved. The binoxide is em- ployed in the preparation of oxygenated water (§ 89). 532 ALKALINO-EARTHY METALS. SALTS FORMED BY PROTOXIDE OF BARIUM OR BARYTA. Sulphate of Baryta. § 536. This salt is found crystallized in nature, forming con- siderable veins in the older rocks, and being remarkable among earthy minerals for its great weight, has been called by mineralo- gists heavy spar. Its density is 4.4. The sulphate is insoluble in water, and scarcely soluble even in water acidulated by nitric or ehlorohydric acid. It may therefore be readily obtained by double decomposition, by pouring a solution of an alkaline sulphate, or even sulphuric acid, into a solution of nitrate of baryta or of chlo- ride of barium. We have seen frequent use made of the insolu- bility of the sulphate of baryta to precipitate the sulphuric acid existing in a solution. To avoid the introduction of another acid into the liquid, the precipitation is effected by a solution of hydrate of baryta. In order that the precipitated sulphate may collect in the form of a heavy powder and readily sink to the bottom of the vessel, the liquid should first be heated to boiling; except where heat would decompose the acid to be isolated. Sulphate of baryta, on precipitating, generally carries with it a portion of the soluble salts existing in the solution, on which ac- count the precipitate requires careful washing. The alkaline nitrates particularly are carried down in quantities. Time should be allowed the precipitate to deposit, after which the clear liquid is decanted and the precipitate boiled with water acidulated by ehlorohydric acid. The sulphate of baryta dissolves in oil of vitriol, but is again deposited when the liquid is diluted. Sulphate of baryta contains, 1 eq. of baryta 76.64 ... 65.71 1 “ sulphuric acid 40.00 ... 34.29 1 “ sulphate of baryta 116.64 ...100.00 Nitrate of Baryta. § 537. We have seen (§ 531) how nitrate of baryta is obtained from the native carbonate and sulphate. The nitrate crystallizes in regular anhydrous octahedrons, which are soluble in 8 parts of cold and 3 parts of boiling water. It is much less soluble in an acid liquid, for upon pouring into its solution a large quantity of nitric acid, it precipitates in the form of a crystalline powder. § 538. Crystallized carbonate of baryta is found in nature, and called witherite by mineralogists. It is obtained by double decom- Carbonate of Baryta. BARIUM. 533 position, by pouring an alkaline carbonate into a solution of nitrate of baryta or of chloride of barium. The carbonate fuses at a white heat, and is then decomposed, parting with its carbonic acid. It is very slightly soluble in water, which dissolves scarcely of it; but rather more when it contains free carbonic acid. COMPOUNDS OF BARIUM WITH SULPHUR. § 539. We have already seen the mode of preparing monosul- phide of barium by calcining the sulphate with charcoal. The residue, treated with boiling water, yields a yellow liquid which deposits white laminated crystals of the monosulphide. The crys- tals produce a colourless solution in water, and the original yellow colour of the liquid is due to its always containing a small quantity of polysulphide of barium. Polysulphides of barium may be obtained by boiling the solution of the monosulphide with sulphur. If the sulphur is in great ex- cess, a pentasulphide BaSs is formed. . The polysulphides may likewise be obtained by heating to redness a mixture of baryta and sulphur. The monosulphide acts the part of a base with the sulphides, furnishing a great number of sulphosalts. COMPOUND OF BARIUM WITH CHLORINE. § 540. Only one compound of barium with chlorine is known, and is easily prepared by dissolving the native carbonate in chloro- hydric acid. It may also be obtained from the sulphate, by first converting it into a sulphide by calcination with charcoal, and then decomposing the solution of the sulphide by chlorohydric acid. The evaporated solution yields a crystallized chloride, with the for- mula BaCl+2HO. The salt readily parts with its water on the application of heat, and the anhydrous chloride fuses at a red- heat.* Chloride of barium dissolves in 2-3 parts of water at 61°, and in 1-3 at the boiling point. On a large scale, it is prepared by calcining, in a reverberatory furnace, powdered sulphate of baryta with half its weight of chlo- ride of calcium arising from the manufacture of ammonia. The * The chloride being frequently employed as an agent, it is important to obtain it in a pure state. The processes recommended for this purpose are to crystallize it repeatedly or to boil its solution with some carbonate of baryta, whereby iron is precipitated. Both of these methods are tedious and imperfect, and I have found it the shortest and best process to make a chloride from the sulphide, by using an ordinary muriatic acid, (not containing much sulphuric acid,) drawing off the liquids from the residue 2 or 3 times, evaporating to dryness, and fusing the dry chloride in a crucible. It may then be dissolved and crystallized. The fusion most effectually renders the iron insoluble.—J. C. B. 534 mass, when withdrawn from the furnace, is reduced to a fine pow- der and briskly stirred with cold water, which dissolves chloride of barium and leaves sulphate of lime. The liquid is rapidly drawn off and evaporated. It is essential that the operation should be done quickly and at a low temperature; for, otherwise, an inverse decomposition would take place, by the regeneration of chloride of calcium and sulphate of baryta, because the latter compound is the most insoluble of them all. We should probably obtain a larger product by adding charcoal to the mixture, so as to form, as in the fabrication of artificial soda, an insoluble oxysulphide of calcium (2CaS + CaO), which would allow the chloride of barium to be wholly separated. ALKALINO-EARTIIY METALS. DISTINCTIVE CHARACTERS OF THE SALTS OF BARYTA. § 541. The salts of baryta are not precipitated by ammonia, provided the latter base be perfectly pure, that is, free from car- bonate or sulphate. Carbonate of ammonia and the alkaline carbonates precipitate baryta in the state of an insoluble carbonate. Sulphuric acid and the sulphates yield, with solutions of the compounds of barium, a white precipitate entirely insoluble in water and in dilute chlorohydric and nitric acids. The latter generally enables us to ascertain the presence of a compound of barium in a solution. However, the salts of strontium and lead present a similar reaction; but the salts of strontium are distin- guished from those of baryta by different characters, to be soon de- scribed ; and as to the salts of lead, they are distinguished by being blackened by sulphuretted hydrogen, while those of barium remain uncoloured. 535 STRONTIUM. Equivalent = 44. § 542. Strontium* is very analogous to barium in all its com- binations, which are obtained by processes resembling those for obtaining the corresponding compounds of barium. Strontium, like barium, is found in nature in the state of a car- bonate and sulphate. The carbonate is called by mineralogists strontianite, from its having been found at Cape Strontian, in Scotland, whence was derived the name strontium, given to the metal. The sulphate is called celestine, and is found in several localities. The gypseous rocks of Montmartre contain flattened nodules composed of small crystals of sulphate of strontia, car- bonate of lime, and gypsum. Strontium is extracted from strontia, precisely as barium from baryta. COMPOUNDS OF STRONTIUM WITH OXYGEN. § 543. Strontium forms two compounds with oxygen, a protoxide SrO, and a binoxide Sr02. The 'protoxide of strontium, or strontia, being procured, like ba- ryta, from the native carbonate or sulphate, we shall not stop to consider it. The strontia prepared by the calcination of the ni- trate, appears in the form of porous masses, of a grayish white, resembling those of baryta. Strontia combines with water with the evolution of considerable heat. The hydrate dissolves in water, and as it is much more soluble in hot than in cold wTater, a greater part of it crystallizes on cool- ing. The formula of the crystallized hydrate is SraO-f 10HO. Sub- jected to heat it soon loses 9 equivalents of water, but retains the last equivalent, even at the highest temperature of our furnaces. The binoxide, obtained by pouring oxygenated water into a solu- tion of hydrate of strontia, is deposited in the form of small crys- talline spangles. SALTS FORMED BY THE PROTOXIDE OF STRONTIUM, OR STRONTIA. Nitrate of Strontia. § 544. The nitrate of strontia, like that of baryta, is prepared from the native carbonate or sulphate, and crystallizes at ordinary temperatures in large regular octahedrons, which are anhydrous. If it be crystallized at a low temperature, it is deposited in another form and in a hydrated state, its formula being SrO,NOs+5HO. Nitrate of strontia is used by the makers of fireworks, from its communicating a beautiful crimson colour to the flame of burning * Strontia was discovered in 1793, by Klaproth and Hope. 536 ALKALINO-EARTIIY METALS. substances, a property possessed by all the compounds of strontia. Red fire is made by burning a mixture of 40 parts of nitrate of strontia, 13 of flowers of sulphur, 10 of chlorate of potassa, and 4 of oxysulphide of antimony. Carbonate of Strontia. § 545. Carbonate of strontia is found in nature, and is easily obtained by double decomposition, by pouring a solution of an alka- line carbonate into a solution of nitrate of strontia. Carbonate of strontia is wholly decomposed at a white-heat, and does not fuse, like carbonate of baryta, before decomposing. Sulphate of Strontia. § 546. The sulphate is the most common state in which stron- tium is found; it may be obtained by double decomposition, by pouring a solution of an alkaline sulphate into a solution of nitrate of strontia. It is very slightly soluble in water, but much less in- soluble than sulphate of baryta; for water in which the sulphate of strontia has been digested, is perceptibly clouded when mixed with a solution of a salt of baryta. COMPOUNDS OF STRONTIUM WITH SULPHUR. § 547. Strontium forms several sulphides, which correspond ex- actly to those of barium, and are obtained in the same way. The monosulphide is a powerful base, forming a great number of sul- phosalts. COMPOUND OF STRONTIUM WITH CHLORINE. § 548. The chloride is prepared by decomposing the native car- bonate, or the sulphide derived from the native sulphate, by chlo- rohydric acid. It is very soluble, and even deliquescent; and dissolves remarkably in concentrated alcohol, which does not dis- solve chloride of barium. Advantage is sometimes taken of this property to separate the two chlorides when mixed. The formula of crystallized chloride of strontium is SrCl+6HO. DISTINCTIVE CHARACTERS OF THE SALTS OF STRONTIA. § 549. The salts of strontia are not precipitated by pure ammo- nia. The alkaline carbonates precipitate the strontia in the state of a carbonate. Sulphuric acid and the sulphates produce in a solution of a com- pound of strontium a precipitate of sulphate of strontia, which resembles that produced by the compounds of barium. But the salts of strontium are easily distinguished from the salts of barium in not being precipitated by a solution of chromate of potassa, which gives a yellow precipitate with the compounds of barium. Silicofluohydric acid precipitates the salts of baryta, and not those of strontia. 537 CALCIUM. Equivalent = 20. § 550. Calcium is a metal very widely diffused throughout nature. Combined with oxygen and carbonic acid, it forms carbonato of protoxide of calcium, or carbonate of lime, which is found in immense strata in all sedimentary groups of strata. Sulphate of lime, called gypsum, or plaster, also forms considerable masses, in- terwoven with secondary and tertiary strata. Lastly, oxide of calcium, combined with silicic acid, enters into the composition of a great number of minerals which form the primary rocks. Lime likewise exists abundantly in organized bodies. The shells of the molluscm are composed of nearly pure carbonate of lime, and the bones of all animals contain a large proportion of phosphate and carbonate of lime. Calcium is extracted from lime, precisely as barium from baryta. It is a white brilliant metal, resembling silver; melts only at a high temperature; rapidly absorbs oxygen from the air, and is converted into an oxide; decomposes water energetically at ordi- nary temperatures, with the evolution of hydrogen, and is converted into hydrated lime. COMPOUNDS OF CALCIUM WITH OXYGEN. § 551. Two compounds of calcium with oxygen are known: a protoxide CaO, called lime, and a binoxide Ca02. Lime is of daily use, not only in the laboratory, but also in the arts, and is the essential principle of the mortar used in building. Lime is obtained by calcining the native carbonate of lime. When only a small quantity is required, Iceland spar, or white statuary marble is chosen, and calcined in a clay crucible in a strong forge-fire. If it be necessary to have the lime absolutely pure, it is preferable to dissolve the carbonate in nitric acid, which is digested cold with the powdered carbonate until effervescence ceases. By boiling the liquid for a short time with a little lime, the foreign metallic oxides are precipitated, such as alumina, or oxide of iron, if any he present. It is then evaporated to dryness, and the nitrate of lime which remains calcined to redness. Lime is a white amorphous substance, presenting the external form of the calcareous stone which produced it, with a density of about 2-3. It has a caustic taste, and blues the tincture of litmus reddened by an acid. It does not fuse at the highest temperature we have ever been able to produce in furnaces, but undergoes a sort of fusion in the hydroxygen blowpipe. 538 ALKALINO-EARTHY METALS. Lime combines with water, evolving a great deal of heat, so that a portion of the water escapes in the form of vapour, and the eleva- tion of temperature is frequently sufficient to inflame gunpowder. The maximum of temperature is produced by adding to lime about one-half of its weight of water. The operation by which water is combined with lime is called slaking, and hydrated lime is said to be slaked, to distinguish it from anhydrous lime, which is called quicklime. Lime, on hydrating, increases considerably in volume. If not too great a quantity of water be added, a monohydrate of lime CaO-fHO is formed, which assumes the form of a light, white, soft powder. By adding a greater quantity of water, a milky paste, called milk of lime, is produced. Water which has remained on lime contains a certain quantity of the base in solution, exerts a strongly alkaline reaction, and is called lime-water. The quantity dissolved is very small; for 1000 parts of water dissolve only 1 of lime. Lime-water, which is fre- quently used in the laboratory, is made by keeping a certain quan- tity of slaked lime in a well-corked bottle, filled with distilled water, and shaking it from time to time, in order to saturate the wa.ter. The hydrated lime in excess falls to the bottom, and the superna- tant liquid can be drawn off by a siphon. Lime-water rapidly attracts carbonic acid from the air, and a white pellicle of the car- bonate forms on the surface of the liquid. Lime-water, evaporated slowly under the receiver of the air-pump, deposits small crystals of hydrate of lime CaO,HO. Lime is less soluble in hot than in cold water; for lime-water, saturated cold, becomes cloudy when its temperature is raised. Quicklime, exposed to the air, attracts its water and carbonic acid, falls into dust, and no longer evolves heat when moistened with water. It is then said to fall in the air. There ensues, in this case, a definite combination of carbonate and hydrate of lime, CaO,C03-f CaO,HO ; but as the atmosphere always contains more vapour of water than carbonic acid, much more of the hydrate than of the carbonate is formed in the same time ; so that the preceding compound remains mixed with a considerable proportion of hydrate of lime. It is only after a long while, the absorption of the car- bonic acid continuing incessantly, that the mass approaches the definite composition of which we have given the formula. § 552. The composition of lime may be deduced from the analy- sis of chloride of calcium in the same way as that of baryta was deduced from the analysis of the chloride of barium (§ 532); but it may also be inferred from the analysis of carbonate of lime. For this purpose, a very pure native carbonate is selected, such as Iceland spar, broken into small fragments, and an exact weight P determined in a platinum crucible. The crucible, covered by its lid, is placed in a second crucible of clay, the cover of which is luted with clay, and which is heated for at least two hours in a CALCIUM. 539 strong forge-fire, in order to be sure that the decomposition of the carbonate of lime is completed. After cooling, the platinum cru- cible, with the quicklime it contains, is weighed. Let p be the weight of the lime; (P—p) will be that of the carbonic acid disen- gaged. Now, as carbonate of lime is formed of 1 equiv. of lime and 1 equiv. of carbonic acid = 22, the equivalent of lime is determined by the proportion, (P— p) : p : : 22 : x. Whence x is equal to 28. But, by hypothesis, 1 equiv. of lime is composed of 1 equiv. of calcium and 1 equiv. of oxygen *= 8. The equivalent of calcium is therefore 20. Consequently, the follow- ing is the composition of lime : 1 eq. calcium 20 ... 71.43 1 “ oxygen 8 ... 28.57 1 “ lime 28 ...lOChOO It is necessary to be certain that the carbonate of lime has been wholly converted into caustic lime by calcination. This is easy; for the lime should dissolve in acids, without disengaging carbonic acid: therefore, if a portion of undecomposed carbonate of lime remain, effervescence will take place during the solution of the substance in the acid. The foregoing analysis may be verified by converting the lime into a sulphate, which is effected by slaking the lime in a small quantity of water, and adding an excess of sulphuric acid to trans- form it into a sulphate. It is heated gently to drive off the water, and the crucible ignited to drive off the excess of sulphuric acid. Anhydrous sulphate of lime Ca0,S03 remains, which is weighed. Let Q be its weight; (Q —p) will be the weight of sulphuric acid which has combined with the weight p of lime, to form sulphate of lime. Now, the equivalent of sulphuric acid weighs 40, which gives the equivalent of lime, by making the proportion, (Q—p) : p : : 40 : x. The value of x, obtained by this proportion, ought to be the same as in the preceding proportion, based on the analysis of the carbonate of lime. § 553. Lime is used in making mortar, of which it is an essential ingredient, and is prepared on a large scale by calcining carbonate of lime, or limestone, in furnaces, technically called limekilns. Cal- careous rocks are rarely pure carbonate of lime, and almost always contain more or less magnesia, oxide of iron, quartz, clay, etc. The quality of the lime depends greatly on the degree of purity of the limestone which yields it and the nature of the foreign mat- ters it contains. When the limestone contains any considerable quantity of these substances, it yields a lime which differs greatly 540 ALKALINO-EARTHY METALS. from the pure lime described in (§ 549). Thus, with water, it evolves but little heat, does not swell much, nor form a soft paste. It is then said to be poor. The lime furnished by a limestone con- taining but a small quantity of foreign matter, nearly approaches, in its properties, lime chemically pure. It swells considerably when moistened, evolves a good deal of heat, and is then called fat lime. We shall see, when discussing the theory of mortars, that these two kinds of lime have special uses. For the present, we shall confine ourselves to the manufacture of fat lime. § 554. Limestone parts with its carbonic acid, at a much lower temperature in an open furnace than in a crucible, owing to the fact that gases are more readily disengaged from their combina- tions in an atmosphere composed of other gases. Thus, a hydrated salt loses readily, and often entirely, its water of hydration when it is kept at a certain temperature in a current of dry air: while it does not perceptibly lose it at the same temperature in an atmosphere of aqueous vapour. Carbonate of lime, calcined in a covered crucible, is constantly in an atmosphere of carbonic acid gas; while, in an open furnace, it is in an atmosphere in which the air, more or less vitiated by combustion, predominates greatly over the carbonic acid. Decomposition is necessarily more rapid in the latter than in the former case.* The various limestones are not decomposed with equal facility, even when composed of carbonate of lime in the same degree of purity, for the degree of cohesion of the stone exerts a powerful influence over the decomposition. Chalk, which is very feebly aggregated carbonate of lime, is much more easily decomposed than marble or Iceland spar, in which the carbonate of lime is aggregated by crystallization. Limekilns are either perpetual or draw kilns, or intermittent kilns. Fig. 366 represents an ordinary lime- kiln. It is about 3 metres (10 feet) in height, and built of common brick, lined with fire-brick. The kiln is generally erected against the escarpment of the lime- stone rock; and is frequently cut out of the rock itself, and merely lined with fire- brick. The kiln has one or more openings below, through which the lime is withdrawn when sufficiently burned. Over the grate on Avhich the fuel is burned, a sort of arch is constructed, with large pieces of limestone supporting the mass of calcareous stone Fig. 366. * Doubtless, the current of air passing through the furnace also facilitates the liberation of the carbonic acid.—J. C. B. CALCIUM. 541 which fills the kiln. The wood, fagots, brushwood, or peat on the grate being kindled, the fire is so regulated at first as simply to heat the whole mass. In 12 hours, the heat is increased, and the process continued until all the limestone is properly burned. It is then allowed to cool, and the lime is withdrawn. Lime-burning in perpetual kilns is much more profitable, as it economizes the heat, to a greater degree, and is exclusively used in localities where lime has a steady sale. It is effected in two ways : 1st. In kilns which are charged with alternate layers of lime- stone and pit-coal. As they gradually descend in the kiln, the lime is withdrawn by the lower openings as fast as it is burned, and replaced by fresh charges of limestone introduced by the upper opening. 2dly. In kilns which are filled entirely with limestone and heated by lateral fires. Figs. 367 and 368 represent one of these fur- naces ; fig. 367 being a vertical section through the axis of the kiln, and fig. 368 a horizontal section, made at the height of the lateral fires g, g', g", of fig. 367. The cavity of the kiln is about 8 or 10 metres (25-33 feet) in height, and lined with fire-brick. The kiln is built against a steep hillside, in order to afford greater facility for feeding it. Three openings o, o', o", slightly inclined outward, are made at the base of the kiln, and used for withdrawing the lime. Three other openings c, c', c", made at a distance of about 2 metres from the ground, commu- nicate with the grates g, g’, g", on which the fuel is burned. The air necessary for combustion penetrates by the side-openings e, and is regulated by dampers. The open- ings o, o', o" and i, i', i" are closed by sheet-iron doors. Working-arches B, in the body of the stack, allow the workmen to draw the lime from the openings for that purpose. Vertical chimneys l leading from the arches carry off the heated air while drawing, which would otherwise be insup- portable. When beginning a firing, the lower part of the kiln is filled writh fagots, as high as the grates g, g’, g" ; limestone is charged from above so as to fill the kiln, and the fagots are set on fire. The limestone immediately above the fire is soon burned, and falls down as the fuel is consumed. Fire is then kindled on the grates, Fig. 367. Fig. 368. 542 ALKALINO-EARTHY METALS. the fuel being peat, or pit-coal of an inferior quality. The heat evolved burns the limestone in the upper portions of the kiln. Every 12 hours, the lime is drawn from the bottom of the kiln, which is kept filled by charges of fresh limestone through the upper opening. This is continued until the kiln becomes so damaged as to be useless. § 555. Common quicklime is frequently used in the laboratory, but is generally mixed with the ashes of the fuel with which it was burned, and the lumps coated by alkaline chlorides and sul- phates, which might injuriously affect chemical investigations. These can be easily removed by slaking quicklime with a small quantity of water, so as to cause it to fall into a pulverulent hy- drate, putting the hydrate on a filter, and washing it with water, until the washings manifest no cloudiness with a solution of nitrate of silver. The hydrate is then calcined in a clay crucible. Quick- lime thus treated is very finely divided, and sufficiently pure for most of the uses of the laboratory. Binoxide of calcium Ca02 is obtained by pouring oxygenated water into lime-water, when the binoxide is deposited in the form of small crystalline plates. This compound is not very fixed, and readily parts with one-half of its oxygen when heated. SALTS FORMED BY PROTOXIDE OF CALCIUM, OR LIME. Sulphate of Lime. § 556. Sulphate of lime is found in nature in two states; in that of anhydrous sulphate of lime CaO,S03, termed by mineralo- gists anhydrite ; and in that of hydrated sulphate of lime CaO, SOs+2HO, called gypsum, plaster of Paris. These two minerals frequently form considerable lenticular masses in the strata of trias, where they are generally associated with rock-salt. Similar collections of gypsum are found in the inferior tertiary rocks. It is in this geological formation that the plaster in the environs of Paris is found, where it is interposed in strata of marl, above the waste limestone [calcaire grossier) which constitutes the building- stone of Paris. The gypsum belonging to this geological epoch is a fresh-water formation, as is proved by the fresh-water shells, the remains of which are found in the adjacent strata. The density of the anhydrous sulphate, or anhydrite, is 2.9. It forms compact masses, of a crystalline texture and some degree of hardness, and is sometimes found in well-formed crystals be- longing to the fourth system of crystallization. Anhydrous sul- phate of lime fuses at a red-heat, and, if allowed to cool slowly, assumes a crystalline texture, the cleavages of which lead to the form of the native crystal. The hydrated sulphate Ca0,S03+2H0 is sometimes found, in CALCIUM. 543 nature, in the state of well-terminated crystals, recognisable, as a mineral, by their want of hardness, so that they can be scratched with the nail. They are generally hemitropes, or twinned crystals, and their form, that of fig. 119, belongs to the fifth system of crystallization (§ 38). Similar, and frequently well-defined crys- tals, are deposited in the graduation-houses, on the twigs on which the "waters of salt-springs are concentrated. Another hemitropic form exhibits flattened lenticular masses, the outer faces of which are slightly curved. The masses are easily cleaved in a direction parallel to the two oblique axes, and the result of cleavage assumes the shape of a spear-head (fig. 369), from which it has been called spear-headed gypsum. It may be divided by a pen- knife into extremely thin laminge, perfectly transpa- rent and colourless, which break readily in the fingers, in two other directions of cleavage, giving them a rhomboidal shape. Crystals of hydrated sulphate of lime, differing from those of gypsum, are often formed in the boilers of high-pressure engines, in which water charged with plaster, and called selenitic water, is used. The formula of this hydrate is 2(Ca0,S03)-|-H0.* The crystals of gypsum are frequently irregularly interwoven with each other, sometimes forming white masses,- at others masses coloured by hydrated oxide of iron. Such gypsum constitutes alabaster, which is used for ornamental purposes, such as the manufacture of vases, clock- cases, etc. Common plaster is also composed of an aggregation of crystals of gypsum ; but foreign substances are generally mixed with it, such as carbonate of lime, clay, or sand. The plaster in the environs of Paris contains, Fig. 369. Sulphate of lime 70.39 Water 18.77 Carbonate of lime 7.63 Clay 3.21 100.00 § 557. Sulphate of lime is but slightly soluble in water; 1000 parts of the latter, at ordinary temperatures, dissolving about 2 parts of it. Its solubility diminishes with the temperature, so that a solution saturated in the cold is visibly clouded when heated to * Similar deposites are formed in boilers by sea-water and by waters which cannot be termed selenitic, because sulphate of lime is not their prevailing con- stituent. The deposite is mostly anhydrous, for its very small percentage of water is easily driven olf, and is not sufficient to make the formula given in the text. The deposite in the boilers of ocean-steamers usually contains sulphate Of magnesia, beside that of lime, or, where a high heat has been employed, caustic magnesia.—J. C. B. 544 4LKALIN0-EA11THY METALS. 212°. Its solubility even exhibits an anomaly similar to that of sulphate of soda, its greatest solubility corresponding to 95°. The same anomaly has been observed in the seleniate of soda, which is isomorphous with the sulphate of soda. The following numbers express the solubility of sulphate of lime at various temperatures: 100 parts of water at 32° dissolve 0.205 of sulphate of lime. “ 41 0.219 “ “ 53.6 0.233 “ “ 68 0.241 “ “ 86 0.249 “ “ 95 0.254 “ “ 105 0.252 “ “ 122 0.251 “ “ 148 0.248 “ “ 158 0.244 “ « 176 0.239 “ « 194 0.231 “ “ 212 0.217 “ A solution of the sulphate, evaporated slowly, deposits small brilliant crystals, presenting the same form as the native hydrated sulphate. § 558. Gypsum, heated to 248° or 266°, parts wholly with its water, and is converted into anhydrous sulphate of lime; but, in this state, it soon recovers the water it has lost, and becomes per- ceptibly warm. This latter property, however, is observed only when the gypsum has not been too highly heated, for if the tem- perature be raised to only 320°, the anhydrous substance recovers its water very slowly. Native anhydrous sulphate of lime, or an- hydrite, does not combine with water, and behaves like gypsum which has been calcined to redness. Sulphate of lime fuses at a red-heat, and solidifies, on cooling, into a crystalline mass, the cleavages of which are the same as those of anhydrite. The use of plaster in building-mortar and in making casts is founded on its property of parting with its water of crystallization at a low temperature, and of recovering it when again mixed with this liquid. By mixing finely powdered dehydrated plaster with ■water, a liquid paste is formed, in which, at first, the particles of anhydrous sulphate of lime are mechanically mixed with the water, but it soon combines with the water and is changed into a hydrated sulphate. A portion of the water disappears in the combination; and the particles which were disaggregated in the liquid paste, aggregate in small crystals at the moment of combining with the water. These small crystals dovetail, as it were, with each other, and the whole substance becomes a solid mass. A paste of plaster, poured into a mould, fills accurately all its cavities, but soon so- lidifies into a single compact mass, when it is said to set, in con- CALCIUM. 545 sequence of the combination of the anhydrous sulphate with water. If, after some time, the mould be removed, a piece of solid plaster can be taken from it, presenting in relief all the cavities of the mould. So, also, by spreading over a wall of rough stone a coat of boiled plaster mixed with water, so as to fill all the irregularities of the wall, a perfectly plane surface is obtained, on which all kinds of moulding can be fashioned, so long as the plaster does not set. A large quantity of plaster is thus used to cover the walls, par- titions, and ceilings of houses. § 559. Plaster for building is calcined in a heap, under a shed (fig. 370). A series of small arches are first made with large pieces of the plaster, on which the material to be calcined is heaped, placing the largest pieces at the bottom. Brush- wood or fagots are burned in the arches, and the flame per- meates the whole mass. The combustion is conducted slowly, in order that the temperature may not rise too high at the lower part, for it was stated that plaster, when too strongly burned, no longer sets with water. When the calcination is completed, which the work- man knows by the appearance of the material, he demolishes the heap, separates the pieces which appear to be too much calcined, or burned, and those which are not sufficiently calcined. The remainder, being reduced to a fine powder by stamping or grind- ing, and then sifted, is packed in small bags and sent to market. § 560. Plaster is usually employed to obtain impressions of ob- jects of which several copies are required; the impressions in basso, which serve as a mould, being used to obtain new impressions in relief. In order to mould a medal, it is surrounded with a border of pasteboard or wax. The medal is then coated with oil, to render the separation of the plaster more easy ; then painted with a brush dipped in a thin mixture of very fine plaster, so as to fill the most delicate cavities of the medal and prevent the admission of air be- tween the medal and the plaster. A thicker coat of plaster is then spread over the medal as high as the border. When the plaster has become solid, the medal is inverted, and a few gentle blows on its reverse will separate the plaster. In order to take the impression of a spherical embossed object, the mould must be made of several pieces which can be easily se- Fig. 370. 546 ALKALINO-EARTIiY METALS. parated. To give an idea of this process, let us suppose we are required to mould a hand. The hand, first covered with a very- thin coat of oil, is laid upon a napkin, and a strong silk thread ex- tended over it. With a brush, a thin coating of plaster is applied, which penetrates into all the folds of the skin ; and before it has time to set, a thicker coat is poured on, which is gradually- increased by successive additions of the same material, until it is several centimetres thick. After waiting a few moments until the plaster has assumed some consistency, the silk thread is raised by one end, thus dividing the plaster into two equal parts. After waiting a little longer, until it is still harder, the two halves of the plaster are separated, and the hand withdrawn. The two halves, again united, are lubricated with oil, and make a mould of tem- pered plaster, which will reproduce the hand as often as required. In a similar way, moulds are constructed for making statues and other ornamental objects; but the mould must be composed of a greater number of pieces, which are held together by an outer framework, or shell. The joints of the various pieces are repro- duced on the cast in the shape of small projecting threads, which are easily scraped off. Plaster intended for moulding delicate objects should be purer than that used in building: it should be carefully calcined, and not come in contact with the fuel. In Paris, the spear-headed gypsum, which forms the fine layers of the gypseous rocks of Mont- martre, is used for this purpose. This gypsum is broken into pieces about as large as a walnut, and calcined in ovens, the heat of which is most carefully regulated. § 561. Stucco, which is used to cover walls, and columns, and in making various ornamental objects in imitation of marble, is made by tempering the best plaster with a solution of gelatine or strong glue. The plaster, baked in an oven, is ground, sifted, and then tempered with a solution of strong glue; but it sets much more slowly than when tempered with pure water. For -white stucco, a colourless glue is employed, such as fish-glue; for coloured stucco, metallic oxides are added, such as the hydrated sesquioxide of iron, of manganese, copper, etc., hydrocarbonates of copper, etc. For marble stucco, different plasters are mixed, tempered with glue, and coloured with the various metallic oxides. The skilful workman makes the pattern at will, by properly regulating the mixture. The plaster, thus tempered, is applied in layers over the object to be covered. When it has become sufficiently hard, its moistened sur- face is rubbed with pumice-stone to render it perfectly smooth. A very thin coat of plaster, tempered with a stronger solution of gela- tine than that originally used, is then spread uniformly over it. When the surface is dry, it is polished with tripoli on a fine cloth. From time to time, its surface is moistened with olive-oil, and the polishing is continued until it is completed. CALCIUM. 547 § 562. Of late years, a plaster calcined with alum, called alumed plaster, has been used in the arts. It becomes harder than ordi- nary plaster, and is more beautiful, being less dead, and possessing a certain degree of translucency. In order to prepare it, the plas- ter is first calcined, to deprive it of its water of crystallization, and then immediately thrown into a water-bath saturated with alum. In six hours, it is withdrawn, and, after having been dried in the air, is again calcined at a dull red-heat, and then ground. This plaster may be used like common plaster, but is frequently tem- pered with a solution of alum instead of pure water. Alumed plaster does not set immediately, like ordinary plaster, but retains its softness for several hours. It may be advantageously sub- stituted for stucco. Mixed with an equal quantity of sand, it pro- duces a substance possessing extreme hardness, and fitted for making flag-stones. Carbonate of Lime. § 563. Carbonate of lime is one of the most extensively diffused substances on the surface of the globe. It is sometimes found in isolated and perfectly terminated crystals, when it assumes one of two incompatible forms, and is one of the first ascertained in- stances of dimorphism. Its most frequent form is that of a rhom- bohedron having an angle of 105° ; but a very considerable number of forms, derived from this, is met with, all presenting three very easy cleavages, which lead to the rhombohedron of 105°. Very large and perfectly transparent rhombohedral fragments of carbo- nate of lime are frequently found in Iceland, hence called Iceland spar, and are highly prized by opticians. The second dimorphic form of carbonate of lime is a right prism with a rectangular base, belonging to the fourth system of crystallization, and is called by mineralogists arragonite. Both forms of the carbonate may be arti- ficially obtained. If an alkaline carbonate be added to a cold solu- tion of a salt of lime, a copious precipitate is formed, which, after some time, becomes granular, and the microscope detects in it small rhombohedrons. If, on the contrary, a boiling solution of a salt of lime be poured into a hot solution of carbonate of ammonia, a dense powder is immediately obtained, in which the microscope shows small crystals of arragonite. If small pieces of arragonite be carefully heated, they soon separate suddenly and fall into powder, and if the temperature has not been raised to redness, the substance undergoes no change of composition, but presents the same weight as before calcination. The disaggregation has been produced solely by a change in the system of crystallization, and the microscope discovers the existence of small rhombohedral crystals in the disaggregated matter. A solution of sugar dissolves a large quantity of hydrated lime; and if the solution be exposed to the air, it absorbs carbonic acid, 548 ALKALINO-EARTHY METALS. and deposits carbonate of lime in the form of small rhombo- hedral crystals, perfectly transparent. If this experiment be made at a low temperature, crystals of hydrated carbonate of lime are deposited; but these crystals are soon converted into ordinary carbonate of lime, at a temperature above 32°. § 564. The waters of a great number of natural springs contain carbonate of lime, dissolved by the assistance of an excess of car- bonic acid. These waters, on reaching the air, soon part with their carbonic acid, and the carbonate of lime separates. Cal- careous incrustations are thus formed, which, after a time, become very large. The fountain of Saint-Allyre near Clermont, pro- duces these incrustations in a short space of time; for if an object be exposed for a few days to the falling water, it becomes covered with a calcareous crust. In this manner the calcareous stalactites and stalagmites are formed (fig. 371) which line the walls of cer- tain grottos. The water, traversing fissures in th** rocks, drops from the roof, but as each drop remaini* suspended for some time before falling, it parts with a portion of its carbonio acid, and, consequently, of its carbonate of lime. The same drop, falling on the floor, deposits another por- tion of calcareous carbo- nate. As the dropping continues in the same spot, a dependent calcareous in- crustation, or stalactite, is formed, which gradually in- creases toward the earth. Immediately beneath this suspended incrustation, a similar one, or a stalagmite, rises from the earth. These incrustations sometimes join each other, and form a con- tinuous column. The carbonate of lime is crystallized in these incrustations, as can be easily ascertained by their fracture. § 565. In saccharoidal marble, the carbonate of lime is likewise crystallized, but the crystals are strongly aggregated to each other. The various limestones formed in all the sedimentary rocks, constituting frequently strata of great thickness, exhibit carbonate of lime in very various degrees of compactness. The limestones of transition rocks are, in general, less compact. The majority of these limestones contain the impressions of shells, and some are wholly formed of them. Chalk is a calcareous rock but slightly aggregated, and belonging to the secondary formations. The shells of the molluscse, and of birds’ eggs, the carapaces Fig. 871. CALCIUM. 549 of the crustacese, are formed of nearly pure carbonate of lime, and the bones of man and animals also contain a considerable propor- tion of it. § 566. Carbonate of lime, subjected to heat, decomposes before fusing; but if it be heated in a gun-barrel hermetically sealed, the high pressure in the tube prevents the escape of carbonic acid, and the carbonate fuses without decomposing. If the barrel be allowed to cool slowly, the fused limestone assumes a crystalline texture, and then resembles exactly saccharoidal marble. Carbonate of lime does not appreciably dissolve in pure water, while water charged with carbonic acid dissolves it in considerable quantity. Nitrate of Lime. § 567. Nitrate of lime is obtained by dissolving carbonate of lime in nitric acid, and concentrating the liquid by heat, when it assumes, on cooling, a crystalline form. It is a deliquescent salt. Phosphates of Lime. § 568. The phosphates of lime corresponding to trihydrated or ordinary phosphoric acid, are those which are best known. When bone-ashes are treated with sulphuric acid, sulphate of lime is formed, and separates because it is but slightly soluble. The liquid contains a phosphate of lime, improperly called biphosphate, which separates in the form of crystalline spangles, if the liquid is sufficiently concentrated. The formula of the salt is (CaO-f 2HO)POs. It was stated (§ 205) that this product is used in the manufacture of phosphorus. If a solution of ordinary phosphate of soda (2Na0 + H0)P05+ 24HO be poured into a solution of a salt of lime, a white gelatinous precipitate is obtained, having the formula (2Ca0 + H0)P05+ 4HO. If this precipitate be digested with ammonia, it parts with a portion of its phosphoric acid, and a precipitate remains of which the formula is 3CaO,POs. The same phosphate 3Ca0,P05 is im- mediately precipitated when an excess of phosphoric acid is poured into a solution of chloride of calcium and the liquid supersaturated with ammonia. Bone-ashes are composed of f phosphate of lime, and \ carbon- ate of lime. The formula of the phosphate of lime of bones is 3CaO,POs. These various phosphates of lime, with the exception of the bi- phosphate, are insoluble in water, but readily soluble in an acid liquid. Phosphate of lime is found crystallized in nature in a mineral called apatite; the phosphate 3CaO,POs found in it being com- bined with a small quantity of chloride and fluoride of calcium. 550 ALKALINO-EARTHY METALS. If the biphosphate (CaO + 2HO)POs be heated to redness, it fuses into a substance which remains vitreous after cooling; and its nature is entirely changed, for it has become insoluble in water. Heat has caused the phosphate to pass from the tribasic to the monobasic modification, and the ignited product is metaphosphate of lime CaO,POs. Chlorate of Lime. § 569. This salt is obtained mixed with chloride of calcium by passing a current of chlorine through milk of lime. There are formed, at first, hypochlorite of lime and chloride of calcium; but, if the chlorine be continued, after the lime is entirely converted into these two products, a new reaction takes place, and chlorate of lime is formed, particularly if the temperature be raised. This liquid may be used for the manufacture of chlorate of potassa, by merely pouring chloride of potassium into it, when a double decom- position takes place, and chlorate of potassa is deposited. Hypochlorite of Lime. §570. This salt is very important on account of its application to bleaching. It is obtained in a state of purity, by adding a solu- tion of hypochlorous acid to milk of lime; but there must be an excess of lime, for as soon as hypochlorous acid predominates, the hypochlorite is decomposed into chlorate of lime and chloride of calcium : 3(Ca0,C10)=Ca0,C10s+2CaCl. A solution of hypochlorite of lime, at first blues the tincture of litmus reddened by an acid; but soon destroys its colour. Chlorine does not act on quicklime; but, if passed slowly over hydrated lime, hypochlorite of lime and chloride of calcium are formed: 2Ca0+2Cl=Ca0,C10 + CaCl. It is essential to leave always an excess of lime; for, if the cur- rent of chlorine be continued after the lime is completely converted into hypochlorite of lime and chloride of calcium, a new reaction ensues, by which the hypochlorite is converted into chlorate of lime and chloride of calcium. This reaction manifests itself, par- ticularly if the temperature be elevated, either from a too copious supply of chlorine or from the substance being too much heated. § 571. The name of chloride of lime is commercially given to a mixture of hypochlorite of lime, chloride of calcium, and hydrated lime, which is obtained by imperfectly saturating slaked lime by chlorine. It is manufactured in large quantities, for it is almost exclusively employed in bleaching. CALCIUM. 551 Chlorine is prepared by the reaction of a mixture of sea-salt, peroxide of manganese, and dilute sulphuric acid, at a gentle heat, in a leaden apparatus, composed of a still abed (fig. 372), enclosed in a casing of sheet iron. Steam is carried between the casing and still, through the pipe o, so as to main- tain the temperature at about 140°. The pipe s serves to withdraw the substances when all their chlorine is disengaged. The cap or top of the still has several tubulures; one at e, through which the materials are introduced ; one at/, for the escape of the gas ; another at g, to which is soldered a safety-tube, furnished with a funnel, through which the acid is poured gradually; and lastly, a tubulure A, traversed by an iron rod, covered with lead and ter- minating in a large paddle mn, also covered with lead, which is used to stir the mixture. The tubulures are constructed with small leaden grooves, into which sulphuric acid is poured in order to seal the joints hermetically. The arrangement is easily understood by reference to fig. 372. The chlorine is conveyed into large chambers of masonry, in which are arranged a great number of wooden shelves, covered by a layer of hydrate of lime about f inch thick. Where the chambers are very low, the hydrate is simply spread over the floor to the thickness of about 2 inches; in which case, it must be constantly stirred with rakes to renew the surface. When the lime has ab- sorbed a sufficient quantity of chlorine, it is withdrawn and packed in casks. By treating chloride of lime with water, the hypochlorite of lime and the chloride of calcium are dissolved, and the excess of hy- drated lime remains in the form of a pulp. The clear liquid may be separated by filtering or decanting. The chloride is often prepared, in the workshops in which it is to be used, in solution, by conveying chlorine into cylinders half filled with lime, which is constantly stirred in order to promote the ab- sorption of chlorine. Hypochlorite of lime is decomposed by the most feeble acids, even by carbonic, and from this property it always exhales the odour of hypochlorous acid, which is expelled by the carbonic acid of the air. An aqueous solution of chlorine exerts an oxidizing agency on all substances capable of higher oxidation, many examples of which have been already presented. By virtue of this oxidizing agency, a solution of chlorine destroys the colour of coloured organic bodies, acting in a similar manner to oxygenated water. Water is decomposed, its hydrogen combining with chlorine, and its oxygen, Fig. 372. 552 ALKALINO-EAliTIIY METALS. in the nascent state, oxidizing the organic matter, which is usually converted into a new colourless body. Moreover, it will be shown hereafter that organic substances, subjected to oxidizing agencies, are finally converted into acids, which may be easily removed by alkalies. The coloured organic substances printed on cloths or muslins by virtue of special affinity, and which, in this state, are in- soluble in water and alkaline lyes, are therefore converted, by an ox- idizing action, into other substances possessing acid properties, and ■which can be readily removed by alkaline lyes. It will be readily seen, that 1 equiv. of hypochlorous acid, or 1 equiv. of hypochlo- rite of lime, in the presence of an acid, must exert the same ox- idizing and decolourizing action as 2 equivs. of oxygen in the nascent state, or 2 equivs. of chlorine dissolved in water; for free hypochlorous acid is readily decomposed into 1 equiv. of chlorine and 1 equiv. of oxygen. Now, to obtain 1 equiv. of hypochlorite of lime, 2 equivs. of chlorine are required to act on 2 equivs. of hydrated lime; so that the liquid obtained by treating chloride of lime with water should exert the same bleaching power as the quantity of chlorine used to produce this chloride of lime. In order to bleach a piece of goods, it is first dipped into a weak solution of chlorohydric acid; then passed through a vat contain- ing chloride of lime ; and lastly, washed with some alkaline liquid. Chloride of lime is also used as a disinfecting agent, and to de- stroy disagreeable odours. The hypochlorous acid is gradually driven off by the carbonic acid of the air, and, like chlorine, it destroys the substances which evolve those odours. The best method of employing it consists in impregnating linen with a strong solution of the chloride, and hanging it up in the place where the air is to be purified. §572. Commercial chloride of lime necessarily presenting va- rious degrees of strength, it is important to the purchaser to be able to ascertain with ease and accuracy its bleaching power, as that alone gives it value. The determination is made by means of the clilorometric analysis about to be described with some minute- ness. In order to compare the merchantable values of the various qualities of chloride of lime found in commerce, the weights of the different chlorides are ascertained which will bleach the same volume of a standard solution of organic colouring matter. The values of the chlorides will be in the inverse ratio of these weights. The colouring matter chosen, is a solution of indigo in sulphuric acid. In order to prepare it, the indigo of commerce is treated with Nordhausen oil of vitriol or fuming sulphuric acid, which dis- solves a considerable quantity of it when it is diluted with water, and produces a deep-blue liquid. The bleaching of this liquid by chlorine is very marked, for the colour passes immediately from a deep-blue to a yellow. The solution of indigo is diluted with water, CALCIUM. 553 until 1 litre of it is exactly bleached by 1 litre of dry chlorine, at the temperature of 32°, and under a pressure of 0.76 m. (29.92 inches.) To effect this, a normal solution of chlorine is first prepared, by means of which the solution of indigo is rated. The normal solu- tion may be prepared in various ways, of which we shall describe the most simple. A ground-stoppered bottle is filled with dry chlorine (§ 168, fig. 223), and the temperature and barometric pres- sure noted at the same time. The inverted bottle is immersed in ' a dilute solution of potassa (fig. 373), the stopper withdrawn a very little, in order to allow a small quantity of the alkaline liquid to enter the bottle, and then replaced. After shaking the bottle without removing it from the solution, a vacuum is formed by the absorption of the chlorine; when the cork is again removed to allow the entrance of a small quantity of the alkaline solution. The bottle is again shaken, and the operation repeated until the absorption of the chlorine is completed. If the bottle contained originally nothing but chlorine, it is evident that it will be entirely filled with the solution of potassa ; but if it contained air mixed with the chlorine, the air will remain after the absorption of the chlorine. In all cases, the volume of alkaline liquid which has entered the bottle is exactly equal to the volume of chlorine absorbed. If, therefore, the chlorine, at the moment of closing the bottle, were under the normal conditions of temperature and pressure, that is to say, at 32°, and under a barometric pressure of 0.760 m., the solution of potassa would contain its normal volume of chlorine, and we will call its standard 100. But, if the surrounding temperature were t, at the moment of closing the bottle, and the pressure H, the so- lution of potassa would only contain a volume of chlorine repre- sented by iq.0.oo367.f ’ tto °f chlorine under normal conditions, and its standard is represented by 100. t+mo367I‘ 760 ’ The solution of indigo must now be diluted with water, so that 50 cub. centim. of the solution shall be exactly bleached by 501+00*367 ; * of the bleaching solution of potassa. To avoid repetition, a preliminary test of the solution of indigo is made, by taking 50 cub. centim. of the solution with the pipette D (fig. 374), which has a mark y at the level of 50 cub. centim., and pouring them into a glass placed on a sheet of paper. The chlorimeter (fig. 375) is filled up to 0 with the standard bleaching liquid, which is poured out slowly until the moment of discolouration. Let n be the number Fig. 373. Fig. 374. 554 ALKALINO-EA11THY METALS. of divisions poured out: it will represent \ cub. centim., as the clilo- rimeter is divided into cubic demicentimetres. The stan- dard of the solution of indigo is therefore n x--, 0q0367 t ■ and must be diluted with water so as to bring it to 100. Supposing that the preceding expression, calculated in numbers, gives 175, sufficient water is added to 100 parts of the solution of indigo to bring the volume to 175, which will make it a normal solution of indigo of the standard of 100. The standard should, however, be verified by an- other experiment, and corrected, if necessary. The nor- mal coloured liquid should then be preserved in a well- stoppered bottle. In order to test a bleaching chloride, small samples of the chlo- ride are taken from various parts of the mass to be tested, so as to make a sample which may be considered as representing the ave- rage of the whole. 10 gm. of this sample being rubbed in a porcelain or glass mortar with a small quantity of water, more water is added and decanted into a filter placed in the vessel A (fig. 376) of 1 litre content. This P- process is repeated several times, and lastly the vessel A is filled with water to the level of the mark a. Filtra- tion may be avoided by employing careful decantation. The liquid, being cleared by repose or filtration, is poured into the chlorimeter as far as the division 0. On the other hand, 50 cub. centim. of the normal solution of indigo, taken up with the pipette D, are poured into a vessel B (fig. 377) placed on a sheet of white paper; and while shaking this vessel with the left hand, the solution of the bleaching chloride is poured slowly into it. When approaching the moment of discoloration, the chloride should be added by drops. Supposing that it required 115 divisions of the chlorimeter to effect the discoloration, the standard of the chloride will be i52-=86.9°. If the inverse method of testing could be made, by pouring the normal solution of indigo, from the alkalimeter into a volume of 50 cubic centimetres of a solution of the bleaching chloride, until the latter assumed a blue colour, it is evident that the standard of the chloride would be immediately given by the number of cubic demicentimetres of the solution of indigo poured out. But this cannot be done, because the solution of indigo contains a large quantity of acid, and because the first portions poured into the bleaching liquid would disengage a quantity of chlorine greater than that necessary to decolorize the indigo with which they im- mediately come into contact. We would thus be liable to the loss of chlorine. But, when the chloride is poured into the solution of indigo, the chloride is always in the presence of an excess of indigo, and is not subject to loss. Fig. 375. Fig. 376. Fig. 377. CALCIUM. 555 § 5TB. The solution of indigo is liable to serious objection which has caused it to be abandoned. It soon changes, and may be the cause of error, when a solution is employed which has been made for some time. A standard solution of arsenious acid in chlorohydric acid is now substituted for the normal blue liquid: the chlorine set free converts the arsenious into arsenic acid, and it is easy to ascertain the moment at which the transformation occurs ; for experience shows that if a solution of arsenious acid be coloured by a few drops of a solution of indigo, chlorine does not bleach it until it has entirely converted the arsenious into arsenic acid. In order to prepare the normal arsenious solution, 4.439 gm. of pure arsenious acid are weighed out, dissolved in chlorohydric acid diluted with its volume of water, and water then added, so that it shall occupy the volume of 1 litre. If doubts exist as to the purity of the arsenious acid, the standard of the arsenious solution must be verified by means of the normal chlorine liquid previously mentioned. The first test should be regarded as only an approximation. A second is made, by pouring immediately into the 50 cub. cntim. of uncoloured arsenious solution, a volume of the solution of chloride, somewhat less than that which effected the decolorization in the first test. A few drops of the solution of indigo are then first added, to colour the liquid, after which the chloride is added, drop by drop, so that the moment of dis- coloration can be ascertained with precision. As there is some danger in filling the pipette D with the arsenious solution by sucking it with the mouth, it is better to fill it by immersion, as in fig. 378.* Fig. 378, * The method commonly practised in England and the United States for ascer- taining the strength of bleaching-salt is to dissolve a given weight of crystallized copperas, protosulphate of iron, in water, and add to it a solution of a given quantity of the bleaching-salt, from an alkalimeter, until all the protoxide is converted into peroxide of iron. A salt of iron fully peroxidized will not give Prussian blue with red prussiate of potash, (ferrid cyanide of potassium,) but the least trace of protoxide yields a blue colour instantly. To determine the com- plete peroxidation of the copperas by the bleaching-salt, many small drops of a solution of red prussiate are put upon a surface of porcelain, and a drop of the copperas solution touched to one of them by a glass rod. If the colour be blue, more chloride is added to the copperas solution, when a drop is again taken out and tried with another drop of the red prussiate. This operation is repeated until the yellow drop of red prussiate is no longer blued. A preliminary experi- ment followed by one more exact makes the whole test shorter than by employing a single test; and the approach of the point of peroxidation is recognised by greenish-blue, green, and light-green colours successively.—J. C. B. A still more exact method, for which less time is required than for either of those mentioned in the text and in the note by Prof. Booth, is the following:—A solution of a given quantity of the bleaching-powder is added to a measured solu- tion of pure protosulphate of iron, in such quantity that only a part of the prot- oxide of iron will be oxidized by the chlorine; the remaining protoxide is then very accurately determined with permanganate of potassa, according to the me- 556 ALKALINO-EARTHY METALS. COMPOUNDS OF CALCIUM WITH SULPHUR. § 574. Calcium forms a great number of compounds with sul- phur. By calcining sulphate of lime with charcoal, it is converted into monosulphide of calcium CaS, a white substance, nearly in- soluble in water. If milk of lime be boiled with the flowers of sulphur, more sulphuretted sulphides are obtained, which remain in solution with the hyposulphite of lime. If the sulphur is in great excess, and the ebullition prolonged, the protosulphide CaSs is obtained, which remains in the solution, and is not depo- sited by the cooling of the liquid. If it is boiled for a shorter time, and the hot liquid filtered, the yellow solution deposits, on cooling, orange-coloured circular crystals of the bisulphide CaS2, but little soluble in cold water, which dissolves only about of its weight. COMPOUND OF CALCIUM WITH CHLORINE. § 575. Only one compound of calcium with chlorine is known, and is prepared by dissolving hydrate or carbonate of lime in chlorohydric acid. Chloride of calcium is produced in large quan- tities in the preparation of ammonia, which is made by heating a mixture of chlorohydrate of ammonia and lime in large cast-iron cylinders (§ 123). The residue from the operation is chloride of calcium, mixed with a small quantity of lime in excess, from which the chloride is extracted by treatment with cold water. The liquid, highly concentrated by evaporation, and allowed to cool, deposits large crystals of the hydrated chloride, the formula of which is CaCl+6HO. These crystals are very deliquescent, and produce a great degree of cold by solution in water; but the greatest depression of temperature is obtained by mixing them with pounded ice, when, as was stated (§ 374), the temperature could thus be reduced to —49°. Hydrated chloride of calcium fuses readily in its water of crystallization, and, when heated to 400°, parts with 4 equivalents of water, leaving a porous mass, which absorbs water with great avidity, and is well fitted for dry- ing gases. Heated still further, it parts with the balance of its water, and then fuses at a red-heat. The fused chloride is gene- rally cast in the form of flat cakes, which are broken up and pre- served in a well-stoppered bottle. It is frequently used in the laboratory, either for drying gases or for removing the water which is mixed with liquids of organic origin. The anhydrous chloride dissolves in water with such an elevation tliod described in \ 804; the quantity of protoxide thus found, subtracted from that contained in the measured solution employed, gives the quantity oxidized by the chlorine, from -which the percentage of the latter is found by a simple and easy calculation.— W. L. Faber. CALCIUM. 557 of temperature that the heat evolved by its combination is greater than that which becomes latent by the act of solution of the hy- drated chloride. It dissolves in large proportion in water, and is also quite soluble in absolute alcohol. The alcoholic solution made by heat deposits, on cooling, crystals of a combination of the chlo- ride with alcohol, an alcoTiolate of chloride of calcium. If a concentrated solution of chloride of calcium be boiled with an excess of hydrated lime, a considerable proportion of the hy- drate is dissolved; and the filtered liquid deposits, on cooling, a crystallized compound of chloride of calcium and lime, the formula of which is CaCl+3CaO+15HO. COMPOUND OF CALCIUM WITH FLUORINE. § 576. Fluoride of calcium is found in nature, either in compact masses of various hues, or in well-defined crystals, which are cubes, sometimes modified by the facets of the octahedron. Mine- ralogists call it fluor-spar. It presents a remarkable phenomenon of phosphorescence. When its powder is heated in an iron spoon, it becomes luminous long before reaching a red-heat, and evolves a violet or green light, according to the specimen. It is used in the laboratory in the preparation of fluohydric acid (§ 204). DISTINCTIVE CHARACTERS OF THE SALTS OF LIME. § 577. The salts of lime are not precipitated by ammonia, which distinguishes them from the earthy metals, and from the second class of metals, properly so called (§ 276). They are precipitated by the alkaline carbonates, a character which distinguishes them from the salts furnished by the alkaline metals. If sulphuric acid or a sulphate be poured into a very dilute solution of a salt of lime, no precipitate is formed; in which case the salts of baryta and strontia would yield a precipitate. If the solution of the salt of lime is more concentrated, a precipitate of hydrated sulphate of lime is formed, which, if left to itself for some time, collects in the form of small crystalline spangles, easily recognisable by the microscope. Salts of lime yield, with oxalic acid and the oxalates, a granular precipitate of oxalate of lime, nearly insoluble in water, and soluble with great difficulty in an excess of acid. Advantage is taken of this property, not only to detect the presence of lime, but also in chemical analyses, to precipitate it from the liquids which con- tain it. 558 MAGNESIUM. Equivalent =12.1. § 578. Magnesium* is obtained by decomposing the anhydrous chloride of magnesium by potassium or sodium. A few globules of potassium or sodium are placed at the bottom of a platinum crucible, and above them the chloride of magnesium broken in pieces. The crucible is covered with its lid, which is fastened down by iron wire, and the temperature then raised by an alcohol lamp. Reaction takes place at a red-heat, with a violent deflagra- tion, which would throw off the lid of the crucible were it not firmly fixed. The potassium combines with the chlorine, and the magnesium is set free. The crucible being allowed to cool, the substance is treated with water as cold as possible, which dis- solves the chloride of potassium and the unaltered chloride of magnesium, leaving the magnesium in the form of metallic globules. Magnesium possesses a certain degree of ductility, and presents the colour and lustre of silver. It changes more slowly in the air than the preceding metals, and is not sensibly decomposed by very cold water; but at a temperature above 86° the decomposition commences, and at about 212° is very active. Heated to a dull red, either in the air or in oxygen, the metal ignites. It be- comes equally incandescent in chlorine. § 579. Only one compound of magnesium with oxygen is known, the protoxide, or magnesia, which is prepared by calcining the hydrocarbonate of magnesia, or the white magnesia of the pharma- ceutist. As this hydrocarbonate is very light, the magnesia pro- duced by it is also very light, and considerable bulk is required for any ordinary weight of matter. This circumstance is very inconvenient in several chemical processes, particularly in those which are effected, in the dry way, in vessels of limited dimen- sions. For these peculiar cases, magnesia is prepared by calcin- ing the nitrate of magnesia, which yields a much heavier oxide. Magnesia is a white powder, infusible at the highest temperatures of our furnaces. It is slightly soluble in water, requiring about 5000 times its weight of that liquid; and yet this solubility is suffi- cient to enable moistened magnesia to blue the tincture of litmus reddened by an acid. It is a powerful base, perfectly saturating COMPOUND OF MAGNESIUM WITH OXYGEN. * Magnesium was first isolated by M. Bussy, by adopting a process by which Woehler has already succeeded in preparing aluminum and glucinum. MAGNESIUM. 559 acids. It is precipitated by lime, but the chief cause of the pre- cipitation is the fact that magnesia is still less soluble in water than lime. Anhydrous magnesia does not produce any sensible degree of heat with water; for, although it forms a combination with it, the action is so slow that evolution of heat is inappreciable. A mono- hydrate of magnesia MgO + HO is formed in this case, and easily restored to the anhydrous state by heat. The same hydrate is precipitated when a solution of potassa is poured into that of a magnesian salt. Caustic magnesia is a powerful antidote in poisoning by arseni- ous acid, with which it combines to form an insoluble compound free from any poisonous effect.* The magnesia for this purpose should be in the hydrated state, or but slightly calcined, nor can its carbonate be substituted for it. § 580. The composition of magnesia may be obtained by the synthesis of sulphate of magnesia, that is, by ascertaining, as in the case of lime (§ 552), the weight of sulphate of magnesia yielded by a given weight of magnesia. It may also be deduced from the direct analysis of the sulphate, by determining the weight of sul- phate of baryta yielded by a known weight of sulphate of magnesia when its solution is precipitated by chloride of barium. SALTS OF MAGNESIA. Sulphate of Magnesia. § 581. Sulphate of magnesia, or Epsom salt, exists in several mineral springs, particularly in those of Epsom, in England, Seidlitz and Pullna, in Bohemia. These waters are used in medicine as a purgative, and owe their efficacy to the sulphate of magnesia which they contain. This sulphate, in mineral springs, appears to arise from the reaction of sulphate of lime in solution on the magnesian limestone which constitutes the formation. The water, charged with sulphate of lime, remaining for a long time on the magnesian soil, reacts on the carbonate of magnesia, carbonate of lime being de- posited, and sulphate of magnesia dissolved. The mineral waters, collected in shallow basins, concentrate by evaporation, and their complete evaporation yields crystallized sulphate of magnesia. Such a formation of sulphate of magnesia, by the reaction of sulphate of lime in solution on carbonate of magnesia, may be proved by direct experiment, by filtering slowly and repeatedly water saturated with sulphate of lime through a thick stratum of magnesian limestone, when the water will finally contain only sul- phate of magnesia. But an inverse decomposition can likewise be effected by operating at a high temperature. If carbonate of lime * Hydrated oxide of iron, made up into a thin mud with water, serves the same purpose, and is preferable on account of its cheapness.— W. L. Faber. 560 ALKALINO-EARTHY METALS. and a solution of sulphate of magnesia be heated to about 400° in a thick glass tube closed at both ends, sulphate of lime and car- bonate of magnesia are formed. The inverse reaction is an import- ant fact in geology, as it serves to explain the formation of native magnesian limestone. It is admitted that carbonate of magnesia has been formed by the reaction of carbonate of lime on sulphate of magnesia dissolved in the hot waters which covered the globe to a great depth, the lower strata of which, consequently, might have attained a very high temperature. The sulphate may also be obtained by adding sulphuric acid to the native carbonate of magnesia or magnesian limestones, such as dolomite, very rich in this carbonate; whereby it would form sul- phate of lime, but slightly soluble in water, and sulphate of mag- nesia, which is very soluble, particularly in hot water. Lastly, it was stated (§ 498 and § 501) that the mother waters of the salines contained considerable quantities of this salt; so that all the Epsom salt used in medicine might be obtained from these waters at a cheap rate. § 582. Sulphate of magnesia crystallizes, at ordinary tempera- tures, in small elongated prisms, having the formula MgO,S03 + 7IIO. If the crystallization takes place at an elevated temperature, the salt deposited contains only 6 equivalents of water, but if at several degrees below 32°, large crystals are obtained, of which the formula is Mg0,S03+12H0. Epsom salt, heated to 464°, still retains 1 equiv. of water, which it loses at a higher temperature. The anhydrous sulphate fuses at a red-heat. At 32°, 100 parts of water dissolve about 26 parts of the salt. The plate at page 407 contains its curve of solubility for the temperatures comprised be- tween 32° and 212°. Sulphate of magnesia combines with the alkaline sulphates and with that of ammonia, forming double salts which readily crystal- lize. The formula of the double sulphate of magnesia and potassa is Mg0,S03 + K0,S03+6II0. Considerable quantities of this salt are deposited during the evaporation of the mother waters of salines (§ 502). The formula of the double sulphate of magnesia and ammonia is Mg0,S03+(NHs,H0)S084-6H0 : it is isomor- phous with the double salt of potassa. Nitrate of Magnesia. § 583. This salt is prepared by dissolving magnesia in nitric acid. It is very soluble in water and deliquescent, and is entirely decom- posed at a red-heat, yielding a residue of pure magnesia. §584. Carbonate of magnesia is found in nature, generally in compact masses, but sometimes crystallized in rhombohedrons. It also exists in nature in combination with carbonate of lime, which Carbonate of Magnesia. MAGNESIUM. 561 is isomorphous with it, and nearly all limestones contain a small quantity of magnesia. The dolomite of mineralogists, which con- stitutes large formations in some countries, particularly in the Alps, is a double carbonate of lime and magnesia, with the formula Ca0,C03+Mg0,C03. When an alkaline carbonate is poured into the solution of a mag- nesian salt, a white gelatinous precipitate is formed, which is a hydrocarbonate of magnesia, that is, a compound of the hydrate and carbonate of magnesia. The proportions of the two compounds vary, according to the quantity of alkaline carbonate used, the state of concentration of the liquids, and their temperature. It is manufactured on a large scale for medicinal purposes, and is called in pharmacy white magnesia, {alba.) It is made as light as possible, for which purpose dilute and hot solutions of sulphate of magnesia and carbonate of soda are mixed together. The liquid is then fil- tered in rectangular baskets lined with muslin, which retains the precipitate. The hydrocarbonate, well washed and dried, presents the shape of square blocks of excessive lightness. White magnesia dissolves in considerable quantity in water charged with carbonic acid. The solution, exposed to the air, loses its carbonic acid, and the hydrated carbonate of magnesia MgO,COa+3HO is deposited.* Phosphates of Magnesia. § 585. A neutral phosphate of magnesia is obtained by decom- posing the hydrocarbonate by phosphoric acid. It is soluble in from 15 to 20 parts of water. Phosphate of magnesia forms, with phosphate of ammonia, double phosphates of very slight solubility. If a solution of phosphate of ammonia be added to a hot solution of sulphate of magnesia, small prismatic crystals of a double phos- phate are deposited on cooling, of the formula [2(NH3,H0)+H0]P05+(2Mg0+H0)P0s+6H0. If, on the contrary, we add to the solution of sulphate of mag- nesia, first chlorohydrate of ammonia, and afterward ammonia, which then forms no precipitate, as we shall see (§ 589), and lastly phosphate of ammonia, a granular precipitate is deposited, insoluble in the liquid in which the precipitation took place, and of which the formula is (NH3,HO + 2MgO)POs+6HO. But since this pre- cipitate is somewhat soluble in pure water, it must be washed with the smallest quantity of water possible.f This double phosphate possesses great interest; for it is often in this state of combination that, in chemical analyses, magnesia is precipitated from its solu- * Beautiful crystals are sometimes formed in well-corked bottles containing the bicarbonated solution.—J. C. B. f It should always be washed with ammoniacal water.—J. C. B. 562 ALKALINO-EARTHY METALS. tions. The same phosphate is likewise occasionally found in the animal economy, forming calculi in the bladder, and is called am- moniaco-magnesian phosphate. Silicates of Magnesia. §586. Silicates of magnesia are found in nature, generally com- bined with water, constituting, in some localities, entire rocks, or veins. The mineral called magnesite, or meerschaum, and talc, are composed of silicate of magnesia MgO,SiOs, combined with water.* Serpentine is also formed of silicate of magnesia combined with the hydrate of magnesia, its formula being 2(3Mg0,2Si08)+ MgO,2HO. Serpentine, which constitutes large masses in certain primitive rocks, is easily attacked by acids, and may be used in the preparation of sulphate of magnesia. It can be worked in a turning-lathe, into various ornamental objects remarkable for their beautiful colours. Silicate of magnesia, combined with other silicates, forms a great number of minerals constituting several primary rocks, such as chrysolite, augite, and hornblende. COMPOUND OF MAGNESIUM WITH SULPHUR. § 587. A monosulphide of magnesium is obtained by heating a mixture of sulphate of magnesia and charcoal in a crucible; but the product is always mixed with magnesia. A purer product is obtained by adding to the preceding mixture an alkaline polysul- phide, or a mixture of carbonate of soda and an excess of sulphur. Sulphide of magnesium is not obtained in the humid way, by boil- ing magnesia and sulphur with water; a behaviour which distin- guishes magnesia from the other alkaline earths and approximates it to the earths. Magnesia in fact forms a transition between the alkaline earths and the earths. COMPOUND OF MAGNESIUM WITH CHLORINE. § 588. Chloride of magnesium is obtained in solution in water, by treating white magnesia with chlorohydric acid. When the solution is evaporated to a high degree of concentration, it depo- sits, on cooling, crystals of a hydrated chloride with the formula MgCl+5HO ; hut if the evaporation be continued to dryness, the chloride is decomposed, chlorohydric acid being disengaged, and free magnesia remaining. In this decomposition, likewise, mag- nesia resembles the earths, the chlorides of which undergo a simi- lar alteration. Anhydrous chloride of magnesium is obtained by heating a mixture of magnesia and charcoal in a porcelain tube, through which a current of dry chlorine is passed; but, the chlo- ride having very slight volatility, remains mixed with charcoal. * Water is often wanting in talc.—J. C. B. MAGNESIUM. 563 To obtain the chloride pure, white magnesia is dissolved in con- centrated chlorohydric acid, sal ammoniac added thereto, and the whole evaporated to dryness. The residue is placed in a platinum crucible, and heated to redness over an alcohol-lamp. The chlo- ride of magnesium combines with the chlorohydrate of ammonia, and acquires sufficient stability to allow the water to be driven off by heat before reacting on the chloride. Heat then decomposes the dried double chloride, disengaging chlorohydrate of ammonia, and leaving chloride of magnesium in a melted state, which, on cooling, solidifies in a crystalline mass. This anhydrous chloride of magnesium is used (§ 578) in the preparation of metallic mag- nesium. Chloride of magnesium exists in sea-water, and the mother waters of the salines (§ 503) contain considerable quantities of it, which they deposit in the form of the double chloride of magnesium and potassium. DISTINCTIVE CHARACTERS OF THE SALTS OF MAGNESIA. § 589. The salts of magnesia yield white gelatinous precipitates with the alkaline carbonates, which distinguishes them from the alkaline salts. Ammonia, poured into a solution of a salt of magnesia which does not contain an excess of acid, nor any ammoniacal salt, yields a white precipitate, which the salts of baryta, strontia, and lime, do not yield under like circumstances. But if the magnesian liquid contains a sufficient quantity of any ammoniacal salt, it is no longer precipitable by ammonia, because the magnesian salt forms a double ammoniacal salt, not decomposable by ammonia. Nor is a precipitate produced if the liquid contains a great excess of acid; for, by adding ammonia to neutralize the liquid, a quan- tity of ammoniacal salt is formed, sufficient to produce the double magnesian salt not decomposable by ammonia. The same phe- nomenon is manifest when the magnesian salt exists in the neutral state in the liquid, a portion only of the magnesia being precipi- tated by the ammonia; for the acid transferred to the ammonia by the precipitated magnesia, forms a quantity of ammoniacal salt, sufficient to produce, with the magnesian salt which remains in the liquid, the double salt undecomposable by ammonia. This property likewise places magnesia between the alkaline earths and the earths. The salts of magnesia are precipitated by lime-water. They are never precipitated by the alkaline sulphates, unless the magnesian liquid be extremely concentrated, in which case, sulphate of magnesia might crystallize; but it is always easy to prove that these crystals are very soluble in water. The salts of baryta and strontian are, on the contrary, precipitated by the 564 ALKALINO-EAltTHY METALS. sulphates, even when their solutions are very dilute ; the salts of lime themselves yield a precipitate of sulphate of lime, easily re- cognised by its appearance, unless the liquid is extremely dilute. The salts of magnesia, heated before the blowpipe with a small quantity of nitrate of cobalt, yield a rose-coloured residue. DETERMINATION OF THE ALKALINE EARTHS, AND METHODS OF SEPARATING THEM FROM EACH OTHER, AND FROM THE ALKALIES. § 590. Baryta and strontia are always determined in the state of sulphates. When they are in solution, the liquid is boiled, a few drops of chlorohydric acid added to it, and then a solution of chloride of barium is poured in. The sulphates are deposited in the form of a granular powder, which is collected on a small filter, well washed with hot water, and dried on the filter. After desic- cation, the precipitate separates readily from the filter, and is carefully dropped into a platinum crucible, which is heated to red- ness over an alcohol-lamp. The filter being suspended by a pla- tinum wire over the crucible, is inflamed, and as it burns in the air, its ashes fall into the crucible. The crucible, with its con- tents, being weighed, the contents are removed, and the crucible cleaned. The crucible being replaced over the lamp, a second filter, of the same size as the first, and made from the same sheet of paper, is burned. This filter, to resemble the first as closely as possible, should have been washed with water acidulated with chlorohydric acid. The crucible is again weighed, and the weights necessary to restore the equilibrium of the scales represent the W'eight of the sulphate. The sulphate, enclosed in its filter, should not be calcined in the crucible, because a certain quantity of sul- phide of barium is always formed, and in order to have only the sulphate, it would be necessary to sprinkle it with sulphuric acid, and calcine it anew.* § 591. When a solution contains only lime, combined with a volatile acid or with sulphuric acid, the lime may be determined in the state of sulphate. To effect this, the liquid is evaporated to dryness in a porcelain, or still better, in a platinum capsule, the residue sprinkled with sulphuric acid, the excess of acid evapo- rated, and the substance heated to redness. The sulphate of lime * It is generally advisable to remove all the contents of a filter, and burn the latter separately, where it can be safely done without loss; but the method of burning the filter in the air over the crucible is objectionable, from the danger of losing particles of ashes of the filter, or of the substance adhering to it, from heated currents of air. A much better method is to burn the filter on the cover of the crucible, or to incline the crucible and burn the filter in it a little in front of the powder; and in either case to begin at a low red-heat, and finish it at a full red. By managing the heat properly, there is no danger of reducing the sulphate of baryta to a sulphide, for even sulphate of lead is burned in the same manner without the slightest detriment to the crucible, which would certainly be injured if sulphuret of lead were formed.—J. C. B. MAGNESIUM. 565 which remains is weighed. At other times, the liquid is evaporated to dryness with a small quantity of sulphuric acid, and treated with dilute alcohol, which does not sensibly dissolve the sulphate of lime, but which can dissolve other saline substances existing in the liquid with the sulphate. The sulphate of lime is washed with alcohol, and weighed after calcination.* Lime may also be precipitated by an alkaline carbonate, or still better, by an oxalate, the oxalate of lime being still more in- soluble than the carbonate, provided the liquid be made alkaline by the addition of a small quantity of ammonia. The precipitate may be determined either as caustic lime, as carbonate, or as sul- phate of lime. If it is to be determined as caustic lime, the oxa- late is calcined at a white-heat, and, after being weighed, it is ascertained whether the lime has become completely caustic, by sprinkling it with nitric acid, which should produce no efferves- cence. It is well, for the sake of exactness, to sprinkle the calcined matter with sulphuric acid, and determine the lime in the state of a sulphate, after a new calcination. In order to determine lime as carbonate, add carbonate of ammo- nia to the matter calcined in a platinum crucible, and heat it only to a dull red-heat, to drive off the excess of carbonate of ammonia. § 592. When magnesia exists alone in a liquid, combined with a volatile acid or with sulphuric acid, it may be determined as sulphate by proceeding exactly as with lime. It may also be pre- cipitated as carbonate by an alkaline carbonate; but it is advisable to evaporate the liquid to dryness and treat the residue with water. The carbonate of magnesia then separates completely: the precipitate is calcined at a red-heat, and weighed in the state of caustic magnesia. Magnesia is often determined in the state of phosphate. In this case, ammonia is first added to the liquid, and afterward a solution of phosphate of ammonia. The precipitate of phosphate of magnesia and ammonia is collected on a filter, quickly washed with water con- taining a little ammonia, and weighed after calcination. The phos- phate of magnesia thus obtained contains 36.6 pr. ct. of magnesia. § 593. Let us now suppose a liquid to contain at the same time alkaline bases, potassa or soda, and the four alkaline earths, ba- ryta, strontia, lime, and magnesia, only volatile acids being present. By adding an excess of carbonate of ammonia, the alkaline earths will be precipitated in the state of carbonates, leaving the alkalies alone in solution. This liquid is evaporated to dryness, after the addition of a small quantity of sulphuric acid. The residue, when calcined to redness and melted in a platinum crucible, will be com- * The best strength of alcohol is a mixture of about 6 measures of commercial alcohol (80 per cent.) 'with 5 measures of water, in which sulphate of lime is wholly insoluble, while the sulphates of magnesia and of the alkalies are soluble in it.—J. C. B. 566 ALKALINO-EARTHY METALS. 4 posed only of the alkaline sulphates, the ammoniacal salts having been driven off by the heat. The alkaline sulphates are weighed, and the proportions of potassa and soda they contain determined by the processes described (§525 bis, 526, and 527). The alkaline carbonates are dissolved in chlorohydric acid, the liquid is sufficiently diluted, heated to ebullition, and sulphuric acid or sulphate of ammonia added, by which baryta and strontia only are precipitated in the state of sulphates. They are weighed after calcination. To ascertain the proportions of these two bases, they are fused in a platinum crucible with three times their weight of pure carbonate of soda, and then treated with water. Baryta and strontia remain in the state of insoluble carbonates, while the sul- phuric acid of the sulphates is found in the alkaline liquids, from which, after adding an excess of chlorohydric acid, it is precipitated by chloride of barium. The weight of sulphuric acid thus obtained, compared with the weight of the sulphates of baryta and strontia with which it was combined, often permits a calcination of the pro- portions of these two bases to be made with sufficient accuracy, at least when they exist in nearly equal quantities in the solution. The proceeding is, in this case, similar to that explained (§ 525 bis) in the analysis of the sulphates of potassa and soda. The carbonates of baryta and strontia are, after being converted into chlorides by adding chlorohydric acid, evaporated to dryness, and treated with concentrated alcohol, which does not sensibly dissolve the chloride of barium, but readily takes the chloride of strontium into solution; thus the two bases are separated, and may be afterward determined in the state of sulphates. The liquid from which the baryta and strontia have been elimi- nated, now contains only lime and magnesia. It is saturated with ammonia until a decided alkaline reaction takes place; oxalate of ammonia is then added, which gives a precipitate of oxalate of lime. In this case, the presence of a large quantity of ammoniacal salts in the liquid prevents the precipitation of the magnesia. The oxalate of lime is determined according to § 591. The liquid, then containing only magnesia and ammoniacal salts, is evaporated to dryness, and the residue is heated to redness, after a small quantity of sulphuric acid has been added; by this operation the ammoniacal salts are driven off, and sulphate of mag- nesia alone remains. The magnesia may also be precipitated by phosphate of ammonia, and the phosphate of magnesia and ammo- nia weighed after calcination. § 594. A mixture of the salts of potassa, soda, baryta, strontia, lime, and magnesia, may also be analyzed by separating the bases in a rather different order. Precipitating the baryta and strontia by sulphuric acid, the lime by oxalate of ammonia, the magnesia by carbonate of ammonia, and afterward evaporating the liquid, the residue will contain only alkalies. 567 III. EARTHY METALS. ALUMINUM. Equivalent = 13.67; (170.9; 0=100). § 595. Aluminum* is one of the substances most extensively spread over the surface of the globe: its oxide, combined with silicic acid and a certain quantity of water, forms the clays. The silicate of alumina, combined with other silicates, constitutes seve- ral minerals, the most important of which are feldspar and mica, two constituent minerals of the granites, that is, of the primitive rocks forming the inner crust of the globe accessible to our means of observation. The name aluminum, given to this metal, is de- rived from alum., a double sulphate of alumina and potassa, which has for a long time been used in the arts. Aluminum is obtained by decomposing the anhydrous chloride of aluminum by potassium; the process is the same as that described for magnesium (§ 578). After the cooling of the crucible in which the chloride of aluminum has been heated with potassium, the sub- stance is treated with cold water, which dissolving the chloride of potassium, leaves the aluminum in the form of a gray powder, showing a metallic lustre when burnished. Aluminum ignites when heated in contact with the air; it does not decompose water at the ordinary temperature, but at 212° the decomposition is very manifest. Aluminum causes an evolution of hydrogen, on being dissolved in dilute acids or treated with a so- lution of potassa or soda; in other words, it decomposes water in the presence of acids or of powerful bases; a circumstance owing to the fact that this substance acts at the same time the part of an acid and a base. COMPOUND OF ALUMINUM WITH OXYGEN. § 596. Only one combination of aluminum with oxygen is known; it is obtained by precipitating a solution of alum by an excess of carbonate of ammonia: the white gelatinous precipitate, after being well washed with boiling water, dried and calcined, yields anhydrous alumina. It may also be obtained directly, by heating ammoniacal alum to a strong red-heat; but it often retains, when thus prepared, a small quantity of sulphuric acid. Alumina is a white powder, insoluble in water, readily soluble in a solution of potassa, soda, baryta, and strontia, except after being heated to * Aluminum was first isolated by M. Woehler. 568 EARTHY METALS. redness, and slightly soluble in a concentrated solution of ammo- nia : in the latter cases, it plays the part of a true acid, and seve- ral aluminates may be procured in a crystallized state. It also dissolves in the acids, yielding salts which invariably show a strong acid reaction. Calcined alumina, on the contrary, is with difficulty dissolved in potassa and the acids. The combination of alumina with the alkalies takes place, in all cases, at a red-heat. Alumina is found crystallized in nature, in the form of minerals, often possessing brilliant colours, Avhich are used by jewellers as precious stones. The crystalline form of these minerals belongs to the rhombohedric system ; their most ordinary form is that of the primitive rhombohedron, or six-sided prism. The names of these minerals vary with their colour; thus, native alumina, when blue, is called sapphire, and when red, takes the name of ruby. These colours are often owing to very minute quantities of colour- ing metallic oxides. Colourless and transparent alumina is known by the name of hyaline corundum. Lastly, it is most frequently met with in the form of opaque six-sided prisms, or even of rounded pebbles, coloured brown by oxide of iron. The density of mineral alumina is considerable, being about 3.9; it is, moreover, after the diamond, the hardest substance occurring in nature. On ac- count of this property, opaque corundum, called emery, is used to polish precious stones and glass. It is finely powdered, and sepa- rated into several sorts, according to its fineness; the powdered emery being suspended in water, the large particles fall to the bottom of the vessel, while the liquid, when allowed to rest for some time, holds the finer emery in suspension. Alumina is infusible in the heat of our furnaces; but it melts before the oxyhydrogen blowpipe, forming colourless and trans- parent globules, which often, on cooling, assume a crystalline tex- ture. In order to obtain artificially fused alumina, it is sufficient to heat common potassic alum, after having previously dishydrated it by heat in the oxyhydrogen blowpipe ; the sulphate of alumina is decomposed, the sulphate of potassa is volatilized at this high temperature, and there remains only alumina, which fuses when the temperature is sufficiently elevated. An addition of a small quantity of chromate of potassa to the alum imparts a red colour to the melted alumina, which then forms a perfect imitation of natural ruby. Alumina, precipitated from a solution of alum by carbonate of ammonia in excess, forms a gelatinous substance, hydrate of alu- mina, which is readily soluble in acids and alkaline liquids, but will not, however, combine with very feeble acids, such as carbonic. It does not lose its water by exposure in a dry vacuum, nor at the heat of boiling water, but must be heated to redness to be obtained perfectly anhydrous. Calcined alumina no longer combines with water, but it is a hygrometric substance, readily condensing the ALUMINUM. 569 moisture of the atmosphere. Hydrated alumina is found in nature: diaspose is one of these crystallized hydrates, with the formula A1203+3H0; gibbsite is also a hydrate of alumina. By subjecting a solution of hydrate of alumina in potassa to slow evaporation, an aluminate of potassa is obtained in crystalline grains, of the formula K0,A1203. Baryta gives a similar com- pound. The mineral called spinell is an aluminate of magnesia, of which the formula is MgO,Al203. Several of these crystallized aluminates may be obtained by mixing together suitable propor- tions of alumina and the metallic oxides we wish to combine, add- ing to the mixture 5 or 6 times its weight of horacic acid, stirring it well, and exposing the whole, placed in a platinum crucible, for several days to a high temperature in a porcelain furnace. The boracic acid first melts and dissolves the alumina and the other metallic oxides, but the tension of vapour of the boracic acid at this temperature being very great, it is evaporated but slowly. The alumina and the metallic bases, being in presence of the same solvent, combine with each other. In proportion as the solvent evaporates, the compound separates, and forms, as it is slowly deposited, small well-terminated crystals. By the same process, several other compounds found in the mineral kingdom, which are infusible in the heat of our furnaces, may be obtained crystallized. § 597. The composition of alumina has been deduced from the analysis of alum. Potassic alum is a double sulphate of alumina and potassa, containing water of crystallization, which it loses at a moderate heat. 10 gr. of anhydrous potassic alum are dissolved in hot water, and the alumina is precipitated by an excess of car- bonate of ammonia: the precipitate, when collected on a filter, is well washed, and then weighed after calcination. 1.986 gr. of alumina are obtained. The liquids are evaporated: the residue, when calcined to redness in a platinum crucible, is composed of sulphate of potassa alone, the ammoniacal salts having been vola- tilized by heat. The sulphate of potassa thus obtained weighs 3.878 gr. 10 other grains of anhydrous alum are then dissolved in boiling water, and the sulphuric acid precipitated by an excess of chloride of barium: in this case, 18.044 gr. of sulphate of baryta are found, which contain 6.188 gr. of sulphuric acid. Now, the 3.378 gr. of sulphate of potassa contain 1.547 gr. of sulphuric acid; the weight 1.986 gr. of alumina is therefore combined with the weight 4.641 gr. of sulphuric acid. This sulphate of alumina is regarded as a neutral sulphate ; knowing the oxygen of the sulphuric acid to be treble that of the oxygen contained in the base, and finding the weight of that con- tained in 4.641 gr. of sulphuric acid to be 2.784 gr., one-third of this weight, that is 0.928, is combined with the 1.986 gr. of alumina. Alumina is therefore composed of 570 EARTHY METALS, Aluminum 1.058 53.27 Oxygen 0.928 46.73 Alumina 1.986 100.00 It now remains for us to discover the formula suitable to alumina. If we suppose that this base presents the same formula as the bases previously studied, the formula should be written AlO, and the equivalent of alumina would be given by the pro- portion, 46.73 : 53.27 : : 100 : x, whence z =113.99. But this formula AlO is contradicted by considerations founded on isomorphism. Alumina never appears as isomorphous with an oxide of the formula RO, but is, on the contrary, always isomor- phous with certain oxides R303, of which the formulas are certain. Thus, a series of alums, having the same crystalline form and very analogous properties, are obtained by combining sulphate of potassa with the sulphates of sesquioxide of iron Fe203, sesquioxide of manganese Mn203, and oxide of chrome Cr903. Native crystallized alumina, or corundum, presents also the same crystalline form as the native sesquioxide of iron, or specular iron, and the sesqui- oxide of chrome. The formula of alumina, therefore, should undoubtedly be written A1203; consequently the neutral sulphate of alumina must take the formula A1203,3S03. The equivalent of alumina is then obtained by the proportion, 46.73 : 53.27 :: 300 : 2x, whence x =170.98. SALTS FORMED BY ALUMINA. Sulphate of Alumina. § 598. The neutral sulphate of alumina has for a long time been manufactured on a large scale, being employed in dyeing, and advantageously substituted for alum. It is obtained by treating clay with sulphuric acid, for which purpose the clays containing the smallest quantity of iron possible, the kaolins, for example, are selected. They are calcined at a dull red-heat in ovens, then ground to powder, and mixed with one-half of their weight of sulphuric acid of the density 1.45: this mixture is heated in another oven, until sulphuric acid begins to be driven off. It is then withdrawn and allowed to rest for several days, when the mass, treated with water, yields a solution of sulphate of alumina. But as this solution almost always contains some traces of a salt of iron, which would destroy its use in dyeing, it is important to separate this ingredient, which is effected by precipitation with prussiate of potash, added to the liquid until a blue precipitate is no longer formed. It is then evaporated; the sirupy residue ALUMINUM. 571 is poured into small leaden basins, where it solidifies in the form of a white mass. Sulphate of alumina is soluble in double its weight of water. A solution saturated when hot deposits the salt in the form of small crystalline spangles, of which the formula is A1303,3S03+18HO. A solution of neutral sulphate of alumina can dissolve an addi- tional proportion of alumina when digested with hydrated alumina: a basic sulphate of alumina, of the formula 2A1303,3S03 is then formed. Lastly, by pouring ammonia into a solution of sulphate of alumina, a tribasic sulphate of alumina is precipitated in the form of a crystalline powder, having for its formula Ala03,S03-|-9H0 —a compound occurring in nature. § 599. Sulphate of alumina is very important on account of the double salts which it forms with the alkaline sulphates and with that of ammonia, a class of salts comprised under the general name of alums. Most frequently, however, this name is given to the double sulphate of alumina and potassa. These combinations are easily prepared by mixing together the solutions of the two sulphates, and evaporating the liquid to allow the double salt to crystallize. Potassic and ammoniacal alum are very slightly so- luble in cold water, and readily crystallize : sodic alum, on the contrary, is very soluble. The best mode of obtaining sodic alum in crystals is by pouring a layer of absolute alcohol on a con- centrated solution of the salt, and allowing it to rest for several days; the alcohol gradually combining with the water, allows the sodic alum to be deposited in the form of beautiful octahedral crystals. These three alums follow the regular system of crystallization: their ordinary forms are the octahedron and cube, or combinations of the two, in which sometimes the octahedron, and sometimes the cube predominates. Their composition is also similar; thus, the formula of Potassic alum is .....KO,SOs-f A1303,3S03+24H0. Sodic alum Na0,S03+Al303,3S03-f 24HO. Ammoniacal alum (NH3,H0)S03+A1303,3S0s+24H0. The basic sesquioxides which are isomorphous with alumina, form, with the sulphates of potassa, soda, and ammonia, perfectly similar salts, also called alums. These new alums crystallize in octahedrons or in cubes, like those formed by the sulphate of alumina, and have similar formulas; thus, the sulphate of ses- quioxide of iron, Fe303,3S03, yields : A ferri-potassic alum K0,S03+Fe303,3S03+24H0. A ferri-sodic alum Na0,S03+Fe303,3S03+24H0. A ferri-ammoniacal alum (NH3,H0)S03-f-Fea03,3S03+24H0. 572 EARTHY METALS. The sulphate of sesquioxide of manganese Mn303,3S03 gives, in the same manner, A mangani-potassic alum K0,S03+Mn303,3S03+24II0. A mangani-sodic alum Na0,S03 + Mn303,3S03 + 24II0. A mangani-ammoniacal alum (NH3,H0)S03+Mn303,3S03 + 24H0. Finally, the sesquioxide of chrome gives the following alums : A chromi-potassic alum K0,S03+Cr303,3S03+24H0. A chromi-sodic alum Na0,S03 + Cr203,3S03 + 24II0. A chromi-ammoniacal alum...(NH3,H0)S03+Cr303,3S02+24H0. We shall frequently refer to the existence of the isomorphous alums in proof of the isomorphism of the sesquioxides. Potassic alum is the one most used in the arts: it is employed in dyeing, and its manufacture has received great attention in all countries. Potassic alum dissolves in 18.4 parts of cold, and in only 0.75 of boiling water; its curve of solubility may be seen in the plate at page 407. It is deposited, on cooling, in beautiful octahedrons, the angles of which are often terminated by the faces of the cube, and is then called octahedral alum; but it may also be obtained crystallized in cubes by pouring carbonate of potassa into an ordi- nary solution of alum, saturated at 122° : a sub-sulphate of alumina is precipitated, which redissolves if the liquid be shaken. On allowing the liquid afterward to cool, the alum crystallizes in its ordinary form, but it then takes the form of cubes, sometimes modified by the faces of the octahedron, the cube, however, always predominating. This alum is called cubic-alum, and is more esteemed in commerce than the octahedral, the latter frequently containing some sulphate of iron, which, as it changes the shades of colours, is very injurious in dyeing. Now, as alum crystallizes in cubes only in liquids containing an excess of alumina, and con- sequently deprived of oxide of iron, the cubic form of alum is a proof of its purity. The taste of alum is, at first, sweet, and like sugar, but it soon becomes very astringent. When heated, it first melts in its water of crystallization ; then, on cooling, solidifies into vitreous masses, called rock-alum. Heated still further, it gradually loses its water and becomes anhydrous. When alum is heated in a crucible, the substance, at first liquid, becomes more and more doughy, as it loses its water; it swells considerably, rising above the crucible, and if it be gradually heated, the anhydrous alum assumes the form of a spongy mass, which rises in a mushroom-shape above the ALUMINUM. 573 crucible (fig. 379). Dishy drated alum is used in medicine as a caustic : it is called burnt alum. Lastly, alum decomposes when heated to redness, disengaging a mixture of sulphur- ous acid and oxygen, and leaving as a residue free alumina and unaltered sulphate of potassa, which latter salt may be separated by dissolving in water. Alum calcined with charcoal, or better, with lamp- black, yields a very finely-divided residue, consisting of alumina, sulphuret of potassium, and charcoal. This residue is a true pyrophorus: it ignites when exposed to a damp atmosphere. § 600. For the manufacture of alum, several methods are em- ployed : 1st. To a solution of sulphate of alumina obtained by the action of sulphuric acid on clay, as stated in § 598, sulphate of potassa or chloride of potassium is added, and the liquids are allowed to cool, being constantly shaken. The alum is precipitated in the form of small granular crystals, which, after being perfectly drained, are washed with a small quantity of cold water: from a solution of these crystals in boiling water, octahedral masses of alum are deposited on cooling. Chloride of potassium is preferable to sulphate of potassa, because it converts the salts of iron mixed with the sulphate of alumina into chloride of iron, which, being much more soluble than the sulphate, is consequently not precipi- tated with the alum. This method is, however, generally too expensive to be adopted in the manufacture of alum. 2d. The greater portion of alum is obtained by the spontaneous or artificial roasting of certain argillaceous rocks, strongly impreg- nated with small crystals of sulphurets of iron. The most common is the bisulphuret FeSa, or pyrites; it is sometimes, however, a sulphuret Fe2S3, or magnetic pyrites. These argillaceous and py- ritous rocks are met with in great abundance in two geological formations : they are found in the transition rocks, where they form schists, which commonly are bituminous, and also occur in the formation of the tertiary rocks, immediately above the chalk. These latter aluminous schists are much less aggregated: their roasting is more easy, and frequently takes place spontaneously in the air. The aluminous schists are placed in large prismatic heaps on a layer of combustible matter laid on an impervious hearth, which is slightly inclined. The combustible is fired, which soon causes the sulphur of the pyrites and the bituminous matter with Avhich the schist is impregnated to ignite also. The combustion must be carefully regulated, so that the temperature may not rise too high in certain parts of the mass: this is done by covering the heaps with powdered schist already calcined, or, on the other hand, by poking up the parts where the combustion is going on too slowly. Small quantities of water are from time to time poured upon the Fie. 379. 574 EARTHY METALS. heap. The combustion ceases after a period of five or six months: the heaps, which then are much reduced in size, are demolished, the substance sprinkled with small quantities of water, and exposed to the air for some time. The solutions arising from these wash- ings, or from the rain fallen on the heap, are conducted into water- tight reservoirs. Lastly, the substance, subjected to a methodic system of lixiviation, yields solutions sufficiently concentrated to be evaporated by fire. The aluminous schists of the tertiary rocks are much more changeable: it suffices to expose them to the air and wet them from time to time, to effect their spontaneous oxidation. Iron py- rites absorbs the oxygen of the air, and is converted into sulphate of iron and sulphuric acid, which, combining gradually with the alumina of the schist, form sulphate of alumina: FeS2+70=Fe0,S03+S03. In Picardy, large quantities of alum are obtained from the ter- tiary schists, which rapidly decompose in the air. They are made into heaps, which are turned from time to time, and occasionally w'etted, if the season be very dry. Oxidation goes on rapidly, and sometimes the heat evolved is even sufficient to fire the mass, which, however, must be avoided, as in this case a considerable quantity of sulphurous acid is disengaged. When the sulphatization is suf- ficiently advanced, the matter is lixiviated, and the washings, which mark 18° or 20° on the areometer, are subjected to evaporation: on being allowed to cool, they deposit a large quantity of sulphate of protoxide of iron, while the mother liquid contains the sulphate of alumina. Chloride of potassium is poured into the hot solutions, and they then are allowed to cool; when alum begins to be de- posited, the crystallization is disturbed by constant stirring. The alum then precipitates in a crystalline sand, which is gradually withdrawn by a rake, and allowed to drain on an inclined plane, from which the solution is conducted into the crystallizing vessel. The washed schists may yield an additional quantity of sulphates, but then the roasting must be assisted by artificial heat, by arrang- ing them in large prismatic heaps on a layer of brushwood which is ignited. The pyrites and the bituminous matter taking fire, soon extend the combustion through the whole mass: the temperature is regulated by making openings in the almost impervious cover- ing of the mass. Soluble sulphates, but principally sulphate of alumina, are again formed, as the greater portion of the sulphate of iron passes into the state of an insoluble sub-sulphate of sesqui- oxide of iron. By treating the roasted schists, which present an ochrous colour, with water, the sulphate of alumina and a certain quantity of the sulphate of protoxide of iron are dissolved; the solu- tion is evaporated to a proper degree of concentration, and then ALUMINUM. 575 treated in the same manner as the first lixiviation, to obtain the alum. The alum thus obtained requires to be purified by recrystalliza- tion, to effect which the impure crystalline sand is washed with a small quantity of cold water, and then dissolved in boiling water. The hot solution, on being allowed to cool in casks, deposits alum in large crystals on the sides of the casks. When the solution is completely cooled, and deposits no more crystals, the mother liquid is run off, the casks are taken to pieces by removing the iron hoops which hold the staves, and a crystalline mass of alum, shaped like the inside of the cask, is removed. This, after being broken into large pieces, and washed with a small quantity of cold water, is ready for sale. 3d. In some localities, principally at Tolfa, near Rome, a rock, called alunite, or alum-stone, is found, from which a highly-valued alum, called Roman alum, is obtained. The formula of alunite is K0,S03+A1303,S03. It is heated in ovens until sulphuric acid begins to be disengaged; by subsequent treatment with water, the ordinary alum is dissolved, leaving a residue of alumina. The liquid, when evaporated, yields cubic crystals of alum, generally tinged to a rose-colour by a small quantity of peroxide of iron, which, however, is not injurious in dyeing, on account of its insolu- bility in water. Roman alum is more valuable than the ordinary kind, as it is certain to contain no soluble iron; but this alum is now made artificially, by adding carbonate of potassa, which pre- cipitates a certain quantity of subsulphate of alumina, to a solution of ordinary alum. By shaking the liquid, and exposing it for some time to the air, the subsulphate is redissolved, and hydrated peroxide of iron remains: by evaporating the liquid, cubic alum deposits. This alum is colourless, but, for a long time, dyers would not make use of it. To make it resemble Roman alum, the manu- facturers then introduced it into casks with a small quantity of pounded brick: by letting the casks revolve for a few minutes, the ordinary colour of Roman alum was imparted to the article. If carbonate of potassa be poured into a boiling solution of alum, a subsulphate of alumina is at first precipitated, but immediately redissolves in the liquid ; however, if the addition of the carbonate of potassa be continued, a granular precipitate, which does not dis- solve by agitation, is soon formed: the composition of this precipi- tate is the same as that of the alunite of Tolfa, and it is called insoluble alum. Silicates of Alumina. § 601. The silicates of alumina exist in great abundance in na- ture, and possess a high degree of interest. They are sometimes found crystallized, but are chiefly important in their hydrated 576 EARTHY METALS. state. Thus, our ordinary clays, porcelain-earth, or kaolin, and the halloysites are merely hydrosilicates of alumina, containing, however, a small quantity of silicate of potassa. These substances are evidently produced by decomposition of the primitive rocks, chiefly the granites: the alkaline silicate of the constituent mine- rals of these have been dissolved, silicate of alumina, more or less pure, has remained, and was drifted off by water, forming deposits in new basins. The feldspars are double silicates, formed by an alkaline silicate and that of alumina: the formula of ordinary or orthose feldspar is K0,Si03+Al303,3Si03. Frequently, however, lime or magnesia takes the place of a part of the potassa. Minerals which have been for a long time confounded with feld- spar, on account of the resemblance of their external charac- ters, or a certain analogy in their chemical composition, are also known. They have been called albite, petalite, triphan, and labra- dorite, according as soda, lithine, or lime takes the place of a part of the potassa. The clays are found in the various geological formations of rocks. The purest clay is that constituting kaolin, or porcelain-earth: it is found in white, amorphous, friable masses, forming with water merely a slightly cohesive paste. Kaolin generally is the result of the decomposition of a feldspathic rock in situ. In some locali- ties, this alteration may be traced from the intact feldspar forming the interior of the rock to the most friable kaolin on the surface. This clay frequently contains small fragments of unaltered feld- spar, which are easily separated by levigation. The formula of kaolin, thus washed, approaches closely that of Al303,Si03+2II0. § 602. The ordinary clays do not differ greatly from this com- position ; but they are frequently mixed with various proportions of quartzose sand, oxide of iron, and carbonate of lime, which affect considerably the physical and chemical qualities of the clay. Pure clay is eminently plastic, that is, it forms a very pliant paste with water, which may be moulded and kneaded into any shape. This is called fat clay; but when it contains any considerable proportion of foreign matters, its plasticity is greatly diminished, and it is then said to be poor. Clay mixed with a considerable proportion of carbonate of lime is called marl. The chemical pro- perties of clay are not less affected than their physical by the ad- mixture of foreign matters; thus, pure clay, which is completely infusible in the highest heat of our furnaces, or also when mixed with sand, becomes fusible when it contains any considerable pro- portion of oxide of iron or carbonate of lime. Certain kinds of clay, known by the name of fullers earth, are used in the scouring of woollen stuffs. These clays are first levi- gated, to separate the quartzose particles they may contain, the fuller’s earth, well dried, is then powdered and spread over the ALUMINUM. 577 cloth to be scoured, and the whole passed over a cylinder. By its capillarity, the clay absorbs all the grease in the cloth. The intimate mixture of clay with hydrated peroxide of iron is called ochre, or ochrous earth. Ochres are used in painting; their shades vary with the quantity of oxide of iron they contain. An addition of hydrate of sesquioxide of manganese imparts a brown hue to them. Sienna earth is a clay of this kind. COMPOUND OF ALUMINUM WITH SULPHUR. § 603. Hitherto sulphuret of aluminum has been obtained only by heating aluminum in the vapour of sulphur, as a blackish gray mass, assuming, when burnished, a slightly metallic lustre: it cannot be obtained in the moist way. When sulfhydrate of ammonia is added to a solution of a salt of alumina, sulphohydric acid gas is evolved, and the alumina is precipitated in the state of a hydrate. COMPOUND OF ALUMINUM WITH CHLORINE. § 604. By dissolving aluminum in aqueous chlorohydric acid, a solution of chloride of aluminum is obtained, which may be crys- tallized in a dry vacuum; very deliquescent crystals, of which the formula is Al3Cl3-fl2HO, are deposited. Their water of crys- tallization cannot he expelled by heat without decomposition: chlorohydric acid is disengaged, and the isolated alumina remains. Anhydrous chloride of aluminum may, however, be prepared by allowing dry chlorine to act on a mixture of alumina and charcoal heated to redness in a porcelain tube. The chlorine will not attack alumina when alone; but, when the alumina is mixed with charcoal, oxide of carbon gas is evolved, and chloride of aluminum, being volatile, condenses in a receiver placed in front of the por- celain tube. In order to obtain an intimate mixture of alumina and carbon, alumina and lampblack are ground together, a small Fig. 380. 578 EARTHY METALS. quantity of oil is added, and the pasty mixture rolled into small balls, which are calcined in an earthenware crucible. These small porous masses are introduced into a porcelain tube, arranged in a reverberatory furnace (fig. 380). Through one end of the tube a current of dry chlorine is passed, while the other enters an allonge which communicates with a well-cooled bottle ; the chloride of aluminum condenses in the allonge and receiver, in the form of small crystalline laminae, of a yellowish-white colour. Larger quan- tities of this substance may be obtained, by replacing the porcelain tube by a tubulated stone-ware retort, which will contain a larger quantity of the mixture of carbon and alumina. The apparatus must be then arranged as represented in fig. 265. Chloride of aluminum volatilizes at a temperature slightly above 212° : it fumes in the air, and rapidly attracts moisture, and should therefore be kept in a ground-stoppered bottle. DISTINCTIVE CHARACTERS OF THE SALTS OF ALUMINA. § 605. The solutions of the salts of alumina are precipitated by ammonia, a characteristic distinguishing them from the alkaline and alkalino-earthy salts, but which may, nevertheless, confound them with the salts of magnesia. We have seen (§ 589) that if a suffi- cient quantity of an ammoniacal salt be added to a magnesian salt, the latter is no longer precipitated by ammonia: a salt of alumina, however, is always precipitated. Caustic potassa and soda precipitate the salts of alumina, but an excess of either of these reagents immediately redissolves the precipitate. This character distinguishes with great accuracy the salts of alumina from those of the alkalies and alkaline earths. The salts of alumina are precipitated by lime-water. The alkaline carbonates and carbonate of ammonia, poured into the solution of a salt of alumina, precipitate hydrated alumina, which, when well washed, will redissolve in acids without effervescence. The sulfhydrates also precipitate hydrated alumina. If sulphate of potassa be added to a concentrated and hot solu- tion of a salt of alumina, octahedral crystals of alum are deposited on cooling: from a dilute solution, the crystals of alum are also deposited by evaporation. The salts of alumina, heated before the blowpipe with a small quantity of nitrate of cobalt, give a substance of a beautiful cha- racteristic blue colour. 579 GLUCINUM. Equiyalent=6.96. § 606. The oxide of glucinum, or glucina * exists in several minerals in combination with silicic acid. Of these, the most common is the emerald, a combination of silicate of alumina and silicate of glucina, of the formula Gl203,Si03-|-Al303,Si03. The crystalline form of the emerald is the regular 6-sided prism, be- longing to the rhombohedric system : the mineral is found in the state of a stone, but presenting a very evident crystallization, in the environs of Limoges. The emerald is rarely found in the transparent state; sometimes it exhibits beautiful colours, and possesses great value as a precious stone. The transparent and green emerald alone is called emerald in jewelry. When it ex- hibits only a pale-green hue, it is called beryl; and lastly, when it is bluish-green, bears the name of aqua marina. Glucinum is obtained, like aluminum, by heating in a closed platinum crucible a mixture of anhydrous chloride of glucinum with potassium: the same process is followed as in the preparation of aluminum and magnesium. Glucinum appears in the form of a grayish powder, which acquires a metallic lustre by burnishing: it decomposes water only at the boiling point. Heated in the air, it becomes incandescent, and is converted into an oxide: in acid or alkaline liquids it dissolves with the evolution of hydrogen gas. COMPOUND OF GLUCINUM WITH OXYGEN. § 607. Only one compound of glucinum with oxygen is known: it is called glucina. Glucina is obtained from the Limoges eme- rald, by finely powdering the mineral, and melting it in a platinum crucible with treble its weight of carbonate of potassa. The sub- stance is afterwards treated with sulphuric acid, and then with water, which dissolves the sulphates of alumina, potassa, and glucina, leaving the silex, which is easily separated by filtration. The liquid is evaporated by boiling: on being allowed to cool, the greater portion of the alumina separates in the state of crystallized alum. An excess of ammonia added to the mother liquid diluted with water, precipitates at once the balance of alumina, sesqui- oxide of iron, and glucina. The moist precipitate is left to digest with a concentrated solution of carbonate of ammonia, which dis- solves only the glucina, from which the residue of alumina and sesquioxide of iron are separated by filtration: the glucina then * Glucina was discovered in 1797, by Vauquelin. M. Woehler first isolated glucinum. 580 EARTHY METALS. precipitates by boiling, as a carbonate, which, when calcined, leaves pure anhydrous glucina. Glucina presents the appearance of a white powder, soft to the touch, insoluble in water, infusible in the heat of our furnaces, of the specific gravity 3.0. It is soluble in a solution of caustic potassa and soda, but ammonia will not sensibly dissolve it. § 608. The composition of glucina has been deduced from the analysis of the chloride of glucinum. It has been found that 10 gr. of chloride of glucinum contain 8.842 of chlorine. The proper formula of glucina still remains to be known. If the formula GIO be assigned to it, and consequently, the formula of G1C1 to the chloride of glucinum, the equivalent of glucinum will be calculated by the proportion, 8.842 :1.158 : : 443.2 : x, whence x =58.04. Assuming, on the contrary, that the composition of glucina is analogous to that of alumina, that is, if its formula is assumed as GL03, the equivalent will be given by the proportion, The question is here much more difficult to decide than in the case of aluminum, as, in the case of the latter metal, we had iso- morphism for a guide, while, for glucinum, no isomorphism of any of its combinations with a corresponding compound of aluminum, or with any such formed by the oxides RO, has been discovered. Thus, chemists do not agree upon the formula of glucina; and, while some assign to this base the formula GIO, and place it aside of magnesia, others give it the formula GlaOs, and rank glucinum with aluminum. 8.842 :1.158 :: 1329.6 : 2x, whence x =87.06. SALTS FORMED BY GLUCINA. § 609. Glucina has a stronger affinity for acids than alumina. Its salts have a sweet taste, from which it has derived the name of glucina, (from y%vxv$, “sweet.”) The hydrate of glucina is obtained by precipitating the salt of glucina by ammonia: it is a white gelatinous substance, readily parting with its water when heated. Glucina forms several compounds with sulphuric acid: the neutral sulphate Gla03,3S03+12110 yields beautiful crystals. COMBINATION OF GLUCINA WITH CHLORINE. § 610. Hydrated glucina dissolves readily in chlorohydric acid: the solution, when evaporated, deposits crystals of hydrated chlo- ride of the formula GlaCl3+12HO. Anhydrous chloride of glucinum is obtained by the process de- scribed (§ 604) for the chloride of aluminum. It volatilizes in the shape of small white crystalline spangles. GLUCINUM. 581 DISTINCTIVE CHARACTERS OF THE SALTS OF GLUCINA. § 611. The salts of glucina are precipitated by ammonia, even in the presence of an excess of ammoniacal salt: solutions of po- tassa and soda also precipitate them, but an excess of alkali redis- solves the precipitate. These two properties distinguish the salts of glucina from the alkaline and alkalino-earthy salts, but confound them with the salts of alumina. The salts of glucina are distinguished from those of alumina by not forming, like the latter, an alum with the sulphate of potassa, and by the property of carbonate of ammonia in excess dissolving the precipitate of carbonate of glucina, which it, at first, produces in glucinic solutions. The alkaline carbonates likewise precipitate the salts of glucina, but the carbonate of glucina is sensibly soluble in an excess of the reagent. The salts of glucina do not turn blue when heated before the blowpipe with a small quantity of nitrate of cobalt. 582 ZIRCONIUM. Equivalent = 34.0. § 612. The oxide of zirconium, or zirconia,* exists in consider- able quantity in a well-crystallized mineral called zircon, a silicate of zirconia 2Zr203,Si03, containing most frequently a small quan- tity of oxide of iron. In order to extract the zirconia, the zircons are heated in a crucible and thrown red-hot into cold water: by this sudden cooling, they become friable, and may be finely pul- verized. The powdered zircon is heated to a strong white-heat in a platinum crucible, with thrice its weight of carbonate of potassa: the mass, when calcined, is treated with chlorohydric acid, the solution is evaporated to dryness, and again treated with water. The silex is evaporated by filtering, and sulfhydrate of ammonia is added to the liquid, which precipitates the zirconia in the state of a hydrate, and the iron as protosulphuret. The clear liquid is decanted off after settling, and the precipitate allowed to digest for several hours with a solution of sulphuric acid, by which the sulphuret of iron is dissolved in the state of a hyposulphite, while the zirconia remains perfectly white: it is calcined after being well washed. Zirconia is a white powder, insoluble in water, and infusible at the temperature of our furnaces. When calcined, it dissolves with great difficulty in the acids: it is, however, readily dissolved in them when in the state of a hydrate. Zirconium is obtained by decomposing the fluoride of zirconium by potassium ; the metal appears in the form of a grayish powder, which assumes, when burnished, a metallic lustre. DISTINCTIVE CHARACTERS OF THE SALTS OF ZIRCONIA. § 613. The solutions of the salts of zirconia are precipitated by caustic potassa and soda; but the precipitate is not redissolved in an excess of the reagent: a characteristic which distinguishes zir- conia from alumina and glucina. Ammonia behaves with solutions of zirconia like as with those of potassa and soda. A concentrated solution of sulphate of zirconia yields, with sul- phate of potassa, a white crystalline precipitate, which completely separates when the liquid is saturated with sulphate of potassa. * Zirconia was discovered by Klaproth, in 1789. 583 THORIUM. § 614. The oxide of thorium, or thorina,* has hitherto been dis- covered only in two very rare minerals, called thorite and pyro- chlore. Thorina is chiefly obtained from thorite by reducing this mineral to a fine powder, and boiling it with chlorohydric acid; chlorine is disengaged; the solution is evaporated to dryness and treated with water. The liquid, when filtered, is subjected to a current of sulphuretted hydrogen, which precipitates a small quan- tity of sulphuret of tin and lead, which is separated by filtration. A solution of ammonia is then added to the liquid, which precipi- tates the thorina mixed with oxides of iron and manganese. The precipitate is then redissolved in sulphuric acid, and the liquid rapidly concentrated by ebullition, when the sulphate of thorina, which is very slightly soluble in hot water, is soon precipitated; it is collected on a filter and washed with boiling water. Sulphate of thorina is remarkable for being more soluble in cold than in boiling water. Calcined, it yields pure thorina. Thorina is a very heavy white powder: its specific gravity is about 9.4, greatly surpassing that of baryta. Thorina contains 11.84 per cent, of oxygen. YTTRIUM, ERBIUM, TERBIUM. § 615. These three metals have been discovered in some rare minerals, to which mineralogists have assigned the names of gado- linite, orthite, and yttrotantalite. Their properties are but little known, and we shall not stop to consider them. The oxides of these metals are called yttria, erbia, and terbia.f CERIUM, LANTHANIUM, DIDYMIUM. § 616. These three metals have been found together in several minerals, the most important of which is cerite.% The three me- tallic oxides exist in it, in combination with silicic acid. We shall not describe the combinations of these metals, as they are but little known, and have hitherto received no application. * Thorina was discovered by Berzelius. f Yttria was discovered in 1794, by Gadolin. Erbia and terbia have been recently discovered by M. Mosander. J Cerium was discovered in 1809, by Berzelius and Hisinger. Lanthanium and didymium have been recently discovered by M. Mosander. 584 DETERMINATION OF EARTHS. DETERMINATION OF THE EARTHS : THEIR SEPARATION FROM THE ALKALIES AND ALKALINE EARTHS. § 61T. We shall here treat only of alumina and glucina; the other earths being so rare that nothing need be said concerning the methods of determining them. Alumina and glucina are always determined in the state of an- hydrous alumina and glucina. To effect this, the bases are calcined to redness in a platinum crucible : it is advisable to allow the sub- stance to cool in a closed crucible, and weigh it rapidly, as it soon absorbs the moisture of the air. Alumina and glucina are generally precipitated from their solu- tions by ammonia; but it is important not to forget that these two hydrated bases are sensibly soluble in liquids highly charged with ammonia; it is therefore better, when possible, to effect the pre- cipitation by sulfhydrate of ammonia. § 618. When the alumina and glucina have been weighed toge- ther after calcination, they are separated by treatment with con- centrated sulphuric acid, which dissolves them when assisted by heat, although but slowly if the substance has been strongly cal- cined. It is evaporated to dryness, treated with water, and then precipitated by carbonate of ammonia, in which the glucina redis- solves. The precipitate of alumina should be digested several times with a solution of carbonate of ammonia, if the glucina is to be dissolved. § 619. When alumina and glucina exist together in a solution with alkalies and alkaline earths, they are separated by supersatu- rating the liquid with highly caustic ammonia, which precipitates only alumina and glucina. Sometimes, however, if the liquid con- tains a great deal of magnesia, a part of this base is deposited, because then a quantity of ammoniacal salt sufficient completely to prevent the precipitation of the magnesia by ammonia has not formed during the saturation. In this case, the moist precipitate is redissolved in chlorohydric acid, and an excess of ammonia is added; the magnesia then remains in the liquid. Alumina and the majority of the earths, precipitated from their solutions, form gelatinous substances, which it is very diffi- cult to wash completely. For this purpose, the washing-bottle represented in figs. 381 and 382 is generally used. This bottle is composed of a flat-bottomed balloon (fig. 381), the neck of which is closed by a cork pierced Fig. 381. Fig. 382. DETERMINATION OF EARTHS. 585 by two tubes—the tube abc, which opens in the upper part of the balloon, is drawn out at c, and the tube def, open at both ends, and descending to the bottom of the balloon. When the bottle rests on its bottom, it communicates with the external air by the tube abc ; when, on the contrary, it is inverted, as in fig. 382, the air enters by the tube def, and the water escapes by the tube abc in a fine stream, which may be directed on the several parts of the precipitate deposited on the filter. The rapidity of the stream may be increased by giving a greater length to the tube abc, thus increasing the difference of the level h, under the influence of which the water flows. Precipitates are, generally, more effectu- ally washed with hot than with cold water. Fig. 383 represents an apparatus by which the washing may be performed without constant manipulation on the part of the operator. This apparatus, which is frequently used for washing pre- cipitates in chemical analyses, is com- posed of a washing-bottle, the cork of which is traversed by a tube abed ar- ranged as seen in fig. 384. The filter being completely filled with water, the balloon A, also filled with water, is in- verted, so that the delicate and curved end d may dip to the distance of about 1 centimetre into the water of the filter: it is kept in this position by means of a stand. The pressure of the atmosphere acts on the liquid of the small lateral tube be, and also on the level of the liquid in the filter, and consequently on the water of the tube abd. The water of the bottle receives an impulse from the weight of the liquid column comprisecMbetween the level of the liquid in the filter and that of the liquid in the lateral tube cb ; but the lateral tube cb being very small, capillary action prevents the air from entering it, and equals the pressure of a small column of water. The water will therefore not flow from the washing-bottle, as long as the capillary action in ab surpasses the hydro- static pressure exerted by the column h. But in proportion as the water escapes from the filter, its level falls, the height of the column li increases, and this very soon overcoming the capillary action in cb, water will flow from the bottle into the filter, air will pass in by the lateral tube cb, and a new equilibrium will be established in consequence of the rise of the level in the filter. By means of this apparatus, the liquid is kept at nearly a constant height in the filter, and the Fig. 383. Fig. 384. 586 DETERMINATION OF EARTHS. upper stratum is always pure water, which is a condition very favourable to efficacious washing. When the quantity of gelatinous precipi- tate is considerable, it is almost impractica- ble to wash it in an ordinary filter, and then it is advisable to employ the arrangement represented in fig. 385. The large opening of the tubulated bell-glass A is closed with a doubled sheet of filtering-paper, kept in its place by a cloth tied around the border of the bell-glass. The bell-glass being placed on a stand over a dish, the liquid holding the precipitate in suspension is gradually poured into it. When the whole of it has been introduced into the bell-glass, a long tube ab is filled to the opening a, through which the water for washing is poured. A large pervious surface is thus offered for filtration, which takes place through a precipitate forming a layer of equal thickness, and under the pressure of a co- lumn of water which may be increased at will by increasing the length of the tube ab. The washing may be made continuous by passing into the bottle a curved tube ab fit- ted to the lower aperture of a Mariotte’s bottle B; the level of the liquid is thus kept at a constant height in the tube ab, and a continuous washing is effected under very favourable circumstances, because the pure water, arriving slowly from above, has no tendency to mix with the inferior strata which have become impure by their contact with the precipitate. Fig. 385. § 620. In connection with the particular study of the alkaline, alkalino-earthy, and earthy metals, we shall enter with some mi- nuteness into the description of the manufacture of several im- portant products which contain the compounds of these metals, namely, the manufacture of gunpowder, that of lime and mortars used in building, the manufacture of glass, and of earthenware. 587 GUNPOWDER. § 621. By mixing intimately saltpetre with charcoal or with sulphur, we obtain substances which, when subjected to a high temperature, deflagrate and suddenly develop a large volume of gas. When the combustion takes place in a contracted space, considerable pressure is exerted on the surrounding walls of this space, and if one of these be movable, it may be projected with more or less force. If, for example, 1 equivalent of nitre K0,N05 is mixed with 1 equivalent of carbon, there are produced, by detonation, 1 equiva- lent of carbonate of potassa, 1 equivalent of nitrogen, and 3 equiva- lents of oxygen: K0,N05+C=K0,C03+N+30; 2 volumes of nitrogen and 3 of oxygen will therefore be dis- engaged. We may calculate by approximation the volume of gas developed by one volume of the detonating mixture. 1 equivalent of nitrate of potassa weighing 1264.3, and 1 equivalent of carbon weighing 25.0, the weight of the mixture will therefore be 1339.3. Assum- ing that this pulverized mixture occupies the same volume as an equal weight of water, we can admit that a weight 1339.3 gm. of the mixture will occupy a volume of 1.339 lit. Now, this weight of the mixture develops 1 equiv. = 175 of nitrogen, and 3 equiv. = 300 of oxygen. 1 lit. of nitrogen at 32°, under a pressure of 0.760 m.weighs 1.257 gm. 1 “ of oxygen “ “ “ “ 1.429 The volume occupied by the nitrogen at 32°, and under a pressure of 0.760 m., will be given by the proportion, 1.257 : 1.000 : : 175: whence x =139.2 lit. The volume occupied by the disengaged oxygen under the same circumstances will be deduced from the proportion, 1.429 :1.000 :: 300 : y, whence y =209.9 lit. Thus a volume of detonating mixture represented by 1.339 lit,., yields 349.1 lit. of gas at 32°, and under a pressure of 0.760 m.: a volume 253 times greater than that of the explosive substance. The volume of gas, at the moment of development, is really much larger than we have just found, being strongly dilated by the high temperature produced by the combustion ; and we may safely admit that the expansion is at least three times greater than that 588 GUNPOWDER. given by calculation, when the gas was supposed to have a tem- perature of 32°. If 1 equivalent of nitrate of potassa is mixed with 2 equivalents of carbon, then 1 equivalent of carbonate of potassa, 1 equivalent of nitrogen, 1 of carbonic acid, and 1 of oxygen are formed: K0,N05+2C=K0,C03+N+C03+0. The equivalent of carbonic acid being represented by 2 volumes, it will be seen that 5 volumes of gas are still disengaged; that is, that the expansion is the same as in the preceding case. The projectile force may, however, be greater, if a high temperature be developed during the combustion. Lastly, if 4 equivalents of carbon are added to 1 equivalent of nitre, then 1 equivalent of nitrogen and 3 equivalents of oxide of carbon are disengaged : K0,N05+4C=K0,C03+N+3C0. 1 volume of oxide of carbon containing only a | volume of oxy- gen, it is evident that 6 volumes of oxide of carbon will be de- veloped : the gaseous volume will therefore be equal to 8. Thus, there will be a greater production of gas than in the two preceding cases. The projectile force might, however, be less, if the heat developed be not so great. Moreover, in the mixture we have just supposed, a great portion of the carbon does not ignite. Mixtures of nitre and sulphur also produce, by detonation, con- siderable volumes of gas. Thus, a mixture of 1 equivalent of nitre and 1 equivalent of sulphur yields 1 equivalent of sulphate of potassa, 1 equivalent of nitrogen, and 2 equivalents of oxygen: K0,N05+S=K0,S03+N+20; 4 volumes of.gas will therefore be formed. With 1 equivalent of nitre and 2 equivalents of sulphur we have K0,N05+2S=K0,S03+N+S03, that is, again 4 volumes of gas; for the equiv. of sulphurous acid is represented by 2 volumes. A mixture of 1 equiv. of nitre with 4 equivs. of sulphur gives K0,N05+4S=KS+N+3S03; 2 volumes of nitrogen and 6 volumes of sulphurous acid will therefore be disengaged; in all, 8 volumes of gas. In fact, how- ever, the gaseous volume is less considerable, owing to the incom- plete combustion of the sulphur. Mixtures of nitre and carbon generally produce a greater volume of gas than mixtures of nitre and sulphur ; but the latter have the advantage of being more combustible. GUNPOWDER. 589 § 622. Experiments have proved that the mixtures possessing the greatest projectile force consist of nitre, carbon, and sulphur. A mixture of 1 eq. of nitre 1264 66.0 1 “ sulphur 200 10.5 6 “ carbon 450 23.5 1914 100.0 gives KO,No5+S+6C=KS+N+6CO ; that is, 14 volumes of gas. But in reality the gaseous volume is less considerable, because a large portion of the carbon escapes combustion, and the temperature does not rise very high. The following mixture possesses a greater projectile force: 1 eq. of nitre 1264 74.8 1 “ sulphur 200 11.9 3 “ carbon 225 13.3 1689 llO We then have K0,N05+S+3C=KS+N+3C02, with the disengagement of 8 volumes of gas. We may calculate by approximation the volume of gas pro- duced by a volume 1 of this mixture. Let us again admit that the mixture occupies the same volume as an equal weight of water. We shall say that 1689 gm. of the mixture, or a volume of 1.689 lit. disengages 175 gm. of nitrogen = 139.2 lit., and 825 gm. of carbonic acid = 417.3 lit. ; total gaseous volume = 556.5 lit. A volume 1 of the detonating mixture will therefore produce 329 times its volume of gas at 32° and under a pressure 0.760 m. § 623. The numerous experiments made in all countries to dis- cover empirically the best composition for powder, show that it should be as approximate as possible to that just now theoretically developed. In France, three different compositions are in use: For war powder Saltpetre 75.0 Sulphur 12.5 Charcoal 12.5 100.0 For sporting powder...Saltpetre 76.9 Sulphur 9.6 Charcoal 13.5 100.0 For blasting powder...Saltpetre 62.0 Sulphur 20.0 Charcoal 18.0 100.0 590 GUNPOWDER. Prussian war powder shows the following composition : Saltpetre 75.0 Sulphur 11.5 Charcoal 13.5 100.0 English and Austrian war powder : Saltpetre 75.0 Sulphur 10.0 Charcoal 15.0 100.0 Swedish war powder: Saltpetre 75.0 Sulphur 16.0 Charcoal 9.0 100.0 Chinese powder: Saltpetre 75.7 Sulphur 14.4 Charcoal 9.9 100.0 French blasting powder is the only one which differs remarkably from the theoretical composition just indicated: this is, because a great projectile force is not required; and the government, which imposes a considerable tax on sporting powder, endeavours to manufacture a blasting powder such that it cannot be substituted for the former. This powder has, indeed, less strength, and fouls the gun very rapidly. § 624. Powder should satisfy several conditions, which vary ac- cording to the weapon in which it is to be used. When it is very explosive, and the explosion of the charge is instantaneous, the reaction on the walls of the weapon is sudden and violent, fre- quently causing the weapon to burst: the powder is then said to be too explosive. If the powder is not sufficiently explosive, the projectile is thrown from the weapon before all the charge is burned; a portion of the latter, therefore, is uselessly inserted and wasted. The powder most suitable for any given weapon is that which, burning perfectly whilst the projectile passes through the chamber of the piece, communicates to it, gradually, and not in- stantaneously, the whole projectile force of which it is capable, Hence the quality of the powder must vary according to the nature of the piece in which it is used. With equal quantities of the in- gredients, the quality of the powder can still be altered, by using charcoal more or less carbonized, by giving the substance a greater or less degree of compactness, or by varying the size of the grain. GUNPOWDER. 591 Before proceeding to study the manufacture of the various kinds of powder, we shall investigate the preparation of its primary com- ponents. Choice and preparation of the primary components. § 625. Saltpetre. -The saltpetre used in the manufacture of powder is the refined nitre of which we spoke (§ 450). This nitre is remarkably pure, and rarely contains more than 2 or 3 thousandths of sea-salt. It comes from the refinery in very small crystalline grains, and in this state is used in the manufacture of powder. § 626. Sulphur.—Powder-mills purchase the refined sulphur in rolls. It must be reduced to an impalpable powder, which is effected in wooden drums (figs. 386 and 387) having on the inside wooden brackets a, b, arranged along the edges of the cylinder. These drums are cylindrical, and about 1.10 m. long, with a dia- meter of about 1.15 m.: they revolve on a horizontal iron axis OO'. Through a door abed, which is fur- nished with ironhandles m' m, the material is introduced. Pulveriza- tion is effect- ed by means of small brass balls, of about 5 or 8 millimetres in diameter, of which each drum contains 150 kilogrammes: 30 or 40 kilogrammes of sulphur are added, and the drum is made to revolve for 6 hours, during which time the balls, rolling with the sulphur, crush it and reduce it to extreme fineness. In order to withdraw the sulphur, the door of the drum is removed, and replaced by a similar door abed, the panels of which are of wire-gauze (fig. 388); by causing the drum to revolve 5 or 6 times, the sulphur escapes through this door, leaving the balls in the drum. The powdered sulphur is sifted in a bolting-machine, similar to that used for bolting flour; the particles which have not been suf- ficiently pulverized are thus separated, as wTell as any small grains of sand, which might occasion accidents in the manufacture of the powder. §627. The charcoal destined for the fabrication of powder must be most carefully selected. All kinds of wood are not suitable for the preparation of this charcoal: the tender and light woods, which Fig. 386. Fig. 387. Eig. 388. 592 GUNPOWDER. yield a friable, porous charcoal, leaving very little ash, are pre- ferred. The woods most esteemed are the black alder and spindle-tree: poplar and chestnut may also be used. Hemp-stalks likewise yield a very good charcoal. The wood of the black alder is exclusively used in France. The branches of about 15 or 20 millimetres in diameter are preferred; and if larger branches are used, they are first split. The bark is always removed, as it gives too much ash. These branches are cut into lengths of from 1.5 to 2 metres, and tied in bundles weighing from 12 to 15 kilogrammes. The carbonization is never effected in kilns, as common charcoal is made, but in pits or in cylinders. § 628. Carbonization in pits.—Cylindrical pits, about 1.5 m. in diameter and 1.2 m. in depth, are excavated in the earth and lined with bricks, and filled with the wood, cut into pieces of 0.30 m. in length, until the heap rises to the height of a few decimetres above the mouth of the pit. Fire is communicated through a hole at the bottom; and as the combustion advances, the branches are raised with a fork, so as to allow the fire to be regularly distributed. The pile gradually sinks, and fresh wTood must be added to keep the pit full. When a flame is no longer seen, the mouth of the pit is hermetically closed by a sheet-iron lid, and the carbonization is then finished without access of air. The pit remains closed for three or four days, in order entirely to extinguish and cool the charcoal. It is then opened, the charcoal removed, and conveyed to the sort- ing-room, where it is most carefully sorted by hand; such branches as have not been sufficiently carbonized and the half-burnt pieces are rejected, as also those which are too much carbonized, and therefore would make bad powder. The good charcoal should be used immediately, as it sensibly deteriorates by exposure to the moist air. By carbonization in pits, about 18 to 20 per cent, of charcoal are obtained. § 629. Carbonization in cylinders.—This process yields a much larger proportion of charcoal; its quality is also more constant and uniform, because the fire can be regulated at will, and the carbonization can be arrested at the proper moment. Fig. 389. Fig. 390. GUNPOWDER. 593 The cylinders C, C (figs. 389 and 390) are arranged in pairs in the same furnace: they are made of cast iron, having 2 metres in length and about 0.70 m. in diameter. One end of the cylinder is closed by a cast-iron lid, having four circular openings, through which pass four sheet-iron tubes, as pq, mn. Three of these tubes, which serve for the introduction of sticks of wood, are closed ex- ternally with wooden plugs, which can be withdrawn from time to time, so as to observe the progress of the carbonization. The fourth is open, and gives exit to the gases which are evolved dur- ing the process. A curved copper tube no is fitted to one end of it, opening above a funnel v, which communicates with a horizontal canal T, ranging along the furnace and opening into the chimney. There are generally twelve furnaces arranged in the same mason- work. The combustible is placed on the grate d, the flame and smoke ascend between the two cylinders, surround them, and descend by vertical pipes u and u' into a horizontal canal VV', which extends under all the furnaces, and opens into a chimney built in the mid- dle of the room. The heat around each cylinder is regulated by registers r and r\ in the vertical pipes u and u'. The part abc of the cylinders, which is more immediately exposed to the action of the fire, is covered with a luting of broken tiles and clay. The maximum of temperature is thus found at the top of the cylinders, favouring greatly the progress of the operation. The sticks of wood to be carbonized are about 1.5 m. in length: when the cylinders are filled with them, the movable end fghi is replaced. This end is made of two sheets of iron, the space be- tween which is filled with ashes: assay sticks are then introduced into the tubes pq, mn. When the cylinders are charged, fire is kindled on the grate: turf is the fuel generally used. Active decomposition of the wood does not begin under four or five hours. The progress of the operation is estimated by the quantity and colour of the smoke which escapes from the pipe no. When the carbonization is sup- posed to be advancing, the assay-sticks are withdrawn, and an opinion formed from their appearance of the progress of decompo- sition in the various parts of the cylinders: if it be more advanced in some parts than in others, the combustible is pushed to the side where the carbonization is slowest. The heat is also regulated by the registers r and r'. In 11 or 12 hours, no vapour escapes any longer from the pipe no; the operation is then terminated, the registers are closed, and the carbonization is completed without further aid. On the following day, the charcoal is withdrawn and placed in sheet-iron extinguishers, (6touffoirs.) Carbonization in cylinders yields from 35 to 40 per cent, of charcoal, which is sorted by hand, and broken into small pieces. The carbonization is not carried so far when the charcoal is in- 594 GUNPOWDER. tended for sporting powder: it is then withdrawn in the state of red charcoal (charbon roux); its colour then is brown. For war powder the carbonization is pushed further, to the state of black charcoal* (charbon noir.) Powder made with red charcoal would be too explosive for muskets or artillery. MANUFACTURE OF POWDER. § 630. The principal processes of the many used in the manufac- ture of powder are the following: 1st. Powder-mills with stampers. 2d. The pulverizing drum and hydraulic press, called also the revolutionary process. 3d. Powder-mills with edge-stones. 4th. The Bernese, or process of Champy, by which round pow- der is made. 1st. Powder-mills with stampers. § 631. These are the oldest; they make good powder, and are still in use in France for the manufacture of war powder. Fig. 391. A battery of pestles (fig. 391) is generally composed of two parallel rows of ten pestles, each of which falls into its own mor- tar. The pestles are made of square pieces of wood (fig. 392) 2.5 m. in length and 0.10 m. square, terminating in a rounded part which fits into a pyriform brass box a. Their weight is about 40 kilogrammes. The mortars are hollowed out of a large block of chestnut-wood * Called also “ distilled charcoal.”—Trans. GUNPOWDER. 595 mm\ about 0.60 m. square; their shape is that represented in fig. 393. The material, under the blows of the pestle, rises along the sides, but soon falls to the bottom, on account of the shape of the mortar. From the position of the latter, the blows of the pestle, falling on the length of the grain of the wood, would very soon destroy the mortars; and to prevent this, a piece of hard wood z is inserted at the bottom. Each row of pestles is set in motion by its own shaft OAB, furnished with cogs C (fig. 391). The two parallel shafts are thrown into gear by two wooden wheels L and the wheel IK, attached to the shaft of the water-wheel, which moves the whole machine. The shafts OAB thus make four revolutions while the water-wheel makes but one. Each pestle has a crosspiece m (fig. 392), by which the cogs ele- vate it. The latter are arranged spirally around the shafts, so that the same number of pestles is always raised, and the machine has constantly the same work to do. Horizontal crossbeams uu', w' regulate the movements of the pestles. The lower crossbeam uu' also serves to hold the pestles suspended while the workman is engaged with the mortars. To suspend the pestle, the workman raises it until the hole o is above the crossbeam mm', and then inserts into this hole a pin s which rests on the crossbeam. The charge of each mortar is 10 kilogrammes. On the one hand, 1.25 kilog. of charcoal in pieces are weighed, and placed in a small wooden bucket; and on the other, 1.25 kilog. of sulphur and 7.50 kilog. of saltpetre, which are likewise placed in another bucket, are accurately weighed and carried to the mill. The charcoal is placed in each mortar, and wetted with one litre of water: the pins s are then removed, and the pestles lowered. They are instantly set to work, by letting water on the wheel. The pestles are allowed to stamp the charcoal for 30 minutes; the wheel is then stopped, the pestles fastened with the pins, and the sulphur and saltpetre placed in the mortars. The three substances are then mixed with the hand, moistened with a half-litre of water, and again mixed. The pestles are lowered, and the battery set in motion. The workman ascertains in a few moments if the mate- rial is behaving properly under the pestles. If the external tem- perature is too great, an additional quantity of water sometimes becomes necessary. After stamping for half an hour, the material in the mortar is changed. After the battery is stopped and the pestles are fast- Fig. 392. Fig. 393. 596 GUNPOWDEK. ened by the pins, the workman removes, with a copper ladle (fig. 394), the material from the first mortar and deposits it in a small wooden box called a trough. In the same way he removes the material from the second mortar, depositing it in the first, which he has just emptied: the material in the third is received by the second, that of the fourth by the third, and that of the fifth by the fourth. Lastly, the fifth mortar receives the material taken from the first. Each workman superintending only five mortars, four men are required for the battery. The pestles are set in motion, and allowed to stamp for an hour: a second exchange is then made precisely like the first. This is continued for 12 hours, an exchange being made every hour; from time to time, the material is moistened, in order to preserve the proper degree of dampness. After the last exchange, the battery is allowed to work for two hours, to give additional consistency to the material. During these fourteen hours, the material has received about 30,000 blows with the pestle: this number has been ascertained to be necessary to give the powder the proper degree of density. Wherever it has been reduced, a powder has resulted which would not bear transportation. § 632. The material, when removed from the mortars, is de- posited in wooden buckets, called tubs, and sent to the graining- house. The object of graining powder is to bring the material which comes from the mill into the shape of grains of a given size. This process is effected in sieves (fig. 395) of 0.60 m. in diameter, the bottom of which is made of leather pierced with circular holes. The sieves are named differently according to the diameter of their holes. The holes of the guillaume are from 5 to 10 millimetres in diameter. Those of the common powder sieve are 4 millimetres, while the diameter of those of the musket powder sieve is only 2 millimetres. Around the graining-house are arranged wooden boxes called maies (fig. 396), through which wooden bars ab are passed for the support of the sieves. The workman places the material in the guil- laume, rests this sieve on the bar ab, and moves it backward and forward in a horizontal position. A portion of the material passes through the holes; but as the motion alone would not suffice to drive all through the apertures, the larger pieces are broken by placing in the sieve a lenticular disk of hard wood t (fig. 395), called a cake. This cake is 21 centimetres in diameter, 55 Fig. 394. Fig. 395. Fig. 39(3. GUNPOWDER. 597 millimetres in thickness at the centre, and only 45 at the circum- ference. The material which has passed through this sieve is deposited in the common powder or in the musket-powder sieve, according to the quantity to be manufactured. This sieve is handled in the same way as the first, the cake always being used. The mate- rial which has passed through the sieve is composed of grains of the requisite size, of smaller grains, and of dust, or meal. The separation of common powder is effected by pouring it over the musket-powder sieve, without using the cake; the fine grains fall- ing through, leaving the common powder to remain in the sieve. The material which has passed through being composed of musket powder and meal, is passed through another sieve, the holes of which are smaller than those of the musket-powder sieve, thus sepa- rating the meal. Still finer grains may be obtained of a uniform size by using a finer sieve: a certain quantity is thus separated, and sold as sporting powder. The meal is carried back to the mill and again stamped, after having been moistened with water. § 633. The grains of powder thus obtained are small angular fragments: they are called angular or green powder (poudre verte). This powder must be dried or glazed, for which two processes are adopted: drying by exposure to the air, or by the application of artificial heat. Drying in the open air can of course only be done in fine wea- ther. The damp powder is spread, 3 or 4 millimetres in thickness on cloths, which are arranged on tables, along a wall having a southern exposure and sheltered from the north. A different surface of the grains of powder is occasionally exposed by stirring. The drying is pretty rapidly effected during the summer, but in spring and autumn is much slower. Drying by artificial heat has the advantage of being more regu- lar, and capable of being performed at all seasons. The powder is spread in thin layers, on cloths stretched over wooden frames, between which a current of warm air passes, heated by means of long tubes in wooden closets, through which a current of hot water is constantly passing. 2d. Revolving process. § 634. This process was adopted for the manufacture of a large quantity of powder in a short time; it is now, however, no longer employed, as the quantity of powder obtained is of an inferior kind. In the first place, the nitre alone, and then the mixture of sul- phur and charcoal, were finely powdered in drums with brass halls. The materials were then properly proportioned, and the mixture deposited in revolving drums, containing halls of tin, which effected an intimate admixture of the ingredients. A moist cloth was then placed on a square copper plate, over 598 GUNPOWDER. which was arranged a wooden frame, intended to hold the powder: this frame was then filled with the preceding mixture. The frame was then removed, and the mixture covered with a second damp cloth, on which a second copper plate was placed, another moist cloth and another layer of material was added, and so on. When the mass was sufficiently thick, it was compressed by a hydraulic press, thus diffusing the water in the cloths through the material, and moistening it uniformly. The cakes arising from this operation were exposed for some time to the air, to dry them, and then grained in the ordinary manner. 3d. Mills with edge-stones. § 635. This process is applied in France to the manufacture of sporting powder, and produces a very compact and superior article. The proportions adopted are the following: Saltpetre 80.0 or 76.9 Sulphur 10.0 ... 9.6 Charcoal 14.0 ... 13.5 104.0 ... 100.0 The charcoal and sulphur are first reduced to powder in revolv- ing drums with brass balls, as stated in §626; but the charcoal, being pulverized with more difficulty, requires a longer time for this operation. The charcoal is therefore first pulverized alone, and the sulphur is not added for some time. 21 kilogrammes of red charcoal, as it comes from the sorting-room, are allowed to revolve for 12 hours in the drum; then 15 kilogrammes of sulphur, in pieces, are added, and the operation is continued for 6 hours. The mixture is then withdrawn in the state of an impalpable and per- fectly homogeneous mixture, by operating as was stated for the pulverization of sulphur alone. The proper quantity of saltpetre is then added to the binary compound, and the whole deposited in another drum, called the mixing-drum in which the three ingredients are intimately mixed together. The drum is divided into 3 compart- ments by oak partitions, which are kept asunder by 12 wooden ribs on the interior ; the latter, being covered with a piece of cow- hide, form a cylinder, the convex surface of which is of leather. The material is introduced and withdrawn through a door held in its place by copper screws. In order to effect the mixing, 60 kilo- grammes of bronze balls of 5 millimetres in diameter, and 26 kilog. of mixture, are allowed to revolve in the drum for 12 hours, at the rate of 25 or 30 revolutions per minute. The material is then withdrawn, as from the other drum, by substituting for the door another of which the panels are of wire-gauze. Fifty kilog. of this mixture are then placed in a box called a mate, GUNPOWDER. 599 and continually stirred, while one litre of water is poured on it with a watering-pot, the rose of which is pierced with very fine holes. In summer, double this quantity of water is sometimes used. It is then carried to the grinding-mill. This mill consists of two vertical cast-iron millstones M, M (fig. 397), weighing 5500 kilog. and resting on a cast-iron platform AB, supported by solid ma- son-work. The diameter of these stones is 1.50 m.; their thickness at their cir- cumference 0.50 m. The diameter of the horizontal platform is 2.0 m. Each stone is traversed by an iron axis CDC', which runs on one side into a vertical cast-iron shaft EF, and on the other into a framework GHIH'G’, firmly fastened to the vertical shaft EF. A horizontal shaft KL, be- low the floor and commu- nicating with the water- wheel, moves the vertical shaft by means of conical gearing. Two small wooden scrapers s,s', shod with copper, are fastened on iron arms t, t' fixed to the movable shaft, and follow the course of the stones. Their object is to push into the track of the wheels any portions of the material which the pressure might drive toward the edges of the platform. Scratchers r, r', shod with copper, rub continually against each wheel, and detach any material which may adhere to it. Fifty kilog. of material are spread evenly on the platform : the mill is then set in motion, the rapidity increasing gradually until the shaft EF makes about 8 revolutions per minute. In about an hour the greater portion of the water added has evaporated, through the considerable rise of temperature during the operation; the material has become dry, and requires again to be moistened. In order to effect a uniform moistening throughout the whole mass, a receiver Y (fig. 398), terminating in a horizontal tube ab pierced with very small holes, is fastened behind one of the millstones: into this receiver 1 litre of water is poured, and allowed to flow at pleasure, by means of a stopcock in the vertical tube. The workman also cleans the track of the millstones with a copper scraper, with- out stopping the mill. After an hour’s operation, he stops Fig. 397. Fig. 398. 600 GUNPOWDER. the mill; and having very accurately examined the material along the track, he turns the mill very slowly, so as to require 8 or 10 minutes for a single revolution of the platform. The stones, thus resting for a long time on one point, forcibly compress the material, thus effecting a great density. This terminates the operation : the cake is removed and sent to the graining-house. § 636. The graining-machine now used consists of a wooden frame AB (fig. 399), of octagonal form, and 2.50 m. in diameter, suspended horizontally by 8 ropes. In the centre is a collar, shod with copper, through which passes the curved iron defgh, called a signole, making at the same time a part of the vertical axis KH, of which the upper end H turns in a socket set in a joint. The lower end K is furnished with a horizontal bevelled cog-wheel, running into a vertical bevelled cog-wheel on the axis MN, by means of which the whole machine is set in motion. It is evident that during the rotary movement of the axis KII, the signole com- municates a similar motion to the frame AB. Fig. 399. Eight graining-machines, having 3 compartments each (fig. 400), are fixed on the frame. The bottom AB of the first is of walnut, 2 centimetres in thickness, pierced with small holes rimed out from below. A wooden plate C, weighing 2 kilog., and made of the service- tree, rests on this bottom; and, at the two opposite points, are two openings of 1 decimeter in width, to which are fitted two in- clined copper planes, in the shape of small troughs, the lower extremities of which touch the surface of the second bottom FGr. This bottom, 3 centimetres distant from the first, is made of GUNPOWDER, 601 metallic gauze, the meshes of which al- low sporting powder to pass through. Lastly, at 3 centim. from this second bottom, there is a third III of silk bolt- ing-cloth, intended to receive the meal powder. The lower part of the graining- machine, which rests on the surface of the frame, is covered with leather: the upper surface is covered by cloth fur- nished wTith a leather tube E for the introduction of the material. On the side of the sieve there are two openings 0, O', each also provided with a lea- ther tube, intended to convey the grained and meal powder into small light boxes X, X'. The graining-sieve being arranged on the suspended frame, the mate- rial is introduced by the small troughs Y, Y', and the frame is set in motion. The plate in the sieve then moving circularly over the wooden plate, breaks up the material, which, after pass- ing through the holes of the wooden bottom AB, falls on the me- tallic gauze FG. All that part which is fine enough to escape through the meshes falls, and the larger particles which remain, on meeting the inclined plane of the troughs, ascend along these troughs to the first plate, by virtue of its circular motion, and are again subjected to the action of the plate. While this is taking place in the two upper compartments, the mixture of grained and meal powder, which has passed through the metallic gauze, falls on the silk bolter, and is bolted. The meal, which passes through the bolting-cloth, falls on the leather bottom of the frame, whence it is conveyed by a leather tube into a small tight box, while the clean grains, which remain on the silk, escape by an opposite opening and fall into a small barrel. The meal is sent to the mill, to be again pulverized: the cake produced is grained as before. § 637. Sporting powder undergoes another operation, called glazing, the object of which is to give it a polished and brilliant surface, which insures its preservation and renders it more dense. The glazing-machine is a cylindrical wooden drum (fig. 401) of 2.70 m. in length, and 1.20 in. in diameter, divided internally into five compartments, by intermediate partitions. Each com- partment has its own door. A wooden shaft traverses the drum, and receives its movement by proper gearing: the inside of the drum contains, like the mixing-machines, 12 projecting wooden ribs. Above the glazing-machine is a large hopper, divided into five Vol. I.—3 A Fig. 400. 602 GUNPOWDER. Fig. 401. compartments corresponding to those of the drum, each of which terminates in a leather tube serving to convey the material into a barrel beneath. 100 kilog. of grains being placed in each com- partment of the drum, it is made to revolve slowly for the first 12 hours. The powder thus rolls continually on itself: the wooden ribs renewing the points of contact, the grains wear off their angles against each other, and become polished. The drum is then made to revolve faster, and in 36 or 40 hours the glazing is terminated. Powder taken from the glazing-machine is dried as usual. Glazing gives powder greater density, but diminishes its inflam- mability : it must, therefore, not be pushed too far, and should be stopped when the grains are sufficiently hard to bear transporta- tion, and are free from dust. § 638. A quality of sporting powder, superior to the preced- ing, is obtained by again grinding the grained powder by the process just described, subjecting it to a second pulverization and graining. This powder, after glazing, is known by the name of poudre royale (royal powder), and is superior to the best foreign powders. 4th. Manufacture of Round Powder by the Bernese process, or the process of Cliampy. § 639. Blasting-powder is made in France by a peculiar pro- cess, first used at Berne, whence it has obtained the name of Bernese process. It is also called the process of Ohampy, in honour of the inspector of powder, to whom great improvements in its working are due. This process is also applied to the manu- facture of cannon and musket powder. For blasting powder, the more highly-burned charcoal, which is unfit for other powder, is used: the great degree of calcination is in this case not injurious to the quality of the powder, as blasting powder should not possess too great an inflammability. Six different operations may be distinguished in the manufac- GUNPOWDER. 603 ture: pulverization, mixing, graining, equalization, glazing, and drying. The pulverization is effected by bronze balls in iron drums, ex- actly as has been previously described, with the only difference that at the same time balls of 4.5 m. in diameter, and some vary- ing from 7 to 15 mm. are used, the charcoal being more difficult to grind. The drum contains 120 kilog. of these balls, with 30 kilog. of sulphur and 27 of charcoal, which is the proportion for 150 kilog. of powder. The door is closed, and the drum made to revolve for 4 hours, at the rate of from 25 to 28 revolutions per minute: the binary mixture, being then sufficiently ground, is removed from the drum. The further mixture is then made as follows:—14.25 kilog. of the substance taken from the drum, exactly weighed, are placed in a barrel, and 23.25 kilog. of saltpetre added. Each barrel then contains 37.50 kilog. of the compound, viz: Saltpetre 23.25 62.0 Sulphur 7.50 20.0 Charcoal 6.75 18.0 37.50 100.0 This compound is carried to the mixing-machine, which are leather drums, containing 60 kilog. of bronze balls of 4.5 mm. in diameter. The 37.50 kilog. of it are introduced, and the machine made to re- volve at the rate of 25 or 30 revolutions per minute. After four hours’ working, the compound is well mixed; the material is then conveyed into a maie, and placed in barrels to be carried to the graining-house. Fig. 402. § 640. The machine (fig. 402) used for the manufacture of round grains consists of two large oak drums AEGB, CI4FD, 1.75 m. in diameter, and 0.63 m. in height. Each of them has only one entire end BE, CF: the opposite end AG, HD being furnished 604 GUNPOWDER. with a circular opening U of 0.60 m. diameter in the centre. The two drums are traversed by the same iron axes 10, supported between two strong vertical beams by two copper chains. Two copper disks aabb', fixed on the iron axis, connect the transverse iron axis 10 with the ends EB and CF of the drum, while four strong cross-pieces, as AB, keep the whole steady. Each drum has a door M, of 0.35 m. by 0.60 m., closed with four copper screws, and used for introducing and withdrawing the material. All the lower part of the machine is surrounded by a large trough N, furnished with inclined copper planes, intended to receive the material when withdrawn by the doors, and conduct it into barrels placed beneath. The drum AEGB is used for graining, and the other CIIFD for glazing the powder. The outer periphery of the granulator AEGB is furnished with 12 small cleats x, x, x, which, during the movement of the drum, move, and cause a small wooden hammer p, fastened by a cord to the side of the trough N, to strike constantly on its surface, de- taching, by its blows, any portion of the material which might adhere to the drum. A copper watering-tube nu, 2 centim. in diameter, and 0.40 m. in length, having one side pierced with very minute holes, enters the granulating machine, a little above its axis, and communicates, by a curved copper tube nms, with a forcing-pump. This pump (fig. 403) is composed of a copper pump-tree P, in which a perfectly well-fitting piston moves: an iron rod tt', fastened to the upper part of the piston, works between two wooden uprights. The piston is set in motion by means of a winch and a rope which passes over a pulley fixed to the iron rod. The lower part of the pump-tree com- municates, on the one hand, with a reservoir of water, and, on the other, with the inject- ing-tube smnu; two stopcocks r, r' closing at will the communicating tubes. When the stopcock r is opened and the piston raised, the lower part of the pump fills with water: if this stopcock be closed and that at r' opened, the piston descends by its own weight, allowing the water to escape through the watering-tube smnu. In order to introduce a charge of the material, the workman removes the door M of the granulating-machine, and pours in 100 kilog. of powder already grained, called the nucleus (noyau), the origin of which will be hereafter explained; he replaces the door, and sets the machine in motion at the rate of 10 revolutions per minute. During this motion, the first sprinkling, of 5 per cent. Fig. 403. GUNPOWDER. 605 of water, is made; the fluid thus wetting the nucleus which occu- pies the lower part of the granulating-machine in the form of a fine rain, and the rotary motion of the drum constantly renewing the surface, all the grains are uniformly moistened. When the first sprinkling is over, he introduces, through the opening U, 50 kilog. of the mixture, as it comes from the mixing- machine, inserting 1 kilog. at a time with a wooden shovel, spread- ing it as evenly as possible in the drum. The movement of the machine rolling the damp grains constantly among the dry meal- powder, causes the latter to adhere to their surface, and each grain thus to increase by concentric layers. Immediately after, a second sprinkling is made, and then 50 kilog. of the ternary mixture are gradually added. After allowing the machine to revolve for a quarter of an hour, the workman ascer- tains if the meal powder is entirely absorbed ; he then empties the machine, by dropping the material into barrels placed beneath. These operations last from 35 to 40 minutes. § 641. The material, when taken from the machine, is com- posed of variously-sized grains, which require to be separated, or equalized. This is done by shaking the grains over two leather sieves; the first, called the equalizer (egalisoir), separates those grains which are too large, while the second, the subequalizer, allows those which are too fine to pass through. The holes in the equalizer are 3.4 mm. in diameter. The grains and irregular pieces which do not pass through are set aside; those which pass through are sifted on the sub-equalizer, the holes of which are 1.2 mm. in diameter. There remain on the latter sieve those grains the diameter of which are comprised between 1.2 mm. and 3.4 mm. and which are suitable for blasting : they are deposited in a barrel to undergo a subsequent operation. All which passes through the sub-equalizer is com- posed of grains smaller than 1.20 mm.: it is considered as a nucleus, because this grain need only be increased in the granulating-machine to make it of the proper size. As each operation yields the quan- tity of nucleus necessary for a succeeding operation, it is sufficient to obtain some for the first operation, for which the angular powder, of the size of musket powder prepared in the stamping-machine, is employed. The grains which are too large and the irregular pieces, which remained on the equalizer, are broken by means of the cake, and used as a nucleus for the succeeding operation. Blasting powder is glazed as well as sporting powder, in order to increase its density. This operation is effected in the second drum CTIFD. 200 kilog. of equalized grains are introduced, and it is turned for 4 hours: by direct experiment it is ascertained when the grain has acquired sufficient density. For this pur- pose 60 gm. of the glazed grains are poured into a graduated 606 GUNPOAVDER. test-glass: the grain is considered as sufficiently glazed when the level of the material rises to a certain division in the instru- ment. The glazed grain is dried in the ordinary way. § 642. Round war powder is manufactured by the same method, the usual proportion of the ingredients for Avar powder, 25 of salt- petre, 12.5 of sulphur, and 12.5 of charcoal being employed. Two kinds of equalized grains are separated : those of which the dia- meter is between 1.2mm. and 2.1 mm. constituting cannon powder; and musket powder, the diameter of the grains of Avhich varies from 1.0 mm. to 1.20 mm. METHODS OF TESTING THE FORCE OF POWDER. § 643. In the French powder-mills, powder is subjected to a series of experiments, intended to ascertain the physical qualities and the ballistic force of each kind of powder. War powder must fulfil the following conditions : The grain must be angular, hard, dry, and equal; the size vary- ing from 2.5 mm. to 1.4 mm. for cannon powder, and from 1.4 mm. to 0.6 mm. for musket powder. It should resist moderate pressure, and leave no dust when rubbed between the hands. The apparent density of powder is measured by a peculiar apparatus, called a gravimeter. It is a measure holding exactly 1 cubic decimetre : it is filled by means of a valved funnel, which fits thereon, and spreads the powder uniformly. The weight of the litre of pow- der, not heaped up, contained in this measure, is its gravimetric density. For war powder, this density is from 0.820 kilog. to 0.830 kilog. § 644. The ballistic force of powder is measured at the same by the eprouvette mortar, or testing-mortar, and the pendulum-gun or pendulum-test. The eprouvette-mortar is a cast iron mor- tar (fig. 404), the axis of which has an in- clination of 45°; its internal diameter is 191.2 mm. , By means of a bent funnel, 92 gm. of powder are introduced into the chamber a of the mortar, and a bronze ball of 189.5 mm. in diameter, weighing 29.4 kilog. is placed thereon. To be satisfactory, the ball must be thrown at least 220 metres. § 645. The pendulum-gun (fig. 405) is composed of two parts— the pendulum-gun AB, properly so called, and the receiver, or ballistic pendulum, CD. The pendulum-gun is made of a musket- barrel ab, the breech of which is replaced by a piece of iron sup- ported at the bottom of an iron frame omn, terminated at its upper part by two gudgeons o, shaped like knife-blades, forming a hori- Fig. 404. GUNPOWDER. 607 zontal axis. Below the pendulum is an axis mn, sustaining a movable mass of lead p, which may be made to slide on a horizontal rod, capable of being itself fixed at different heights. This mass is placed in such a position that the centre of oscilla- tion of the compound pendulum is over the axis of the gun, and over the vertical axis which passes through the centre of gravity. The pendulum-gun has a rod i, which drives a movable slider over a graduated arc cd: this slider marks the recoil of the pendulum. Eig. 405. The ballistic-pendulum is composed of a conical bronze box D suspended to an iron frame-work gh, which terminates itself, at the upper part, in a horizontal axis made of two knife-shaped gudgeons. This system therefore forms a second pendulum, as movable as the first. The axis of suspension of these pendulums should be accurately parallel. The box contains a mass of lead, into which the ball penetrates; it has also an appendage r, which moves on a graduated arc H, and drives a slider over the arc. The distance to which the slider is driven by the impinging of the ball measures the momentum communicated by the ball to the pendulum. The apparatus is so arranged that the centre of oscillation of the pen- dulum is over the axis of the conical box. Mathematical formulae permit a calculation of the initial rapidity of the ball, either by the space traversed by the slider of the ballistic pendulum, or by that of the slider of the gun-pendulum. 608 GUNPOWDER. The weight of each of these pendulums, when mounted, is 25 kilogrammes. The charge of powder in the gun weighs 10 gm.; the diameter of the ball is 16.3 mm. The initial rapidity of the ball, inferred from the two observations just indicated, should be 450 metres per second. For. sporting powder only 5 gm. are used: they should give the following initial degrees of rapidity: For fine powder 330 metres “ superfine 350 “ “ royal 375 “ In many French powder-mills, a cannon-pendulum,* arranged on exactly the same principles as the gun-pendulum, is used. § 646. War powder is likewise subjected, from time to time, to several other experiments, intended to test its hardness, its capa- bility of transportation, or the more or less rapid changes it under- goes by exposure to the moisture of the atmosphere. ANALYSIS OF POWDER. § 647. The analysis of powder is a tedious and delicate opera- tion, when the proportions and nature of its components are to be ascertained very exactly. The first operation is to determine the proportion of hygrometric water the powder contains; for which purpose a known weight of powder is exposed for several days in a dry vacuum, and the loss it experiences ascertained; or else the substance is placed in a U-shaped tube, kept at a temperature of 60° or 70° and traversed by a current of dry air. The appa- ratus is arranged as described (§ 261) for oxalic acid. Ten gm. of dry powder are then treated with hot water, which dissolves the nitrate of potassa. The insoluble residue, composed of sulphur and charcoal, is collected on a small filter, which has been previously dried and weighed. When this residue has been properly washed, it is dried with the filter at a moderate tempera- ture, and weighed: by subtracting from this weight that of the filter above, the weight of the sulphur and charcoal is obtained. After separating, as carefully and completely as possible, the sub- stance from the filter, it is again weighed in a small bottle and treated with sulphuret of carbon, which may be mixed with an equal volume of ether, without too much impairing its solvent power. The charcoal which remains isolated is collected on a small filter previously dried, and weighed after a second desiccation, after hav- * This method, the invention of the Citizen liegnier, made in the year VII. of the republic, still bears his name.—Translator. GUNPOWDER. 609 ing been well washed in a mixture of sulphuret of carbon and ether. The weight of the sulphur is thus obtained by the difference. It may, however, also be weighed directly after evaporating, at a low temperature, the solvent which contains it. The charcoal of the powder is not pure carbon : it contains, as the carbonization is always imperfect, a considerable proportion of oxygen and hydro- gen ; but, as the chemical nature of the charcoal exerts great in- fluence on the quality of the powder, it is important, in an accu- rate analysis, to ascertain the amount of carbon exactly. Char- coal is analyzed, like organic substances in general, according to the process explained (§ 260) for the analysis of oxalic acid. An idea may be formed of the composition of charcoal by the following numbers, obtained from the analysis of the red charcoal used in the manufacture of sporting powder: Carbon 71.42 Hydrogen 4.85 Oxygen and Nitrogen 22.91 Ashes 0.82 100.00 § 648. The quantity of sulphur contained in powder may also be determined by operating directly on the powder itself. To effect this, 10 grammes of dry powder are dissolved in a small quantity of hot water, nitric acid is added, and, after allowing the fluid to boil, small quantities of chlorate of potassa are gradually introduced. Under the influence of these oxidizing agents, the sulphur is dissolved in the state of sulphuric acid, which is precipitated, after the liquid is filtered, by chloride of barium. The precipitate is allowed to settle, the clear liquid poured on a filter, and the pre- cipitate, after being boiled for a few moments with chlorohydric acid, to dissolve the nitrates it might contain (§ 536), is collected on the same filter and weighed after calcination. Ten grammes of dry powder may also be mixed with an equal weight of nitrate of potassa and four or five times its weight of chloride of sodium: the mixture being thrown, in small quanti- ties at a time, into a platinum crucible, deflagrates slowly, without any loss of the material. It is subsequently treated with water, and the sulphuric acid is precipitated by chloride of barium, after supersaturating the liquid with chlorohydric acid. It has also been proposed to dissolve the sulphur of the mixture of sulphur and charcoal, by a solution of monosulphide of so- dium, or of hyposulphite of soda; but this process is useless, be- cause the charcoal, being always considerably attacked by these alkaline liquids, gives off a peculiar acid, called ulmic acid. § 649. Very frequently, only the quantity of saltpetre contained in powder is to be ascertained. This is easily done by treating 610 GUNPOWDER. 50 grammes of powder with 200 grammes of hot water, and filter- ing the liquid into a test-glass having a mark at the level corre- sponding to 500 cubic centimetres. The material is washed with water on a filter, until the filtrate reaches the level. The liquid is then cooled to 60°, the contraction it undergoes by cooling being compensated by the addition of a small quantity of water : it is then well shaken to render it homogeneous, and a peculiar areo- meter, graduated so that its level will mark immediately the hundredths of nitrate of potassa contained in the 50 grammes of powder, is dipped in. In this manner the proportion of ni- trate of potassa may easily be determined, to very nearly a half- hundredth. 611 LIME AND MORTARS. BUILDING MATERIALS. § 650. The material used in building is of two kinds: natural, or building-stones, and artificial, or bricks. "VYe shall now study only the natural material, and return to the artificial wrhen treating of earthenware. Generally speaking, those stones are selected for building which are the cheapest, and possess at the same time sufficient resistance to the action of rain and frost. Very often the prefer- ence is given to those which are light, easily worked, and will well bind with the mortar. The preference given to any kind of stone depends essentially on the use for which it is intended ; thus moles, breakwaters, etc., which are constantly washed by the waters of the ocean, can be built only of very hard stone, capable of resisting the corroding action of salt-water. For the foundation of houses in damp lo- cations, a hard stone, not likely to nitrify, is required. Building-stones may be divided into three classes, according to their chemical nature: 1st. The stones formed by the alkaline and earthy silicates, such as granite, porphyry, certain trachytes, and basalts. As these stones are very difficult to cut, being extremely hard, they are used, in the form of hewn stone, only for special constructions, demanding great solidity and subject to continual wear, such as sea-dikes, footways, pavements, etc. Moreover, they do not bind well with mortar. Being capable of a fine polish, and often pre- senting beautiful shades of colour, they are used for pedestals, obelisks, columns, and other large architectural ornaments. Many volcanic rocks also furnish a material highly valued as building-stone, as possessing lightness combined with great solidity. Certain volcanic pumice-stones and scoriae yield a light material, very valuable in the construction of inside arches. 2d. The quartzose rocks found in various geological formations also yield a good material for building. The most important of these are the sandstones. Graywacke, the old red sandstone, and the variegated sandstone furnish excellent stone for cutting. In the tertiary formation in the environs of Paris, a quartzose rock, called millstone (meuliere,) which, being porous and light, is nevertheless very solid, is frequently used for the foundation of houses, because it arrests with great efficacy the dampness of the earth, and cannot nitrify. The quartzose pebbles found in layers in the various strata of cretaceous rocks are also sometimes used. 612 LIME AND MORTARS. 3d. The limestones furnish very valuable building material. White marble and certain coloured and veined transition lime- stones are used for ornamental purposes, such as mantel-pieces, hall-floors, etc., or for monumental and artistic purposes. The tertiary limestones and those of the jurassic formation fur- nish a material highly prized for cutting. They may be divided into compact and granular limestone; the first, being hard, re- sists wear, nitrifies with difficulty, and is susceptible of a high polish. The limestone of Chateau-Landon is of this class, and is extensively used in Paris for monuments, especially in those parts intended to be sculptured. The ordinary building-stone of Paris is a conchiferous lime- stone, called coarse limestone. The different strata of this rock yield stones varying in value, the inferior qualities and their strata are used for ashlar-work. The chalk formation also yields a moderately good building- stone : the chalk tufa of Touraine is used for building throughout a large portion of central France. The famous travertin in the environs of Rome, which has been used in the construction of the greater portion of the monuments of Italy, is a fresh-water calcareous tuff, belonging to the tertiary formation. The compact limestones may be used in building immediately after being quarried, which is not the case with the other lime- stones : as they are more or less porous, they must be exposed to the atmosphere for several months, or even years, in order to evaporate their quarry water. These stones are often very soft when taken from the quarry, and harden in the air : chalk tuff, which, when recently extracted, may easily be cut with a knife, does not become hard until after several years’ exposure. § 651. Building materials are divided into two classes, accord- ing to their form: regular material, such as hewn stone, bricks, etc., and irregular material, as rubble stones and large pebbles. Buildings may be constructed, with regular materials, without the interposition of any substance to unite’ their surfaces, pro- vided these surfaces be hewn so as to be in pretty close contact. Walls constructed in this manner are called dry walls. But, with irregular materials, a solid building can only be erected by inter- posing a substance called mortar, intended to fill the interstices, and bind the materials to each other. It is necessary that the mortar should acquire, after some time, sufficient hardness and adhesion to prevent its falling off, or being washed out by rain. Even with regular materials, a thin coat of mortar is interposed to close the interstices; but in this case the mortar is not required to fulfil the same conditions as when used with the irregular ma- terials : it need not require the same hardness, at least with regu- lar materials of large size. BUILDING MATERIALS. 613 We shall divide mortars into three classes: 1st. Common mortar, made with non-hydraulic lime. 2d. Hydraulic mortars. 3d. Mortars eminently hydraulic, or cements. COMMON MORTARS, MADE OF FAT LIME. § 652. A paste of lime and water, exposed to the air, will dry after some time ; the greater part of the water evaporating, leaves a cracked and friable mass of hydrated lime. But if a very thin layer of this paste be laid between well-dressed and porous stones, the greater part of the water soaks into the stones, and the thin layer of hydrated lime which remains becomes consistent and adheres strongly to the stones. The water should not soak up too rapidly, for in this case the lime sets too quickly, and never be- comes very hard: for this reason the stone is moistened before applying the diluted lime. As the adhesion of the hydrated lime with the stone is greater than that with its own particles, the layer must not be too thickly spread. A much more consistent material is obtained by mixing slaked lime with two or three times its wTeight of quartzose sand, or some ground stone, and tempering the whole with water. This mixture is applied, with a trowel, on the moistened stone: another stone is laid thereon, pressing it down as much as possible, so as to squeeze out the superfluous cement, and obtain only a very thin layer of mortar. Each grain of sand is thus enveloped in a small pellicle of lime, and adheres to it strongly. The addition of the sand presents another advan- tage, in preventing a too great contraction of the substance on drying, which would thereby split and become too friable. The solidification of this kind of mortar does not depend on the eva- poration of the water alone, but also on the combination of the lime with the carbonic acid of the air. The portions in immediate contact with the air are entirely converted into carbonate of lime; but the inner parts only reach the condition of a compound of carbonate and hydrate of lime, which acquires a great degree of hardness (§ 551). This change requires a long lapse of time, for, after many years, the lime in walls is still found almost entirely hydrated : these mortars, therefore, should not be used in the inte- rior of thick walls, where no opportunity of drying is afforded them. It will readily be understood that the quartzose sand mixed with lime has exerted no chemical action; for if the solidified mortar be dissolved in an acid, no gelatinous si'lex is separated, which would probably take place if the sand had partially combined with the lime, forming a silicate. The quality of the mortar depends greatly on its mode of pre- paration, the quality of the sand, the quantity of water used, and the more or less perfect mixing of the compound. Sharp sand is 614 LIME AND MORTARS. preferable to smooth. In all cases, the mortar should solidify slowly : it even has been remarked to become more consistent wffien laid in the fall than when used in summer, when the water evapo- rates too rapidly. Common mortar, made with fat lime, is used not only for regular materials, such as hewn stone or bricks, but also for rubble-stone. Care must be taken, however, to insert small pieces of stone be- tween the others, when the spaces are too large. In dry places, common mortar, made with fat lime, does not set until after some time; but it solidifies with difficulty in any case, and not at all in water. In this last case, the mortar is soon di- luted with wTater, and in a short time entirely w7ashed aw’ay. For buildings in damp locations or under water, peculiar mortars are necessary, which set, not in consequence of their desiccation, but by virtue of a special chemical action. These are known by the name of hydraulic mortars. HYDRAULIC MORTARS AND CEMENTS. § 653. We have seen (§ 553) that pure carbonate of lime, or that containing only a few hundredths of foreign substances, yields on calcination a lime, the properties of which are closely allied to those of pure lime. This lime, called fat lime, develops a great degree of heat with water, and swells considerably, its volume be- coming three or four times that of anhydrous lime. But when the limestone contains a greater proportion of foreign substances, its properties are remarkably changed, and it acquires new ones, which have been turned to good advantage in the art of building. If of the impurities occurring in carbonate of lime, the most promi- nent are the oxides of iron and of manganese, or quartzose sand, such a limestone yields, when burned, a lime which swells but little, and will form no adhesive paste with water: when tempered, it hardens in time, hut falls to pieces in the water. If the foreign substance mixed with the limestone be clay, or silex in a certain state of division, and if its proportion is as much as 10 or 15 per cent, of the limestone, the lime produced from it is a poor lime, but pos- sesses the remarkable quality of setting under water after a longer or shorter time, provided it has not been too strongly calcined. This kind of lime is called hydraulic lime. The setting of hy- draulic lime is owing to a chemical combination between the lime and the silex of the clay : the manner in which these two sub- stances communicate this property to the lime becomes manifest from the following experiments: If we preserve for some time, in a bottle well corked, lime-water and dried clay at a temperature of 300° or 400°, the clay will be found to abstract the lime from the water; and, if the contact be sufficiently prolonged, the water no longer turns tincture of litmus BUILDING MATERIALS. 615 blue. Gelatinous silex, when substituted for the clay, also takes up the lime, though less actively than the clay. Hydrated alumina likewise removes some of the lime, while magnesia, the oxides of iron, and manganese are nearly inert. This experiment shows that alumina, silex, and particularly clay, have an affinity for lime suf- ficient to separate it from the water, and form with it an insoluble compound, while magnesia and oxide of iron do not possess this property. Silex, in the state of quartzose sand, is equally inert. If a gelatinous silex, previously dried to the state of a mealy powder, is mixed with lime, the whole worked up with water, al- lowing the paste to stand for some time, a portion of the lime combines with the silex, as the water no longer dissolves the whole of the lime; and, if the substance be treated with an acid, a por- tion of the silex separates in the gelatinous state, which proves it to have been in combination with the lime. Lastly, by subjecting a very intimate mixture of carbonate of lime and clay to a moderate heat, a substance which hardens after some time with water is obtained, in which the greater portion of the lime is combined with silicate of alumina; for it is only par- tially soluble in water, and leaves a residue of gelatinous silex when dissolved in a weak acid. The clay is attacked by the weak acid by being burned in contact with carbonate of lime, while in its original state it is unaffected by them. These experiments show that the solidification of hydraulic lime under water is caused by the combination of the hydrate of lime with the silicates of alumina and lime: a new aggregation of the material is thus effected, and at the same time the lime is rendered insoluble in water. § 654. Intimate mixtures of limestone and clay are found in na- ture, argillaceous limestones, which yield hydraulic lime immedi- ately on burning. Experiment has shown that a limestone must contain at least 10 or 12 per cent, of clay in order to possess hy- draulic properties. The lime produced by such a stone, when tem- pered with water, hardens in moist places, or under water, in about twenty days; but the hydraulic properties are much greater when the limestone contains 20 to 25 per cent, of clay: then the tem- pered lime acts in two or three days. Lastly, if the limestone con- tains 25 to 30 per cent, of clay, it sets in a few hours: this last kind of lime has been called Roman cement, or lime cement. The nature of the clay exerts a great influence on the hydraulic qualities of lime : a fine division of the clay and a slight combi- nation of the silex with the alumina are indispensable conditions. The best clays are those which give off a portion of their silex to a solution of caustic potassa. The first Koman cements were made at London, from the pebble- stones found in the bed of the Thames: precisely similar pebbles have since been found on the seacoast, in the vicinity of Bou- 616 LIME AND MORTARS. logne. Subsequently, thick strata of limestone, belonging to the jurassic rocks, and yielding an excellent cement, have been dis- covered in Burgundy, in the vicinity of Pouilly and Yassy. In those countries where the jurassic rocks are abundant, chemical analysis has led to the discovery of similar limestones. We here subjoin the analysis of the principal limestones con- taining hydraulic lime, and cements, which are used in building. Moderately Hydraulic Limestones. Of Macon. Of St.Germain (Ain). Of Bigna. Carbonate of lime 89.2... 85.8 .. 83.0 “ of magnesia 3.O.... 0.4 .. 2.0 <( of iron —... 6.2 Clay or silex 7.8.... 7.6 .. 15.0 100.0 100.0 100.0 Highly Hydraulic Limestones. Carbonate of lime “ of magnesia.., u of iron Of Metz. . 77.3..., . 3.0.... . 3.0.... Of Senonches. 80.0.... 1.5.... Of Lezoux (Puy-de-Dome). 72.5 4.5 “ of manganese Clay or silex . 1.5.... . 15.2.... 18.5 23.0 100.0 100.0 100.0 The limestone of Senonches contains only very finely divided silex. Limestones containing Cement. Of Boulogne- Of Pouilly O f Argenteuil, sur-mer. OfLondon. (Cote-d’Or). near Paris. Carbonate of lime 63.6 ... ... 65.7.. .... 57.2... ... 63.0 d. Crystal. § 686. Crystal is a kind of glass used only for the fabrication of articles of luxury; it must therefore be very transparent, per- fectly homogeneous and colourless, and the greatest care must be exercised in the selection of the materials for its composition. Crystal is a double silicate of potassa and oxide of lead, the com- position varying greatly in the different factories: the proportion 638 GLASS of the oxygen of the silicic acid to that of the united bases ranges from 6 : 1 to 9 : 1. The ratio of the oxygen of the potassa to that of the oxide of lead ranges between still wider limits, viz. from 1:1 to 1: 2.5. By increasing the proportion of oxide of lead, greater density and higher refracting and dispersing powers are imparted to the crystal, which produce in cut-glass the beautiful effects of colour by transmitted light. But the proportion of the oxide of lead cannot be increased indefinitely, because the crystal, in that case, acquires a yellowish tinge. The finest and purest sand is chosen for the manufacture of crystal: the carbonate of potassa employed is refined; and the ordinary oxide of lead or litharge is not used, because it always contains some particles of metallic lead, which would be scattered through and injure the glass. Minium, an oxide of lead of a degree of oxidation superior to the protoxide, only is used : this oxide can- not contain metallic lead, and the oxygen it evolves when heated prevents the reduction of any lead by the carbonaceous dust or particles of other substances which may fall into the pot. The ordinary proportions for tumblers, decanters, &c., are 300 parts of pure sand, 200 “ minium, 100 “ purified carbonate of potassa. Crystal-glass furnaces are generally heated with wood ; m some, however, coal is burned, but in that case the shape of the pots must be changed. Coal produces a very fuliginous smoke, the deoxidizing action of which it would be very difficult to prevent, if the glass were melted in open pots; peculiarly shaped pots (fig. 431), called covered cru- cibles, or muffles, are therefore used: their vertical opening is placed in front of the working-hole of the furnace. Many articles are made of crystal by blowing, but it is also cast in great quantities in bronze or wooden moulds, which latter are kept moist, so as not to car- bonize too rapidly. § 687. The glass tubes used by chemists, and also thermometer- tubes are made by a particular process, which we shall briefly de- scribe. The workman gathers on the end of his pipe a certain quantity of glass prepared as usual; he then blows it into the shape of a pear (fig. 432), which he makes larger or smaller, thicker or thinner, according to the size and thickness of the tube required. Another workman has also gathered some melted glass on the end of a pipe, and applies it to the bottom of the bottle (fig. 433); the two workmen then recede rapidly from each other. The Fig. 431. Fig. 432. Fig. 433. 639 glass pear is then drawn out, as seen in figs. 434 and 435, and is converted into a tube terminating into two swollen extremities. GLASS, Fig. 434. Fig. 435. Tubes of 30 or 40 metres in length are thus made: they are laid on a wooden floor, and divided into lengths of 1 metre each. It will be seen that the external diameter of these tubes is not equal throughout its whole length, being generally smallest toward the centre; neither is the internal calibre more regular, and it is rare to find a tube possessing the same internal diameter throughout its whole length. MANUFACTURE OF GLASS FOR OPTICAL PURPOSES. Crown-glass and Flint-glass. § 688. Two kinds of glass are used for optical instruments: one, called crotvn-glass, is analogous in its composition to Bohemian glass, w'hile the other, called flint-glass, is a species of crystal. This glass must be as colourless as possible, and perfectly homo- geneous : great care is therefore required in the choice of the materials entering into its composition, and they must be refined expressly. Ordinary flint-glass is manufactured of 100 parts of white sand, 100 “ minium, 30 “ very pure carbonate of potassa. The density of this flint is about 3.5. A more refracting flint, but one slightly coloured yellow, is made of 225 parts of wdiite sand, 225 “ minium, 52 “ carbonate of potassa, 4 “ borax, 3 “ nitre, 1 “ peroxide of manganese, 1 “ arsenious acid, 89 “ cullet of the preceding flint. The melting-furnace (fig. 436) contains only one covered crucible or pot, into which the mixture is gradually introduced by small portions at a time, always waiting until the preceding charge has become perfectly fluid. Eight or ten hours are required for the wThole charge of a pot. A strong blast is then applied, and kept up for four hours, to render the mixture perfectly fluid. When this is effected, a hollow cylinder a5, made of fire-clay, previously heated to redness, and which does not sink in the melted glass, on 640 GLASS, account of its greater lightness, is introduced into the pot. Into the cavity of this cylinder a curved iron bar fe is passed, the end of which is heated to redness: by resting this bar on an iron gallows kl, the clay cylinder may be moved in any direction, so as to mix intimately the various parts of the liquid mass. The bubbles of air are thus driven out, and the whole rendered perfectly homogeneous: this operation must be frequently repeated, to make the glass as perfect as possible. The clay cylinder is then removed, and the furnace allowed to cool slowly for 8 days. The pot is then taken out, and is broken after cooling, to retract the glass, on which small polished facets are cut, here and there, so as to judge of its quality in various parts. This mass is then broken into pieces, and those that are perfect are selected, and heated in a muffle to soften them; they are then rolled into balls with pincers, and afterward carried to moulds which give them a lenticular shape. Lastly, they are allowed to cool slowly in an annealing-furnace. Crown-glass is made exactly in the same way, of Fig. 436. 120 parts of white sand, 35 “ carbonate of potassa, 20 “ carbonate of soda, 20 <£ chalk, 1 “ arsenious acid. By joining two lenses, properly cut, one of crown, and the other of flint-glass, achromatic lenses are obtained, which are remark- able for their property of giving the same convergence to all the coloured rays, so that a colourless object produces, in the focus of the compound lens, an image equally colourless, the edges of which are free from the coloured fringes always presented by images seen through simple lenses. This property, however, is very manifest only in those rays which do not depart very far from the axis of the lens. Strass. § 689. A peculiar kind of crystal is sometimes made, very dense and refracting, resembling the diamond, when it has been properly cut. By colouring this glass with various metallic oxides, coloured GLASS 641 glasses closely imitating the precious stones are obtained. This crystal, called strass, should be made of the purest materials, and requires great care in fining: generally, a certain quantity of borax is added. The manufacture of artificial jewels has in mo- dern times reached great excellence. Enamel. § 690. The name of enamel is given to a species of glass, ren- dered opake by an addition of certain metallic oxides. Peroxide of tin or stannic acid is generally used for this purpose : however, arsenious acid, phosphate of lime, or antimoniate of oxide of anti- mony may also be employed. Enamel is generally made of a very fusible crystal. An alloy of 15 parts of tin and 100 parts of lead are oxidized in a reverberatory furnace, by which a stan- nate of oxide of lead is formed, which is purified by levigation. 100 parts of this plumbeous stannate are then mixed with 100 parts of very pure sand and 80 parts of carbonate of potassa: an addition of small quantities of certain metallic oxides to this mix- ture gives coloured enamels. OF THE IMPERFECTIONS AND ALTERATIONS TO WHICH GLASS IS SUBJECT. § 691. We have seen that objects made of glass are kept for some time in a furnace at a dull red-heat, and then allowed to cool slowly: this process, called annealing, is a very essential operation, for glass cooled suddenly after blowing, is so brittle as to be useless. It frequently happens that common tumblers, which are imperfectly annealed, break suddenly on a slight change of temperature : such glass sometimes, also, is fractured when exposed to the current of air from an open door. This property is highly developed in the lachrymce Batavicce or Prince Rupert's drops. These are drops of glass suddenly cooled, and made by allowing drops of melted glass to fall into cold water: they thus become suddenly solid, in the form of tears, (fig. 437), terminating in a long tail; and as the outer surface solidifies while the interior is still at a high temperature, it retains nearly the shape it had in the liquid state. The internal particles are kept in an abnormal condition by those of the surface surrounding them: if this resist- ance of the surface particles be removed, at only one point, the whole mass bursts with noise, and falls into dust. This occurs, for example, if the tail of the drop be broken off. A similar effect is produced in a small glass apparatus, long known as the philosopher s phial, a kind of glass tube, thick, and of a pyriform shape : the master-blower frequently makes them on his pipe, when trying the metal in the pot. If any hard substance, Fig. 437. 642 GLASS, a small ball for example, be dropped into this phial, which has not been annealed, the shock is sufficient to reduce the phial to dust. The workmen apply this tendency of glass to break in a given direction when touched with a cold body, to detach the pipe from the objects blown, or to crack the glass in any direction required. § 692. When glass has been exposed for a long time to a high temperature, it loses, by volatilization, a considerable portion of its alkali, and becomes less and less fusible, at the same time ac- quiring the property of readily crystallizing by slow cooling. Thus masses of glass of a crystalline structure are often found in the worn-out pots which have been for a long time in the furnace and cooled slowly : at other times, the crystallization is developed only in some parts of it, the remainder being vitreous ; the vitreous portion always containing more alkali than that rendered opake by crystallization. This alteration of the glass takes place not only at its fusing point, but also at a lower temperature. If a glass bottle be left for several days in a furnace, at a degree of heat approaching that which effects the softening of the glass, it entirely loses its transparency and resembles a porcelain bottle. The glass thus altered, devitrijied, is much less fusible than when transparent. A peculiar art was attempted to be founded on this property, which consisted in making objects of blown glass, and then destroying their fusibility by devitrification. This devitrified glass was called Reaumur's porcelain; but the manufacture of it has been abandoned. § 693. Glass containing a large proportion of alkali changes by exposure to moist air, its surface becoming rugose and cracked. Frequently an excessively thin pellicle of altered glass forms on it, producing the same play of colours as a soap-bubble, or a drop of oil on a large surface of water; an alteration produced by the surface of the glass parting, after a long time, with a portion of its alkali: it is particularly remarkable in pieces of glass which have remained buried for years in a damp soil. These pieces are some- times found to have entirely lost their transparency, to be swollen, and cleavable into very thin lamellae: then they exhibit the same play of colours as mother-of-pearl. OF GLASS-WORKING IN THE LABORATORY. § 694. Various small objects are made of the glass tubes of com- merce ; for this purpose, an oil-lamp, generally made of tin (fig. 438), fed by a bellows, and called an enameller’s lamp, is used. The wick is of cotton, and does not project very high. The bellows is worked with the foot: the blast of air is conveyed by a pipe which Fig. 438. GLASS, 643 can be turned in various directions. By properly arranging the wick, and modifying the inclination of the pipe, and adapting a proper aperture to it, a flame of any size may be obtained at pleasure. When working with a plumbeous glass, or crystal, the flame must be made oxidizing by admitting a greater quantity of air; for, if the flame were reducing, oxide of lead would be brought to the surface of the glass in the state of metallic lead, and the glass would be blackened. It is important not to heat the glass too suddenly, lest it should break ; it is therefore first held for a few moments before the flame, and brought by degrees into the hottest part. § 695. In order to bend a glass tube, it is heated to the distance of 3 or 4 centimetres on each side of the point of flexion, turning it constantly, so that its whole periphery may be uniformly heated. As soon as the tube is sufficiently soft to yield to a slight force, it is bent; but it is important not to make the curve too short, be- cause the tube would be misshaped and brittle. The tube is there- fore not heated at the point where it was begun to be bent, but the flame is directed on the adjacent part, so as to make a small arc of a circle. Tubes can be bent in an alcohol-lamp even more readily than in an enameller’s lamp, for it is better not to have the glass too hot. § 696. In order to close a tube at one end, a longer tube is heated in the enameller’s lamp, at the point of closure, turning it con- stantly in the flame : as soon as it is perfectly soft both ends are gently drawn out, still turning it. The tube thus takes the shape of fig. 439. The point of the flame is then directed to the point a of the narrow part, and the two halves of the tube are sepa- rated, each of which will fur- nish a tube closed at one end; the ends are then rounded and made more uniform in thiekness. To do this, the end is again heated in the lamp, blowing into it occasionally, to round it. Lastly, a border is only required to complete it; which is made by simply heating the sharp edges until they are rounded by fusion. If the edges are to be widened, or a mouth made to pour liquids, it is done by applying an iron wire against the softened edges, by which means the aperture can be fashioned at will (fig. 440). When the end is to be closed, this end is heated in the lamp, and the heated end of another tube applied to it. The two tubes are soldered together, and the operation is then continued as just de- scribed. § 697. It is frequently necessary to solder a smaller tube cd (fig. 441) to the end of a larger one ab. The larger tube is then drawn out, in the lamp, till it is of the size of the smaller one, and Fig. 439. Fig 440. 644 GLASS the tube ab is next closed at the point b of the narrow part, by placing the part b in the point of the flame, and turning the tube between the fingers. Then, after having heated the closed end to soften it, a very thin sphere, which bursts by blowing through the opening a, is formed at the end b. By means of a file, the glass is separated so as to leave only a widened edge-border at the end b. The same is done to the end c of the small tube; the ends b and c are then exposed to the flame, opposite to each other, turning them constantly, after having previously closed the end a with a cork. When these ends are sufficiently softened, they are pressed firmly against each other, the joint is equally heated throughout, and from time to time the operator blows through the small tube, in order to prevent the solder from forming a ring. Lastly, it is drawn out slightly, so that no swelling may exist at the point of union. § 698. If a narrow tube cd (fig. 443) is to be soldered to the side of a larger tube ab, the point of the flame, after having rendered it as sharp as possible by a proper arrange- ment of the pipe and lamp-wick, is directed on the point e (fig. 442) of the tube ab. When it is sufficiently softened, the end of a glass point, also heated, is fastened and drawn quickly forward: thus a point ef is formed on the tube ab. This point is closed in the lamp ; then, having stopped the end K with wax, the point ef is again intro- duced into the flame, and when it is in fu- sion, a very thin sphere, which bursts, is formed by blowing through the open end l. A portion of the glass is filed off, the edges of the aperture are melted (fig. 443), and after having closed the end l with wax, the end e of the small tube, also heated, is brought in contact with the opening e. The joint is formed by gradually heating all its parts, and blowing from time to time through the opening d\ § 699. If a globe is to be blown at the end of a tube, the tube is closed in the lamp, and by continuing the action of the flame a mass of glass, large enough to make the globe required, is collected at this end. This mass of glass being very soft, the tube is gradually extended by blowing gently into it. It is then heated again uniformly, and afterward, by constantly turning the tube and blowing gently, a globe of any size may be produced at pleasure. When the globe is to be large, and still be at the end of a nar- row and thin tube, it is better to blow the globe separately on a larger tube, and then solder it to the narrow one. To do this, the Fig. 441. Fig. 442. Fig. 443. GLASS, 645 larger tube is drawn out between two points (fig. 444), by the process stated in § 696; one end a is closed in the lamp, and then the part A heated in the flame so as to soften it completely. Lastly, the operator blows through the end b, turning it constantly, until the globe has attained the size required: the globe is then soldered to the tube, as described (§ 697). But as the globe is still terminated by a point, the latter is placed in the flame, and, by blowing gently after having softened this part of the globe, it is distended so as to cause the small piece of glass to disappear. The bottles which are to contain the volatile liquids intended for analysis (§ 269) are blown in the same way. § 700. In order to fashion a funnel at the end of a tube, as, for example, on safety-tubes, a globe drawn out between two points (fig. 445) is soldered to the end of the tube, as in § 699, and then the point ab is detached (fig. 446). The part a, as wTell as the end of Fig. 444. Fig. 445. Fig. 446. Fig. 447. Fig. 448. the globe, is heated, and when they are very soft, a smart blow of wind through the tube is given: thus a second irregular and very thin globe (fig. 447), fastened to the first, is produced; this is broken and the glass de- tached by means of a file (fig. 448), so as to leave only an edge, which is melted in the lamp, and properly widened by an iron rod (fig. 449). Small bottles, intended to hold definite quantities of volatile liquids for analysis (§ 269), are made as in § 690, but of narrow and thin tubes. § 701. In order to break a glass tube at any given point, a mark is made on it with a gun-flint or a very sharp three-edged file; the tube is then pulled in the direction of its length, and it separates at the mark. If the tube be large, it must be slightly bent at the same time. In order to separate thicker and larger portions, as, for example, to shorten the neck of a retort or flask, a mark with a file is Fig. 449. 646 GLASS. made at the proper point, and followed with a point of red-hot iron ;* it then cracks in the direction of the mark.f COLOURED GLASS AND PAINTING ON GLASS. § 702. Glass dissolves the greater part of the metallic oxides, and while it preserves its transparency, is often tinged with the most beautiful hues: on this property the manufacture of coloured glass is founded. It suffices to mix intimately with the metal of which the glass is to be made, a given quantity of the metallic oxide, to produce coloured melted glass: with certain metallic oxides, how- ever, peculiar care is required. Protoxide of iron FeO imparts to glass a deep or'bottle-green colour, while the sesquioxide Fe203 produces a yellow tinge. Oxide of copper CuO and oxide of chrome Cr203 yield a beautiful green, but of different shades. Oxide of cobalt CoO gives a brilliant blue ; sesquioxide of manganese Mn203 a violet. A mixture of equal parts of oxide of cobalt and oxide of iron colours the glass black. Protoxide of copper Cu20 yields a very beautiful red colour, but so intense that the glass nearly loses its transparency if the oxide be in the proportion of a few hundredths. A fine purple is obtained by mixing a certain quantity of oxide of tin with finely powdered crystal, soaking the mass in a solution of chloride of gold, and melting it, when dried, in a crucible. When the metallic oxide to be used as a colouring agent can be deoxidized in the furnace, as, for example, the sesquioxide * A red-hot coal, held with a forceps, carried round the intended line of separa- tion, answers the same purpose: care must only be taken to blow away the ashes as soon as they form by contact with the cold glass, so as always to present a red- hot point to the surface of the glass.— W. L. F. f The process of dividing a tube by friction, described in Hare’s Chemistry, is so much superior to that adopted by our author, that the translator has not hesi- tated to substitute it for the French mode:— “ Some years ago, Mr. Isaiah Lukens showed me that a small phial or tube might be separated into two parts, if subjected to cold water, after having been heated by the fx-iction of a cord made to circulate about it, by two persons alternately pull- ing in opposite directions. I was subsequently enabled to employ this process for dividing large vessels, of four or five inches in diameter; and likewise to render it in every case more easy and certain, by means of a piece of plank forked like a boot-jack, and also having a kerf or slit cut by a saw, parallel to and nearly equidistant from the principal surface of the plank, and at right angles to the incision forming the fork. “ By means of the fork, the glass is held steady by the hand of the operator. By means of the kerf, the string, while circulating about the glass, is confined to the part where the separation is desired. As soon as the cord smokes, the glass is plunged into water, or if too large to be easily immersed, the water must be thrown upon it. This method is always preferable when the glass vessel is so open that, on being immersed, the water can reach the inner surface. As plunging is the most effectual method of employing the water, I usually, in the case of a tube, close the end which is to be sunk in the water, so as to restrict the refi'igeration to the outside.”—Hare's Compendiumi, ed. 4th, p. 60. GLASS, 647 of manganese can be, a small quantity of nitre is added to the mixture. A beautiful yellow glass is made by adding lampblack to a mix- ture which would produce common white glass. By varying the proportion of lampblack, several intermediate shades between a bright and a purple yellow can be produced. § 703. When it is wished to make glass of clear colours with metallic oxides which possess powerful colouring action, it is diffi- cult to obtain the shade desired by adding the proper quantity of the colouring oxide to the mixture in the pot. Glass is then set in layers, (verre plaque,) so that it is formed of white glass throughout the greater part of its thickness, and has one face only formed by a thin layer of coloured glass; and in order to vary at will the in- tensity of the colour, the layers are made of suitable thickness. This kind of glass is made as follows :—Two pots are placed in the oven, one filled with white, and the other with coloured glass. The workman first takes up with his pipe a certain quantity of white- glass ; then, when it begins to assume the proper degree of con- sistency, he dips it into the coloured glass, and thus fastens a layer of this on the white mass. He then blows the whole into cylinders, in order to make muffs for flatting (§ 688). The inside of the cylin- der is necessarily white, the layer of coloured glass being only external. Painting on glass is done with very fusible and finely powdered coloured glass. The composition of this glass varies with the nature of the colouring oxide; for the majority of them, a mixture of 2 parts of quartz, 2J of oxide of lead, and one of bismuth is used; but as certain colouring oxides are altered by the oxides of lead and bismuth, in this case a mixture of 2 parts of quartz, lj of melted borax, of nitre, and of carbonate of lime is used. The colouring oxide is added to these mixtures, and they are melted in a muffle-furnace ; the glass obtained is reduced to an impalpable powder, ground in turpentine, and the paint thus prepared is applied with a pencil. The painted glass is then heated in a muffle, at a temperature sufficient to melt the coloured glass, but not to affect the object on which the painting has been made. In order to form the groundwork of the picture, glass coloured in the paste is generally used, the outlines and shades being painted on one of the surfaces. The various pieces of glass are then dex- terously fitted together by means of small sheets of lead, each small pane harmonizing with the outline and shades of the figure de- signed. The painted surface of the glass is placed outside, so that the picture is seen through the coloured glass. The numbers and divisions marked on enamel dial-plates are applied in the same way. 648 GLASS ANALYSIS OF GLASS. § 704. We will suppose that the glass to be analyzed contains, or may contain, silex, potassa, soda, lime, manganese, alumina, oxide of iron, oxide of manganese, and oxide of lead. Five grammes of the glass, reduced to an impalpable powder, are intimately mixed with about three times its weight of pure carbon- ate of soda: the mixture having been weighed in a platinum cru- cible, the latter is covered with its lid, and heated in an alcohol- lamp having a double current of air, so as to completely melt the carbonate of soda. For this purpose, it is well to surround the crucible with a small sheet-iron chimney extending a few centi- metres beyond it: the chimney, at the same time increasing the draught, forces the flame completely to envelop the crucible. The carbonate of soda is kept melted for at least 20 minutes, and then allowed to cool. By using a thin crucible, the alkaline cup may be detached by the pressure of the fingers, and is received in a porcelain saucer, containing a certain quantity of water, and co- vered by an inverted funnel. Water, acidulated with nitric acid, being poured into the platinum crucible, and then into the saucer, the alkaline cup dissolves with effervescence, the funnel preventing any loss of substance, by the projection of the small liquid pel- licles surrounding the gaseous bubbles which burst on the surface of the fluid. Toward the close the liquid is acidulated with an excess of nitric acid, and evaporated to dryness at a moderate heat. Hot water, acidulated with nitric acid, is poured on the dried matter: it is allowed to digest for some time hot, and then diluted with water: all the metallic oxides then dissolving, leave the silex alone as an insoluble residue. It is collected on a filter, calcined after being well washed, and weighed. A current of sulphuretted hydrogen is passed through the liquid, which precipitates only the lead in the state of a sulphuret; and finally, the liquid is heated to ebullition, still keeping up the cur- rent of sulphuretted hydrogen, in order to facilitate the deposit of sulphur. The sulphuret of lead is collected on a filter, and, after having washed it, the filter is burned in a platinum crucible, and the substance sprinkled with nitric acid, mixed with a small quantity of sulphuric, in order to convert it into sulphate of lead : lastly, it is calcined to redness. The weight of the oxide of lead is deduced by calculation from the weight of the sulphate of lead obtained. Sulfhydrate of ammonia is then poured into the liquid to pre- cipitate the alumina and the sulphurets of iron and manganese; the wet precipitate is redissolved in chlorohydric acid, and the separation of the two oxides effected by the process described in § 659. The liquid, which then contains only lime, magnesia, and the alkaline salts, is boiled to drive off the excess of sulfhydrate of am- GLASS 649 monia, and chlorohydric acid added to decompose that which still remains. Lastly, it is supersaturated with ammonia, and the lime precipitated in the state of oxalate of lime by oxalate of ammonia; the presence of ammoniacal salts in the liquid (§ 589) keeping all the magnesia in solution. The solution is then concentrated by evaporation, an excess of carbonate of soda added, and it is evaporated to dryness, to decom- pose the ammoniacal salts, and drive off the ammonia as carbonate: it is then treated with water, which leaves the magnesia in the state of insoluble carbonate. § 705. In the analysis just described, the proportions of all the various components of the glass have been ascertained successively, with the exception of those of the alkalies, which must be found by a particular process. The glass is first dissolved in fluohydric acid. As this acid is difficult of preservation, it is better to prepare it freshly for each analysis, which is done in the following manner: Into a small platinum retort (fig. 450) made of two pieces, very finely Fig. 450. powdered fluor-spar is introduced and sulphuric acid added: on the other hand, 5 gm. of glass in impalpable powder are placed in a large platinum crucible, with a certain quantity of water, and co- vered with a sheet of platinum pierced with two openings. The neck of the platinum retort passes through one of those openings ; the other, much smaller, is traversed by a platinum wire, flattened into a spoon at its end, and used for stirring the material in the crucible. On gently heating the retort the fluohydric acid dis- solves in the water of the crucible, attacks the vitreous matter, and a large quantity of fluoride of silicium is disengaged. The material is stirred from time to time with the platinum spoon, and when the glass is entirely dissolved, the crucible is gently heated, to drive off the excess of acid and evaporate the water: sulphuric 650 GLASS, acid is then poured upon the residue, completely to expel the fluo- hydric acid and convert all the oxides into sulphates. When the greater part of the sulphuric acid has been driven off by heat, the substance is treated with water, which leaves the silex and sul- phate of lead as a residue. The liquid is filtered and an excess of carbonate of ammonia added, which precipitates the alumina, the lime, the oxide of iron, a part of the oxide of manganese, and the magnesia: an addition of a small quantity of sulfhydrate of ammonia completes the precipitation of the manganese. The liquid, when filtered, contains only the alkaline salts, a small quantity of magnesia, and salts of ammonia : it is evaporated to dryness, the residue calcined at a strong red-heat, and the alkaline bases are weighed in the state of sulphates. The magnesia is overlooked for the moment, until the termination of the analysis; the potassa is separated by perchloride of platinum (§ 527), and the soda is de- termined by calculation from the difference obtained. The magnesia must be sought in the solution remaining after the precipitation of the double chloride of potassium and platinum. The platinum is then precipitated by sulfhydrate of ammonia, and the liquid, filtered with an excess of carbonate of soda, is evapo- rated : the carbonate of magnesia is then separated by treatment with water. This base may also be precipitated by phosphate of ammonia (§ 592).* * A much better method of separating the magnesia from the alkalies is the fol- lowing, when the bases can easily be obtained as chlorides:—The liquid contain- ing magnesia and the alkalies is evaporated to dryness in a platinum crucible, after having condensed its volume by evaporation in a porcelain capsule, out of which the very concentrated solution is carefully washed, with as little water as possible, into the platinum vessel; a small quantity of pure red oxide of mercury is then added, and the crucible subjected to a strong white-lieat over a spirit- lamp, until all the mercury is volatilized. Care must be taken not to inhale the fumes. The magnesia then all remaining as insoluble caustic magnesia, is sepa- rated by filtration from the alkalies, which then may be determined by weighing them together, determining the potassa by precipitation with chloride of pla- tinum, and finding the weight of the soda by the ditference. Phosphate of soda, with the addition of some ammonia, effects the precipita- tion of magnesia much more perfectly than phosphate of ammonia.— W. L. F. 651 POTTERY. § 706. The term pottery is applied to all objects made of an argillaceous earth, to which a certain consistency is given by burning. The art of pottery is also called the ceramic* art, and the earthy pastes used in the manufacture are termed ceramic pastes. Clay is the base of all the ceramic pastes, and is plastic in the highest degree: when reduced to a proper state by water, it may he kneaded, fashioned, and moulded to any form, and when, by drying, it has become more consistent, may be worked on the lathe and cut with edged tools; and lastly, burning gives it a great degree of hardness. These various properties render clay highly adapted to the manufacture of hollow-ware. Burnt or merely dried clay adheres strongly to the tongue: this physical property is owing to the fact that the substance is traversed by innumerable capillary canals, which rapidly absorb the moisture of the tongue, so that it sticks closely to the clay. In consequence of this porosity, vessels of baked clay allow water to soak through them, and must, therefore, to be rendered imper- vious to fluids, be covered by a varnish, called glaze. The glaze of fine pottery, as porcelain, is always formed of a vitreous sub- stance, very analogous in composition to the material of the paste itself: it should not be very fusible, and still should melt at a temperature below that at which the vessel would lose its shape : the glaze incorporates itself so closely with the paste; that the line of separation cannot be seen, if a piece of burnt porcelain be broken. To produce this effect, however, a very high tempera- ture and a large quantity of fuel are required, so that such a glazing is applicable only for high-priced ware. The glaze of common earthenware is much more fusible. Fine pottery, such as porcelain, is made of very carefully se- lected materials, and should be colourless after burning, so that the glaze may retain its transparency. Common pottery, on the con- trary, is made of impure clays, which are frequently ochreous, and much less rare than pure and colourless clay. As this pottery, after burning, becomes red, the colour is hidden by making the glaze opake, or giving it a very deep colour: in this kind of ware, the varnish is not incorporated with the material, but forms a dis- tinct layer, which is readily seen by breaking a piece. Pure clay, diluted in water, forms a paste eminently plastic and * Derived from Kipz/uo;, “potter’s clay,” as if from ma>, “to burn,” and ipa, “ earth.”—Trans. 652 POTTERY. easily worked ; but it shrinks greatly on burning, and it is difficult to prevent the vessels made of it from losing their shape and cracking. This inconvenience is remedied by adding another ma- terial, called cement, to the clay, which is then said to be scoured. In common pottery, the cement is generally a more or less ferru- ginous quartzose sand: powdered brick, or any powdered baked earth, is sometimes used. The addition of this material diminishes greatly the plasticity of the paste, renders it more difficult to work, and at the same time more porous. The glazing is there- fore more necessary if the vessel is intended to hold water. If a substance which begins to fuse at the temperature of burn- ing pottery, be intimately mixed with clay, a substance which remains translucid after fusion is obtained: the vessel, however, has not lost its shape, because it has not softened much at that temperature, and only the material added has undergone fusion. A similar phenomenon ensues when melted wax is dropped upon paper, the latter remaining translucid after the solidification of the wax. The aggregation of the paste by burning renders it hard and compact, and it would be unnecessary to add glazing to make it water-tight; but it is generally glazed, to improve its ap- pearance and remove the roughness of its surface. The verifi- able material added is often feldspar; at other times, lime, which, by combining with a part of the clay, forms a double silicate of alumina and lime, more fusible than the simple silicate. Oxide of iron produces the same effect; but, as it discolours the paste, is only used for common pottery. The proportion of verifiable ma- terial which can be mixed with the clay is limited, because it greatly diminishes the plasticity of the paste, and makes it harder to work. § 707. We shall divide earthenware into two grand classes: the first will contain that of which the paste softens by burning, and thus becomes compact and impervious to liquids; to this class belong the various kinds of porcelain* and stone-ware. The second will comprise those kinds of which the paste remains porous after burning : this class includes earthenware, fayence,f delft-ware, etc. POTTERY THE PASTE OF WHICH BECOMES COMPACT BY BURNING. § 708. Let us first examine porcelain : being the most expensive and beautiful of all the various kinds of pottery, its manufacture requires the greatest care. Porcelain. §709. The clay used in the manufacture of porcelain, called * Porcelain, from porcelana, the Portuguese word for a cup. f Fayence, from Fayenza, in Italy, where this ware was first made. PORCELAIN. 653 kaolin * is a product of decomposition of the igneous rocks of primitive origin, and, as it always proceeds from the change of a feldspathic rock, is most generally yielded by granites very rich in feldspar, though sometimes also by the porphyries, rarely by the trachytes. In these rocks, the feldspar has been more or less altered: in some, the silicate of potassa has entirely disappeared, while in others a small quantity still remains : in the latter case, fragments of unaltered feldspar, increasing the fusibility of the material, are frequently found in the midst of the earthy mass. To separate these fragments, as well as the quartzose particles, the material is washed in a vat: as the kaolin is generally very fri- able, this operation is easy; were it otherwise, it would be neces- sary to grind it previously, either in a mill or by stamping. The material is mixed with the water, by means of paddles moved by machinery: the largest particles fall to the bottom of the vat. The liquid mud is poured into a second vat below the first, where it is allowed to rest for a few moments, that the quartzose or feld- spathic particles may settle : it is then transmitted into a third vat still lower, where the water is allowed to settle for a long time, and deposit all the clay it holds in suspension: lastly, the clear water is drawn off, and the argillaceous mud at the bottom of the vat dried. The kaolin of Saint-Yrieix, near Limoges, which is almost ex- clusively used in the porcelain manufactories of France, presents, on an average, the following composition, after the levigation just described: Silex 48.00 Alumina 37.00 Potassa 2.50 Water 12.50 100.00 The washed kaolin of Mori, near Halle in Saxony, which is used in the porcelain-factories of Berlin, and which is produced by the decomposition of a quartziferous porphyry, contains, after cal- cination— Silex 71.42 Alumina 26.07 Peroxide of iron 1.93 Lime 0.13 Potassa 0.45 100.00 It is easily seen, with a lens, that this latter kaolin is not homogeneous, and that it contains a large quantity of pure sili- * Kaolin, from kao and lin, two Chinese words signifying porcelain-clay.—Trans. 654 POTTERY. ceous particles. In order to convert it into porcelain-clay, the addition of a certain quantity of finely powdered feldspar is re- quired. The kaolin of Saint-Yrieix must, on the contrary, be mixed with quartzose sand, reduced to an impalpable powder, and a cer- tain quantity of carbonate of lime. At the porcelain factory of Sevres, near Paris, different proportions are used, according to the quality of the porcelain to be made : For domestic For ornamental purposes. purposes. Washed kaolin .... 64.0 62.0 Chalk from Bougival .... 6.0 5.0 Sand from Aumont .... 20.0 17.0 Fine or feldspathic sand. .... 10.0 Quartzose feldspar .... — 17.0 100.0 100.0 § 710. The feldspar and quartz which are to be mixed with the clay, must be first rendered more friable, by being heated to red- ness and thrown into cold water: they are then reduced to an impalpable powder in a mill with edge-stones, and afterward levi- gated in order to separate the grosser particles. The paste of kaolin and that of the quartz and feldspar are mixed wet, and as intimately as possible: it is then dried, in order to give it the degree of consistency fit for further working. This desiccation is effected, either by compressing the liquid pap in a press, in muslin bags, or by heating it in peculiar ovens, or by leaving it for a long time in plaster-boxes, the porosity of which assists the evaporation. The paste which has become more consistent should be worked for a long time, in order to effect a more uniform mixture of the ingredients. This operation is generally effected by tramping in round vats, that is to say, by letting a man stamp it with his naked feet: it is then pounded with wooden stampers, after being rolled into balls. This paste is sufficiently worked when no bubbles of air can be seen on breaking it. These various mechanical operations require great care and cleanliness on the part of the workman. lie must prevent the in- troduction of dust or any organic matter into the paste; for a single hair will effectually destroy a piece of porcelain, as the gas disengaged by the decomposition of the organic matter produces blisters or cracks. Porcelain may be made of the paste thus prepared; but it has been found to improve by being kept for several years in damp places. It then undergoes what is called rotting; it becomes black inside, and disengages an appreciable smell of sulphuretted hydrogen. The small quantity of organic matter in the paste is PORCELAIN. 655 destroyed by spontaneous combustion, in the damp air: it reacts at the same time on some traces of the sulphates, which are also found in it, and transforms them into sulphurets, which, in their turn, disengage sulphuretted hydrogen while changing into car- bonates at the expense of the surrounding carbonic acicl. § 711. Before working up the paste, it is again mixed with the hand, and squeezed into balls, which are forcibly thrown on the table on which this work is done. The air-bubbles which formed in the paste during the rotting are, in this way, driven out. It is formed into articles of various forms, by several processes, of which we shall distinguish: 1. Throwing on the potter’s lathe. 2. Press-work. 3. Moulding, properly so called, or casting. The potter’s lathe (fig. 451) consists of a vertical axis, inserted at its lower part into a disk of wood, which the workman moves Fig. 451. with his foot: on the upper end of this axis is a smaller disk, sup- porting the paste to be worked. The workman, seated on a bench, places a certain quantity of paste on the upper disk, causes it to revolve by means of his foot, and fashions it into the form in- tended : when the piece is large, he adds an additional quantity of paste, and so on, until the proper size is attained. He gene- rally uses a pattern and several measures to guide him in shaping the piece. This first operation is called hollowing out the stuff (ebauchage), 656 POTTERY. and rarely produces a shape sufficiently regular to be immediately burned. The process is completed by shaping (tournassage), an operation frequently performed on the same lathe. The article hollowed out is in this case allowed to dry spontaneously for some time, in order to acquire more consistency; it is then made to re- volve on the lathe, and worked with a cutting-tool, precisely like turning in wood. Its outlines thus become well defined, and it is reduced to the proper thickness. Fig. 451 represents a workman in the act of finishing a vase by throwing. The fragments of paste detached during the operation are called turnings; they are mixed with fresh paste, to which they impart peculiar qualities. § 712. These operations may often be abridged by combining the moulding with press-work: let us, for example, study the manufacture of a dinner-plate. The workman, having deposited a proper quantity of paste on the upper disk of the lathe, fashions it with his fingers into a cylindrical vase of no great height; he then brings down the upper edges of the vase, and shapes out roughly the form of a plate. He stops the lathe, and, by means of a brass wire (fig. 452), cuts off the base of the plate, and detaches it from the platform of the lathe: after allowing the rough plate to dry for a short time in the air, to become more consistent, he inverts it on a plaster mould (fig. 453), which exhibits in relief the shape of the inside Fig. 452 Fig. 453. of the plate. By compressing the paste forcibly against the mould, so as to effect an exact impression, and then giving the wheel a circular concentric motion, he brings it under a brass or steel knife c, the edge of which presents the semi-profile of the outer surface of the plate. He gradually depresses this knife, so that it cuts into the plate to the proper thickness, of which he judges by marks on the knife. In some factories, the workman simply prepares a plate of paste, of proper thickness, compresses it by a sponge, on the plaster mould (fig. 453), and completes it by means of the knife, as has just been described. § 713. In moulding, the ceramic paste is applied to the mould, the shape of which it is to take: these moulds, which are generally PORCELAIN. 657 made of plaster, and always of some porous substance, are fa- shioned on a plaster, earthen, or even a metal pattern, when many are required. The mould is often composed of several pieces, which can be separated in order to remove the article made : they are held, until that time, in a kind of plaster box, moulded itself on the outside of the mould, and called a coat. As the ceramic paste must contract somewhat in consequence of the absorption of its water by the porous walls of the mould, the article moulded is easily extracted, provided the sections of the mould are so com- bined as to present no obstacles themselves. The projections at the lines of junction of the various parts of the mould are removed by a sharp instrument: these lines must be judiciously disposed, so as not to be too apparent, as they sometimes show on the pieces after burning. Moulds intended for the making of round objects, as handles and columns, are made of two equal parts which fit each other exactly. Half of the object is moulded in each of these parts, and, while the paste is yet soft enough to adhere, the two halves are united. The workman waits for a few moments, until the paste is partly dried by the absorption of the water through the porous sides of the mould, and then separates the two parts of the latter. § 714. In order to unite the various component parts of an object, the workman generally does not wait until they are thoroughly dried, but marks on the principal pieces the points of junction of the pieces to be added, and engraves thereon cross- cuts, to render them rough: he then applies with a pencil a thick pap, formed of the ceramic paste suspended in water, and called slip (barbotine); and then quickly applies the pieces. It requires a skilful workman to do this. In fact, ceramic objects, turned in the lathe, experience a contractive influence by the circular motion by which they were made, and even by the direction in which the pressure was applied. The piece, in burning, contracts concen- trically on itself; and if the handle of a vase has been accurately applied in the vertical position, it leans to one side on the burned piece: therefore, in order to obtain a vertical position after burn- ing, the handle must be slightly inclined, so as to counteract the effect of this twisting motion. The proper inclination depends on the length of the handle, and, to a certain degree, on the shape of the vase. The workman must foresee all these effects. § 715. A certain number of pieces of a peculiar shape is made by casting. If a liquid pap of ceramic paste thinned with water is poured into a mould of porous plaster, the mould absorbs a great portion of the water of the pap, and part of the paste ad- heres to the internal surface of the mould. In four or five minutes, the fluid pap is allowed to run off: the layer of paste adhering to the mould, to the thickness of 2 or 3 millimetres, becomes more consistent in consequence of the absorption of the water by the 658 POTTERY. sides of the mould. In a few moments, this layer is sufficiently dried to act as an absorbent on an additional quantity of slip. If, therefore, the mould be filled anew, a second coat of paste is formed, which adheres closely to the first, and this process is con- tinued until the sides of the object are sufficiently thick. In this way the porcelain tubes and retorts are made which are used in chemical laboratories, and also many hollow pieces, such as the spouts of tea-pots. As an exemplification, we shall select a porce- lain tube. The mould is formed of two equal parts (fig. 454), each presenting a semi-cylindri- cal canal, terminating into two small canals a, b. The two parts of the mould are joined by screw-collars l, l (fig. 455), and a cylindrical canal is thus formed, terminating by apertures. A coating of very clear slip is painted over each part of the mould, with a badger’s-hair pencil, and the two halves are fitted together. The slip intended for casting is contained in a bucket furnished with a stopcock, above another bucket having a cross-piece, in the middle of which is a conical leather bung. The lower end of the mould is then rested on the bung, which closes it exactly, the upper opening being, of course, just beneath the stopcock. As the latter is opened, and the canal filled with slip, the level soon sinks in the mould in consequence of the absorption of the water, and is restored by an additional quantity of slip; the mould is then removed from the bung, and the non-adherent slip falls off. As the adherent layer is not sufficiently thick, it is set aside for a few moments, long enough to fill three or four other moulds; the first is then filled anew, after having inverted it. If the tube is not yet thick enough, a third easting must be performed, always in- verting the mould. In 3 or 4 hours, the mould may be separated: the beard and blisters on the tube are then removed with a sharp instrument. § 716. The porcelain articles made by these various processes are first baked, so as to dry them completely and impart to them a certain degree of consistency; but the material is still very porous. They are then glazed, and finally burned. We spoke, in §706, of the glaze applied to porcelain, and the principal conditions it must fulfil. We saw that the material of the glaze must have a certain affinity for the ceramic paste, in order to cover the pieces perfectly and leave no part exposed; this affinity, however, must not be too great, or the glaze would penetrate into the paste, and not leave enough on the surface. The glaze must be more fusible than the ceramic paste; but the difference of fusi- bility again must not be too marked, for if the glaze should melt before the paste was burned, it would flow toward the bottom of the Fig. 454 Fig. 455. PORCELAIN. 659 pieces, or enter the substance of the paste. A last condition, and one of the most difficult to fulfil for pottery in general, is to give the glaze nearly the same dilatability by heat as the paste, as other- wise it would crack, and start in every direction. The glaze of Sevres porcelain is made of a feldspathic rock, mixed with a certain quantity of quartz. No other substance is added to it, but the rock is selected with regard to the quantity of quartz it contains, and the degree of fusibility of the glaze required. The glazing is generally done by immersion. The feldspathic rock is ground in water in mills, and then purified by levigation: the material, very finely divided, is suspended in water, to which a small quantity of vinegar is added, because this acid effectually prevents the precipitation of the powdered matter. This clear pap, called slip, is placed in large buckets, into which the workman dips quickly and dexterously the pieee to be glazed : the piece, from its porosity, absorbs the wyater, and the verifiable matter sus- pended in the water is deposited on its surface. By this rapid and simple process, the thickness of the glazing becomes uniform throughout, if one part of the piece has not been allowed to remain longer in the liquid than another. With a knife and a piece of felt, the glaze is removed from those parts which do not require its ap- plication. As the part by which the workman holds the piece is necessarily not glazed, it is afterward painted over with slip. In order that biscuit porcelain should be properly glazed, its surface must be perfectly clear, and especially free from all greasy substances; hence the workman should avoid touching them with his hands. Advantage is sometimes taken of this property to pre- vent certain parts of the piece from taking the glazing ; they are covered with a mixture of wax and tallow. Lastly, when it is de- sired that a piece, or a portion of it, be less highly glazed than another, it is more or less soaked with water with a pencil, before glazing; the absorbent action of the paste is thus diminished, and a thinner coat of glaze deposited. Glazing by immersion can only be done on porous pieces, such as biscuit porcelain; but if it is required to glaze pieces which, having been highly burnt, are no longer sufficiently porous, it is done either with a brush or by sprinkling. § 717. Porcelain-kilns are generally composed of 2 or 3 stories. In the upper story, where the temperature is lowest, the biscuit is burned, and in the lower, or two lower stories, if there be three, the last burning of the porcelain is effected. Figs. 456 and 457 represent a three-storied kiln, in the manu- factory at Sevres. Fig. 456 gives an external view, and fig. 457 represents a vertical section through the axis of the kiln. In the twTo stories L and L' the porcelain is burned, and in L" the biscuit is baked. Each of the compartments L, L' is heated by four outer 660 POTTERY. Fig. 456. Fig. 457. furnaces immediately adjoining the kiln, and called alandiers. The flame in these furnaces is inverted: they are composed of a rec- tangular vat/, terminating below in a grate. The face they have in common with the kiln has several rectangular holes g, through which the flame enters the kiln ; the ash-holes, as well as the open- ings o, may be closed internally. When the porcelain is deposited in the kiln, in the manner to be described, a few live coals are placed upon the grate, and above that, wood split into short pieces: the door of the ash-hole is then closed. The draught of air is through the kiln itself, which acts as a chimney; the fresh air enters through the upper hole of the alandier, which is open, and the inverted flame passes into the oven through the openings g. The flame and current of hot air pass from the lower to the upper story, through the holes c made in the roof, and escape through the upper aperture t, which can be regulated at will by a register. Birch and aspen wood are used in the alandiers; pit-coal has not PORCELAIN. 661 yet been successfully employed, at least for fine porcelain, as it makes too fierce a fire in front of the working-holes g, and it is very difficult to render, with this fuel, the temperature of each compartment nearly uniform. Pit-coal, also, burns with a smoky flame, which frequently discolours the porcelain and diminishes its value. The kiln, made of refractory bricks, is firmly held together by an iron framework, which will be easily understood by an inspec- tion of fig. 456. In each compartment there is a large door P for charging the kiln, which is closed by brick-work during the burn- ing. In this temporary mason-work several small holes m (fig. 456) are made, through which small fragments of glazed porcelain, called time-pieces or watches (montres), are introduced : these are intended to be withdrawn from time to time, in order to judge of the progress of the baking. § 718. The porcelain articles cannot be placed unprotected in the kiln, for they would be exposed immediately to the current of hot air, carrying with it a considerable quantity of ashes, which would stick to the melted glaze. The various pieces must also not touch each other at any point, as otherwise they would adhere ; each piece must, therefore, be placed in a vessel called a seggar (cazette, or gazette). The seggars are made of refractory clay; they should be less fusible than the porcelain. Their paste should be coarse, that they may resist the immediate and unequal action of the fire without breaking, and can be used several times. They are composed of very pure plastic clay, carefully levigated, and freed from all par- ticles of quartz, limestone, or pyrites: to this clay a certain quantity of fragments of broken seggars, reduced to an impalpable powder, is added as a cement. At Sevres the proportions are generally 40 parts of washed plastic clay and 60 of cement. The seggars and supports are made in the same way as the pieces, but more roughly. The paste is stamped, in order to incorporate the cement with the various clays of which it is composed; it is then fashioned on the potter’s lathe, and turned, but only hollowed out. Seggars are generally made of two pieces: an external cover- ing, usually cylindrical, and a flat bottom, on which the porcelain rests; but their form varies with the use to which they are to be applied. Fig. 458 represents a series of plates; each seggar will be seen to be composed of two parts, a cylindrical covering t and a kind of ves- sel i, having nearly the shape of the plate, and on the bottom of which the bottom of the plate rests. The seggars are arranged Fig. 458. 662 POTTERY. above each other, so as to form a perfectly vertical pile, called a bung. The charging of the furnace requires particular care on the part of the workman. He should endeavour to fill the kiln as com- pletely as possible, without closing the working-holes, and still to preserve between the pieces the spaces necessary for a proper dis- tribution of the flame through the furnace: he places near the working-holes g, those pieces which, from their size or peculiar nature, require the highest temperature. The piles of seggars are fastened to each other by small caps of burnt clay. § 719. Fig. 459 represents very accurately the arrangement of the pieces in a kiln: some of the seggars are supposed to be divided, in order to show the porcelain inside. Fig. 459. Although the glazing may be removed with great care and very perfectly from those parts which come in contact with the supports, the porcelain paste might adhere at certain points, if, between the support and the part denuded of glazing, sandy argillaceous coat- ing wTere not interposed, of such composition as to prevent all ad- hesion : this is called the terrage of the supports. The spaces between the seggars are closed with an argillaceous luting, composed of 30 parts of plastic clay, and 70 of quartzose sand. The biscuit itself is placed in the seggars; but their arrange- ment is more simple, because there is no fear of their adhering to the seggars, and a great number of pieces may be placed in the kiln. The kiln being charged, the doors are walled up and the firing commenced. First the alandiers of the upper chamber are closed, and only those of the lower story heated. When the porcelain in PORCELAIN. 663 the first compartment is supposed to be burned, the alandiers of the second story are opened, and a small fire kept up for about an hour, without completely extinguishing the fire below. All the apertures of the lower alandiers are then closed hermetically, and those of the second story are partially, and afterwards entirely, stopped. The fire is carefully managed, until the pieces in the second story are perfectly baked; the furnace is then allowed to cool, and after having removed the brick walls which closed the doors P, the porcelain pieces are removed from the kiln. The baked porcelain is very carefully sorted : while the faultless pieces are considered as of first quality, the others are divided into several classes, according to the nature of their defects. § 720. The porcelain, the manufacture of which we have just de- scribed, is called hard, or Chinese porcelain. Other qualities are made, called tender, or French china, which require a lower tem- perature, and may be sold at a cheaper rate : the paste of this por- celain should be more fusible than that of the hard porcelain, a character which is easily given to it by introducing larger propor- tions of alkaline material, either in the state of feldspar, or even in that of alkaline carbonates and nitrates. The glaze of this por- celain should also be much more fusible than that of hard porce- lain : with this intention, a certain quantity of oxide of lead is introduced. Sometimes, in the manufacture of the paste of tender porcelain, the clay which is the essential base of the hard porcelains is not used. Thus, the paste of tender porcelain formerly made at Sevres, and now called Old Sevres china, was made by first frit- ting in an oven, Sand from Fontainebleau 60.0 Fused nitrate of potassa 22.0 Sea-salt 7.2 Alum 3.6 Alicant soda 3.6 Gypsum from Montmartre.... 3.6 100.0 Seventy-five parts of this frit were mixed with 17 of white chalk and 8 of marly limestone from Argenteuil, and black soap or gum was added to the paste, to make it more binding. The glaze was made of Calcined sand from Fontainebleau 27 Calcined silex 11 Carbonate of potassa 15 “ soda 9 Litharge 38 664 POTTERY. Stoneware. §721. This is a kind of porcelain, which differs from the finer sort merely in the paste being always more or less discoloured, and less carefully worked. The base of this ware is clay, which in gen- eral contains a large proportion of oxide of iron, and is thus more fusible than kaolin: sometimes the fusibility is still further increased by the addition of a certain quantity of lime, or of alka- line salts. Baked clays or quartzose sand are added as cement. The clay is scarcely ever washed, except for fine pottery; the larger pieces of quartz or limestone are merely separated by hand. The paste is fashioned on the potter’s lathe ; and the pieces, dried in the air, are baked in furnaces, at a temperature nearly equal to that used for the turning of porcelain. This kind of pottery is seldom glazed, but an ingenious substitute is employed: when the pottery has attained a very high temperature, a few handfuls of damp sea-salt are thrown into the furnace. This salt volatilizes, the vapour is decomposed by the presence of the water and the contact of the argillaceous walls, chlorohydric acid is disengaged, and the sides of the pieces are covered with a silicate of soda, which, by combining with the silicate of alumina, produces a very fusible double silicate, forming a varnish over the surface of the pieces. The paste of stoneware is of itself impervious after burning; therefore the only use of glazing is to give it a better finish. Some of this ware, however, merely dried in the air, is glazed, either with a brush or by sprinkling, with the scoriae of blast-furnaces or very fusible volcanic lava. POTTERY THE PASTE OF WHICH REMAINS POROUS AFTER BURNING. § 722. This division comprises several kinds, such as the various fayences and earthenwares used in cooking. Earthenware, or Fayence. § 723. The clays belonging generally to the secondary forma- tions are used for the manufacture of fine earthenware. When these clays do not contain any colouring metallic oxides, as the oxides of iron and manganese, t'he paste remains white after burn- ing ; but as they often contain large proportions of these oxides, the paste generally becomes red or brown by burning. Earthen- ware is always glazed, and subjected successively to two fires : they are first burnt at a high temperature, lower however than that for burning hard porcelain; and, after this first burning, they are covered, by immersion, with an easily fusible glaze, and exposed to a second fire, generally much less intense than the one first used. EARTHENWARE. 665 As the paste of earthenware should not soften in the first fire, it must be very slightly fusible: it is made of a plastic clay which has been carefully levigated for fine earthenware, and to which a greater or less proportion of quartz, reduced to an impalpable powder, has been added. The proportion of quartz is often greater than that of the clay, for in certain kinds of earthenware, a mixture of 70 parts of quartz and 30 of clay is used. The paste of earthenware is more easily worked than that of porcelain, being more plastic; in other respects it is moulded nearly in the same way. Earthenware is burned in kilns similar to those used for porcelain, but the charging is more simple : a much greater number of pieces can be introduced, because the paste does not soften, and there is no danger of their becoming misshaped. Thus, in the first firing of plaster, they may be placed on each other, and the whole pile surrounded by a cylindrical seggar. More care is requisite for the second firing of glazed earthenwares, because the pieces would adhere to each other. They must be supported by three points; and for this purpose the seggars have three holes, disposed in the same horizontal circle, through which small pieces of baked earth are introduced, on which the edges of the plates rest. Fine earthenware is glazed with a glass of alkalies and oxide of lead, the proportion of the latter being increased when a very fusible glaze is required, combined with economy of fuel; but the glaze is, in that case, very tender, and can be scraped off with a knife, so that the plates soon become scratched. The highly plum- beous glasses are, moreover, easily attacked by chemical agents, especially at the parts injured by the knife; they also soon blacken when in contact with substances which disengage sulphuretted hydrogen. It is sufficient to cook stale eggs or fish in these vessels to give them a brown tinge. Therefore, in fine earthen- ware, the proportion of the oxide of lead should be diminished as much as possible: but the glaze being the less fusible, the price of the ware is greatly increased, and approximates that of common porcelain, wThich is always preferable. The glaze of common French earthenware is made by melting together in a crucible, Quartzose sand 100 Carbonate of potassa or soda 80 Red-lead 120 to 150 One or two parts of smalt, or glaze coloured blue by oxide of cobalt, are commonly added to the mixture, in order to give a slight blueish tinge to the glaze, more agreeable to the eye than dead white. For a very fine article, such as the English earthenware, a very small proportion of oxide of lead is used. 666 POTTERY. The glaze of coloured earthenware should be opake, in order to conceal the disagreeable colour of the paste: the opacity is pro- duced by the addition of a certain quantity of oxide of tin. This glaze, a true enamel (§ 690), is then frequently coloured with metallic oxides. Earthenware is much less patient of- fire than porcelain; the glaze cracks and blisters very readily, as, for instance, when the vessels are washed in hot water. Common Earthenware. §724. Very common pottery, such as flower-pots, is made of impure clays, often very ochreous, and mixed with more or less sand. These clays are more generally used as they are taken from the earth, after merely separating the pebbles and lumps which are not easily crushed. They are rarely levigated, as this operation would render them too costly, but are stamped, and frequently allowed to moulder for years in pits, to increase their plasticity. The pieces are fashioned on the potter’s lathe, dried in the air, and then burned at a low temperature, without being glazed. § 725. The common earthenware used in cooking is made of clay, to which a certain quantity of lime in the state of marl, and also quartzose sand, is added. This ware is glazed generally with a plumbeous glass, coloured by some metallic oxide. The glaze is composed of 6 or 7 parts of litharge and 4 or 5 of clay. §726. In warm countries, chiefly in Spain, a very porous kind of pottery is made with clay mixed w7ith large quantities of sand or of baked clay. In this way the bottles called alcarazas are made, which, wThen filled with water, are easily permeated by this fluid, and present an outer surface constantly moist and exposed to the drying agency of the air. When these vessels are suspended in a current of air, the water evaporates so rapidly from their sur- face that the temperature of the water falls several degrees below that of the surrounding air: alcarazas are therefore used in hot countries to cool water; but they are not of much use in temperate climates, in which, during the summer, the temperature of the cellars is lower than that produced by the spontaneous evaporation of water. Common Building-bricks.— Tiles. § 727. Common bricks are made of clay burned at very various temperatures. In hot countries, they are merely dried in the sun ; but, in that case, they are always very friable, and can only be used in buildings where great solidity is not required. Most frequently, they are burned in a kiln, and sometimes the tempera- ture is high enough to effect a sort of superficial fusion of the ma- EARTHENWARE. 667 terial. Burned bricks are generally of a reddish colour, owing to the oxide of iron which frequently abounds in the clay. Nearly all the sedimentary and alluvial earths, containing argillaceous strata, are adapted to the manufacture of common bricks. When the clay is too fat, sand is added: the clay, when dug out of the earth, is left for some time in pits, and then stamped. The bricks are either made by hand, or in rectangular wooden frames, or sometimes by machinery : they are dried in the air, and then burned in kilns, with some cheap fuel. Roofing and paving tiles are made in the same way. Fire-bricks for the construction of furnaces ; and Crucibles. §728. Refractory clays, which consequently should contain no large quantity of oxide of iron, nor -of carbonate of lime, are used in the manufacture of these bricks. These clays are not common, and fire-bricks can therefore be made only in certain favoured localities, the best of which in France is Burgundy. The clay is coloured with quartzose sand. The greatest care is required in the selection of the clay for melting-pots ; for example, those used in glass-houses. These clays are nicely levigated, and scoured by the addition of a considerable proportion of cement, generally yielded by the broken pots, care- fully sorted to exclude all pieces of glass. The paste frequently contains more than and sometimes -§, of this cement. In the same way excellent crucibles are made, which resist a forge-fire as well as the small earthen furnaces used in chemical laboratories. Hessian crucibles, for a long time the most esteemed for the melting of metals, are made of a refractory clay, mixed with a coarse quartzose sand occurring in the tertiary formation near Almerode. These crucibles readily resist sudden changes of tem- perature without cracking. Crucibles highly valued for the melting of steel are also made of 1 part of fire-clay, and 2 of graphite or plumbago. This paste can be easily worked, and the crucible is susceptible of a high polish. During the burning of the crucibles, it undergoes no change internally, as only the surface of the graphite burns. Plumbago is found only in a few localities: within the last few years, it has been replaced by very finely powdered coke, and crucibles of good quality have been thus made. ORNAMENTS AND PAINTING. § 729. Fine pottery is often ornamented by the application of colours or metallic coatings to their surface: sometimes they are painted, thus adding greatly to their value. 668 POTTERY. The colours applied to porcelain are made of colouring metallic oxides, mixed with more or less fusible vitreous substances. The mixture, when melted, is reduced to an impalpable powder, and then ground with essence of turpentine or lavender: these pastes are applied with a brush, and the pottery is then subjected to a temperature high enough to vitrify the colours. These colours must fulfil several conditions; we shall enumerate some of the most important: 1. They must melt at a temperature which is not sufficiently elevated to effect a chemical decomposition, which would change the colour: in other words, the fusibility of the vitreous flux must be in proportion to the fixedness of the colouring matter, and the temperature to which the painted piece is subjected must be such as not to injure the most fugitive colour present. 2. They must adhere with sufficient firmness to the pottery after burning, to resist friction. 8. They must retain a vitreous aspect after burning. 4. They must be unchangeable by water, atmospheric air, and even by the liquids intended to be contained in the vessel. 5. They must bear a proper ratio of dilatability with the paste of the pottery, and especially with its glaze. § 730. The colouring materials may be divided into four classes: The first, comprising the most important and numerous colours, includes the metallic oxides. The second is composed of those mineral substances which re- tain an earthy and opake aspect after burning, and which obtain their lustre only from the general glazing which covers them. They are called engobes. The third class contains the metals; chiefly gold, silver, and platinum. They are applied in the metallic state, mixed merely with a small quantity of some fusible material, to cause them to adhere to the surface of the pottery. They are then polished by burnishing. The fourth class comprehends the metallic lustres. These are very finely divided metals, applied in excessively thin layers, and often produce a fine play of colours. Two kinds of verifiable colours are distinguished, according to the temperature at which the pottery is burned : refractory colours, which do not change even at the high temperature at which glazed porcelain is burnt; and muffle colours, which do not bear this tem- perature without alteration. The latter are vitrified at much lower temperatures in peculiar furnaces, called muffle-furnaces. Refractory colours may be applied under the glaze, or may be mixed with it, and then burned immediately in the high fire of the porcelain-kiln. Muffle colours, on the contrary, are only applied to glazed porcelain. ORNAMENTS AND PAINTING. 669 Refractory colours are not very numerous: they consist of the cobalt blue yielded by the oxide of cobalt CoO; chrome-green, produced by the oxide of chrome Cra03; the browns, made from the sesquioxides of iron and manganese ; the yellow, obtained from oxide of titanium; and the blacks, furnished by protoxide of uranium. Muffle colours are more numerous, and the palette of the porcelain- painter is nearly as richly provided as that of the portrait-painter. These colours are made by mixing in a crucible the metallic oxides with colourless glasses, called fluxes, the fusibility of -which is regulated by the temperature to which the paintings may be ex- posed without detriment to the most fugitive colour. The com- ponents of these fluxes are quartz, feldspar, borax or boracic acid, nitre, the carbonates of potassa and soda, red-lead and litharge, and oxide of bismuth. At Sevres, seven kinds of fluxes, which suffice for all colours, are used, the majority of which are composed of quartz, oxide of lead, and boracic acid ; to some, a small quantity of carbonate of soda is added. The flux for me- tals is composed of oxide of bismuth mixed with one-tenth of its weight of melted borax. We shall not here enter into the composition and mode of pre- paration of the various colours used in porcelain-painting, but merely indicate the chemical nature of the principal colouring sub- stances. The blues are always produced by oxide of cobalt, their shades being varied by an addition of oxide of zinc, or by that of small quantities of colouring metallic oxides. The greens are furnished by the oxide of chrome Cr303 and by protoxide of copper CuO, the shades being varied by adding other colouring oxides: they are rendered yellow by the oxides of anti- mony and lead; brown, by the sesquioxides of iron and manganese; blue, by the oxide of cobalt, etc. etc. The yellows are given by oxide of uranium U203, chromate of lead PbO,Cr03, sesquioxide of iron and antimoniate of potassa: they are mixed with the oxides of lead, zinc, and tin. The reds are produced by protoxide of copper Cu30, and by sesquioxide of iron. The violets and rose-colours are obtained from the purple of Cassius, which is an intimate mixture of metallic gold and per- oxide of tin in various proportions. The blacks are furnished by protoxide of uranium, or by the metallic oxides of cobalt and manganese. The whites are produced by ordinary enamel (§ 690). The gold is prepared by precipitating a solution of perchloride of gold by protosulphate of iron : the pulverulent gold is mixed with one-twelfth of its weight of oxide of bismuth, to which a little borax has been added; the whole is diluted with some essential 670 oil, and painted on the glazed porcelain. After burning, the gold assumes a metallic lustre, hut it is dead: it is polished by rubbing it, first with an agate, and then with a blood-stone burnisher. Gold lustre is obtained by precipitating a solution of gold in aqua regia by ammonia. The precipitate, called fulminating gold, is mixed when moist with essence of turpentine ; it is spread without any flux over the surface of the porcelain. The piece is exposed to the fire, and the lustre of the gold brought out by friction Avith a piece of linen or muslin. A lustre, remarkable for its beautiful play of colours, and called cantharides lustre, is obtained from chloride of silver, which is par- tially decomposed by combustible vapours. A mixture of plumbeous glass, a small quantity of oxide of bismuth, and chloride of silver is applied with a pencil, and the piece is heated in a muffle; and Avhen it is red-hot, a smoky vapour is introduced into the muffle, by which the chloride of silver is partially decomposed. The burning of painted porcelain is an extremely delicate opera- tion : it is done in muffle-furnaces (fig. 460), the fire of Avhich is regulated Avith the utmost care. The AA’orkman is governed in the manage- ment of the fire, by the examination of small watches of porcelain, introduced into the muffle with the porcelain, and withdrawn from time to time: on these Avatches some of the most fugitive co- lours on the porcelain are painted, the rose-colour, for example ; and also the substances which Avould not adhere unless exposed to a sufficiently high temperature, the gold coating, for in- stance, are applied on them. The work- man must therefore so manage the fire, as to cause the gold to adhere firmly, without changing the shades of the most delicate colours. Painted porcelain is ahvays exposed to two firings: after the first, which is considered only as biscuit, the painter retouches it, to correct any faults in the colouring ; it is then subjected to the second heating. Very highly finished paintings are often burned a greater number of times. Only fine pottery, such as porcelain, is painted; but engravings are transferred on earthenware, even of the commoner sorts. An ordinary engraved copperplate is used for this, and the ink em- ployed consists of glass, coloured broAvn, black, red, or blue, etc., reduced to an impalpable powder, and ground in linseed-oil. The engraving is printed on a sheet of thin paper, the engraved side POTTERY. Fig. 460. 671 of which, after moistening the sheet, is impressed on the dry pottery; by carefully removing the paper, the design is left on the vessel. The oil is then driven off by heat, and the glazing done in the ordinary way. ORNAMENTS AND PAINTING. § 731. The paste of earthenware contains the same elementary substances as glass, and differs from the latter only in the propor- tions of its components : the ceramic pastes are therefore analyzed by the processes indicated in § 704. CHEMICAL ANALYSIS OF EARTHENWARE. END OF VOL. I.