STUDIES IN CENTRAL AMERICAN PICTURE-WRITING BY EDWARD S. HOLDEN Professor of Mathematics U. S. Naval Observatory [EXTRACTED FROM THE FIRST ANNUAL REPORT OF THE BUREAU OF ETHNOLOGY] WASHINGTON GOVERNMENT PRINTING OFFICE 1881 SMITHSONIAN INSTITUTION BUREAU OF ETHNOLOGY. J. W. POWELL, DIRECTOK. STUDIES IN CENTRAL AMERICAN PICTURE-WRITING. BY EDWARD S. HOLDEN, PROFESSOR OF MATHEMATICS, U. S. NAVAL OBSERVATORY. LIST OF ILLUSTRATIONS Page. Figure 48.—The Palenquean Group of the Cross 221 49. —Statue at Copau 224 50. —Statue at Copan 225 51. Synonymous Hieroglyphs from Copan and Palenque 227 52. —Yucatec Stone 229 53. —Huitzilopochtli (front) 232 54. —Huitzilopochtli (side) 232 55. —Huitzilopochtli (back) 232 56. —Miclantecutli 232 57. —Adoratorio 233 58. —The Maya War-God 234 59. —The Maya Rain-God 234 60. —Tablet at Palenque 234 STUDIES IN CENTRAL AMERICAN PICTURE-WRITING. Edward S. Holden. I. Since 1876 I have been familiar with the works of Mr. John L. Ste- phens on the antiquities of Yucatan, and from time to time I have read works on kindred subjects with ever increasing interest and curiosity in regard to the meaning of the hieroglyphic inscriptions on the stones and tablets of Copan, Palenque, and other ruins of Central America. In August, 1880, I determined to see how far the principles which are suc- cessful when applied to ordinary cipher-'writing would carry one in the inscriptions of Yucatan. The difference between an ordinary cipher- message and these inscriptions is not so marked as might at first sight appear. The underlying principles of deciphering are quite the same in the two cases. The chief difficulty in the Yucatec inscriptions is our lack of any defi- nite knowledge of the nature of the records of the aborigines. The pa- tient researches of our archaeologists have recovered but very little of their manners and habits, and one has constantly to avoid the tempting suggestions of an imagination which has been formed by modern influ- ences, and to endeavor to keep free from every suggestion not inherent in the stones themselves. I say the stones, for I have only used the Maya manuscripts incidentally. They do not possess, to me, the same interest, and I think it may certainly be said that all of them are younger than the Palenque tablets, and far younger than the inscriptions at Copan. I therefore determined to apply the ordinary principles of deciphering, without any bias, to the Yucatec inscriptions, and to go as far as I could certainly. Arrived at the point where demonstration ceased, it would be my duty to stop. For, while even the conjectures of a mind per- fectly trained in archseologic research are valuable and may subse- quently prove to be quite right, my lack of familiarity with historical works forced me to keep within narrow and safe limits. My programme at beginning was, first, to see if the inscriptions at Copan and Palenque were written in the same tongue. When I say u to see,” I mean to definitely prove the fact, and so in other cases j second, to see how the tablets were to be read. That is, in horizontal lines, are 208 CENTRAL AMERICAN PICTURE-WRITING. they to be read from right to left, or the reverse1? In vertical columns, are they to be read up or down ? Third, to see whether they were phonetic characters, or merely ideographic, or a mixture of the two— rebus-like, in fact. If the characters turned out to be purely phonetic, I had determined to stop at this point, since I had not the time to learn the Maya language, and again because I utterly and totally distrusted the methods which, up to this time, have been applied by Brasseur de Bourbourg and others who start, and must start, from the misleading and unlucky alphabet handed down by Landa. I believe that legacy to have been a positive misfortune, and I believe any process of the kind attempted by Brasseur de Bourbourg (for example, in his essay on the MS. Troano) to be extremely dangerous and difficult in application, and to require a degree of scientific caution almost unique. Dr. Harrison Allen, in his paper, u The Life Form in Art,” in the Transactions of the American Philosophical Society, is the only investiga- tor who has applied this method to Central American remains with suc- cess, so it seems to me ; and even here errors have occurred. The process I allude to is something like the following: A set of char- acters, say the alphabet of Landa, is taken as a starting point. The variants of these are formed. Then the basis of the investigation is ready. From this, the interpretation follows by identifications of each new character with one of the standard set or with one of its variants. Theoretically, there is no objection to this procedure. Practically, also, there is no objection if the work is done strictly in the order named. In fact, however, the list of variants is filled out not before the work is begun, but during its progress, and in such a way as to satisfy the nec- essities of the interpreter in carrying out some preconceived idea. With a sufficient latitude in the choice of variants any MS. can receive any in- terpretation. For example, the MS. Troano, which a casual examina- tion leads me to think is a ritual, and an account of the adventures of several Maya gods, is interpreted by Brasseur de Bourbourg as a record of mighty geologic changes. It is next to impossible to avoid errors of this nature at least, and in fact they have not been avoided, so far as I know, except by Dr. Allen in the paper cited. I, personally, have chosen the stones and not the manuscripts for study largely because variants do not exist in the same liberal degree in the stone inscriptions as they have been supposed to exist in the manuscripts. At any one ruin the characters for the same idea are alike, and alike to a marvelous degree. At another ruin the type is just a little different, but the fidelity to this type is equally great. Synonyms exist; that is, the same idea may be given by two or more utterly different signs. But a given sign is made in a fixed and definite way. Finally the MSS. are, I think, later than the stones. Hence the root of the matter is the inter- pretation of the stones, or not so much their full interpretation as the discovery of a method of interpretation, which shall be sure. HOLDEN.] THE MAYA HIEROGLYPHS. 209 Suppose, for example, that we know the meaning of a dozen characters only, and the way a half dozen of these are joined together in a sentence. The method by which these were obtained will serve to add others to the list, and progress depends in such a case only on our knowledge of the people who wrote, and of the subjects upon which they were writing. Such knowledge and erudition belongs to the archaeologists by profes- sion. A step that might take me a year to accomplish might be made in an instant by one to whom the Maya and Aztec mythology was famil- iar, if he were proceeding according to a sound method. At the present time we know nothing of the meaning of any of the Maya hieroglyphs. It will, therefore, be my object to go as far in the subject as I can pro- ceed with certainty, every step being demonstrated so that not only the archaeologist but any intelligent person can follow. As soon as the border-land is reached in which proof disappears and opinion is the only guide, the search must be abandoned except by those whose cultivated and scientific opinions are based on knowledge far more profound and various than I can pretend or hope to have. If I do not here push my own conclusions to their farthest limit, it must not be assumed that I do not see, at least in some cases, the direc- tion in which they lead. Rather, let this reticence be ascribed to a de- sire to lay the foundations of a new structure firmly, to prescribe the method of building which my experience has shown to be adequate and necessary, and to leave to those abler than myself the erection of the superstructure. If my methods and conclusions are correct (and I have no doubts on this point, since each one lias been reached in various ways and tested by a multiplicity of criteria) there is a great future to these researches. It is not to be forgotten that here we have no Rosetta stone to act at once as key and criterion, and that instead of the accurate de- scriptions of the Egyptian hieroglyphics which were handed down by the Greek cotemporaries of the sculptors of these inscriptions, we have only the crude and brutal chronicles of an ignorant Spanish soldiery, or the bigoted accounts of an unenlightened priesthood. To Cortez and his companions a memorandum that it took one hundred men all day to throw the idols into the sea was all-sufficient. To the Spanish priests the burning of all manuscripts was praiseworthy, since those differing from Holy Writ were noxious and those agreeing with it superfluous. It is only to the patient labor of the Maya sculptor who daily carved the symbols of his belief and creed upon enduring stone, and to the lux- uriant growths of semi-tropical forests which concealed even these from the passing Spanish adventurer, that we owe the preservation of the memorials of past beliefs and vanished histories. Hot the least of the pleasures of such researches as these comes from the recollection that they vindicate the patience and skill of forgotten men, and make their efforts not quite useless. It was no rude savage that carved the Palenque cross; and if we can discover what his efforts 210 CENTRAL AMERICAN PICTURE-WRITING. meant,-his labor and his learning have not been all in vain. It will be one more proof that human effort, even misdirected, is not lost, but that it comes, later or earlier, u to forward the general deed of man.” II. MATERIALS FOR TIIE PRESENT INVESTIGATION. My examination of the works of Mr. J. L. Stephens has convineed me that in every respect his is the most trustworthy work on the hieroglyphs of Central America. The intrinsic evidence to this effect is very strong, but when I first became familiar with the works of Wal- deck I found so many points of difference that my faith was for a time shaken, and I came to the conclusion that while the existing representa- tions might suffice for the study of the general forms of statues, tablets, and buildings, yet they were not sufficiently accurate in detail to serve as a basis for the deciphering I had in mind. I am happy to bear witness, however, that Stephens’s work is undoubtedly amply adequate to the purpose, and this fact I have laboriously verified by a comparison of it with various representations, as those of Desaix and others, and also with a few photographs. The drawings of Waldeok are very beautiful and artistic, but either the artist himself or his lithographers have taken singular liberties in the published designs. Stephens’s work is not only accurate, but it contains sufficient material for my purpose (over 1,500 separate hieroglyphs), and, therefore, 1 have based my study ex- clusively upon his earliest work, u Incidents of Travel in Central America, Chiapas, and Yucatan,” 2 vols., 8vo. New York, 1842 (twelfth edition). I have incidentally consulted the works on the subject contained in the Library of Congress, particularly those of Brasseur he Bourbourg, Kingsborough, Waldeck, and others, but, as I have said, the two vol- umes above named contain all the the material I have been able to utilize, and much more which is still under examination. One fact which makes the examination of the Central American anti- quities easier than it otherwise would be, has not, I think, been suffi- ciently dwelt upon by former writers. This is the remarkable faithful- ness of the artists and sculptors of these statues and inscriptions to a standard. Thus, at Copan, wherever the same kind of hieroglyph is to bo represented, it will be found that the human face or other object employed is almost identically the same in expression and character, wherever it is found. The same characters at different parts of a tablet do not differ more than the same letters of the alphabet in two fonts of type. At Palenquo the type (font) changes, but the adherence to this is equally or almost equally rigid. It is to be presumed that in this latter HOLDEN. J THE MAYA HIEROGLYPHS. 211 case, where work was done both in stone and stucco, the nature of the material affected the portraiture more or less. The stone statues at Copan, for example, could not all have been done by the same artist, nor at the same time. I have elsewhere shown that two of these statues are absolutely identical. How was this ac- complished? Was one stone taken to the foot of the other and cut by it as a pattern ? This is unlikely, especially as in the case mentioned the scale of the two statues is quite different. I think it far more likely that each was cut from a drawing, or series of drawings, which must have been preserved by priestly authority. The work at any one place must have required many years, and could not have been done by a single man; nor is it probable that it was all done in one generation. Separate hieroglyphs must have been preserved in the same way. It is this rigid adherence to a type, and the banishment of artistic fancy, which will allow of progress in the deciphering of the inscriptions or the com- parison of the statues. Line after line, ornament after ornament, is re- peated with utter fidelity. The reason of this is not far to seek. This, however, is not the place to explain it, but rather to take advantage of the fact itself. We may fairly say that were it not so, and with our present data, all advances would be tenfold more difficult. m. SYSTEM OF NOMENCLATURE. It is impossible without a special and expensive font of type to refer pictorially to each character, and therefore some system of nomencla- ture must be adopted. The one I employ I could now slightly improve, but it has been used and results have been obtained by it. It is suffi- cient for the purpose, and I will, therefore, retain it rather than to run the risk of errors by changing it to a more perfect system. I have num- bered the plates in Stephens’s Central America according to the follow- ing scheme: ENGRAVINGS OK VOLUME I. Page. Stone Statue, front view, I have called Plate I - Frontispiece. Wall of Copan, Plate II 96 Plan of Copan, Plate III 133 Death’s Head, Plate III * 135 Portrait, Plate IIIb 136 Stone Idol, Plate IV 138 Portrait, Plate IV a 139 Stone Idol, Plate Y 140 Tablet of Hieroglyphics, Plate Va 141 No. 1, Sides of Altar, Plate VI 142 No. 2, Sides of Altar, Plate VII , 142 Gigantic Head, Plate VIII „ , - 143 212 CENTRAL AMERICAN PICTURE-WRITING. Page. No. 1, Stone Idol, front view, Plate IX 149 No. 2, Stone Idol, back view, Plate X 150 Idol half buried, Plate XI 151 No. 1, Idol, Plate XII 152 No. 2, Idol, Plate XIII 152 No. 1, Idol, Plate XIV 153 No. 2, Idol, Plate XV 153 Idol and Altar, Plate XVI 154 Fallen Idol, Plate XVII - - - 155 No. 1, Idol, front view, Plate XVIII 156 No. 2, Idol, back view, Plate XIX 156 No. 3, Idol, side view, Plate XX 156 Fallen Idol, Plate XX * 157 Circular Altar, Plate XX b 157 No. 1, Stone Idol, front view, Plate XXI 158 No. 2, Stone Idol, back view, Plate XXII 158 No. 3, Stone Idol, side view, Plate XXIII 158 Great Square of Antigua Guatimala, Plate XXIIIa 266 Profile of Nicaragua Canal, Plate XXIIIb 412 ENGRAVINGS OF VOLUME II. . Page. Stone Tablet, Plate XXIV Frontisjriece. Idol at Quirigua, Plate XXV 121 Idol at Quirigua, Plate XXVI 122 Santa Cruz del Quich6, Plate XXVII 171 Place of Sacrifice, Plate XXVIII 184 Figures found at Santa Cruz del Quiche, Plate XXIX 185 Plaza of Quezaltenango, Plate XXX 204 Vases found at Gueguetenango, Plate XXXI 231 Ocosingo, Plate XXXII 259 Palace at Palonque, Plate XXXIII 309 Plan of Palace, Plate XXXIV 310 Stucco Figure on Pier, Plate XXXV 311 Front Corridor of Palace, Plate XXXVI 313 No. 1, Court-yard of Palace, Plate XXXVIII 314 No. 2, Colossal Bas-reliefs in Stone, Plate XXXIX 314 East side of Court-yard, Plate XXXVII - 314 No. 1, Bas-relief in Stucco, Plate XL 316 No. 2, Bas-relief in Stucco, Plate XLI 316 No. 3, Bas-relief in Stucco, Plate XLII 316 Oval Bas-relief in Stone, Plate XLIII 318 Bas-relief in Stucco, Plate XLIV 319 General Plan of Palenque, Plato XLV .1 337 Casa No. 1 in Ruins, Plate XLVI 338 Casa No. 1 restored, Plate XLVII 339 No. 1, Bas-relief in Stucco, Plate XLVIII 340 No. 2, Bas-relief in Stucco, Plate XLIX 1 340 No. 3, Bas-relief in Stucco, Plate L 340 No. 4, Bas-relief in Stucco, Plate LI 340 No. 1, Tablet of Hieroglyphics, Plate LII 342 No. 2, Tablet of Hieroglyphics, Plate LIII 342 Tablet on inner Wall, Plate LIV 343 Casa di Piedras, No. 2, Plate LV 344 Tablet on back Wall of Altar, Casa No. 2, Plate LVI 345 Stone Statue, Plate LVII 349 HOLDEN.] THE MAYA HIEROGLYPHS. Page. Casa No. 3, Plate LVIII 350 Front Corridor, Plate LIX 351 No. 1, Bas-reliefs in Front of Altar, Plate LX 353 No. 2, Bas-reliefs in Front of Altar, Plate LXI 353 Ad oratorio or Altar, Plate LXII 354 Casa No. 4, Plate LXIII 355 House of the Dwarf, Plate LXIV 420 Casa del Gobernador, Plate LXV 428 Sculptured Front of Casa del Gobernador, Plate LXVI 443 Egyptian Hieroglyphics, Plate LXVIII 441 Top of Altar at Copan, Plate LXVIII = Va 454 Mexican Hieroglypbical Writing, Plate LXIX 454 In each plate I have numbered the hieroglyphs, giving each one its own number. Thus the hieroglyphs of the Copan altar (vol. i, p. 141) which I have called plate Ya, are numbered from 1 to 36 according to this scheme— 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 And the right hand side of the Palenque Cross tablet, as given by Rau in his memoir published by the Smithsonian Institution (1880), has the numbers— 2020 2021 2022 2023 2024 2025 2030 2031 2032 2033 2034 2035 2040 2041 2042 2043 2044 2045 2050 2051 2052 2053 2054 2055 ####*# *##### 3080 3081 3082 3083 3084 3085 These are consecutive with the numbers which I have attached to the left-hand side, as given by Stephens. Whenever I have stated any results here, I have also given the means by which any one can number a copy of Stephens’s work in the way which I have adopted, and thus the means of testing my conclusions is in the hands of every one who desires to do so. In cases where only a part of a hieroglyphic is referred to, I have placed its number in a parenthesis, as 1826 see (122), by which I mean that the character 1826 is to be compared with a part of the character 122. The advantages of this system are many: for example; a memo- randum can easily be taken that two hieroglyphs are alike, thus 2072 = 2020 and 2073 =2021. Hence the pair 2020 — 2021, read hori- zontally, occurs again at the point 2072 — 2073, etc. Horizontal pairs will be known by their numbers being consecutive, as 2020—2021; verti- cal pairs will usually be known by their numbers differing by 10. Thus, 2075 — 2085 are one above the other. CENTRAL AMERICAN PICTURE-WRITINC. This method of naming the chiffres, then, is a quick and safe one, and we shall see that it lends itself to the uses required of it. I add here the scheme according to which the principal plates at Pa- lenque have been numbered. PLATE XXIV (left-liand side). r I7 37 38 30 04 96 08 100 102 104 106 1 &e 1800 See 1800 See 1806 : .. l 40 40 41 42 95 07 90 = 127 101 103 105 107 43 = 1810 43“ = 46“. 44 45 108 See 91 46=1810 46“ = 43“ 47 48 40 50 51 In the middle of the 52 52* = 1820 ? 53 54 plato at the top. 55 56 = 1840? 57 58 100 115 "" See 1802 50 60 61 62 = 58? 110 116 See 2020 63 64 651 66 111 117 See 2025 67 68 60 70 112 118 See 1911 71 72 = 281 73 74 113 110 See 2020 75 76 = 67 77 78 114 120 70 80 81 82 83 84 85 86 = 56? 86* 86* 87 88 89 00 01 92 03 * Accidental error in numbering liere. t Possibly Muluc—a Maya day; the meaning is “reunion.” HOLDEN.] THE MAYA HIEROGLYPHS. 215 PLATE XXIV (right-hand side). 121 See 74,86* 122=86?) 123 = 87 124 = 88’ See 01,1822 125 1261 See 1940 127 = 00 12S See 1940 See (44), 64 12!) 130 131 = 147 132 See 50, 58, 62 133 134 135 136=47? 137 138 See 39, 91 130 140 See 1811 141 142 § See 54 143 144 See 50, 58, 62,132 145 146 147 = 131 148 See 71 14!) 150 See 56,1882 151 152 153 154 See 53 155 156 See 50,58,132 157* 158 See 68 150 160 See 38 See 46», 49*, 52* 161 = 50 162 See 58, 62,132 See 50, 73,1882 1163=1036 164 See 57 See 58, 62 165 166 See 81 ? 167 168 16!) See 68? 170 171 ' 172 173 174 See 67,76,90,1910 175 176 See 57 See 126 177 178 See 43* 170 180 See 50, 58,62 181 182 See 57,163,1936 183 184 185 * Possibly Ymix— t Possibly Ohuen- 1 Possibly Abau— § Possibly Ezanat -a Maya day. -a Maya day; meaning “a board,” “a tree.” -a Maya day; meaning “king.” —a Maya day. 216 CENTRAL AMERICAN PICTURE-WRITING. PLATE LII. 200 201 202 203 204 205 206 207 208 209 >Line 1. 210 211 212 214 215 216 217 218 219 See 2020 220 221 222 223 224=2060 225 226 227 228 229 See 2030 See 2060 See 1811-2 • Lino 2. 230 See 1822 231 232 234 235 236 237 238 239 240 241 242=2020 243=1951 244 245 246 247 248 249 • Line 3. 250 251 252 254 255 256 257 258 259-1943 J See 214 2(50 261 262 263 264 265 266 267 268 269 See 2020 See 2021 See 2022 • Line 4. 270 271 274=244 275 276 277 278 See 204 279 280 281 = 72 282 283 284 285 286 287 288 See 1820 See 385 • Line 5. 290 294 295 296 297 298 299 300 301 302 303=360 304 305 306 307 See 203 ■ Line 6. 310 311 314 315 316 317 318 319 320 321 322 323 324=1824 325 326 327 328 329 See 203 See 204 See 285 See 305 ■ Line 7. 330 331 332 See 209 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 See 209 See 322 • Line 8. 350 351 352 354 355 356=1822 357 358 359 See 267, 298 See 230 360=303 361 362 363 364 365 366 367 368 369 370 See 351 See 303, 360 • Line 9. 371 375 376 377 378 379 J 380 390 381 382 383 384 385 See 286, 1822 386 387 388 389 Line 10. 391 392 394 395 396 397 398 399 400 401 402 403 = o«7 404 405 406 407 40S 409 See 326 See 360 • Line 11. 410 411 412 414 415 416 417 418 419 See 326 See 324 420 421 422 423 424 425 426 See 324 427 1 ■Line 12. 430 432 434 435 436 437 438 439 HOLDEN.] THE MAYA HIEROGLYPHS. PLATE LIII. [The upper left-hand square is hTo. 500, the upper right is 519, the lower left-hand is 720, the lower right is 739. All the squares from 500 to 508, 520 to 528, 530 to 538, etc., up to 720 to 728, are obliterated (and their numbers omitted here) except a few.] 509 510 511 512 See 1967 513 514 515 See 509 516 See 510 517 518 519 529 See 3012 530 531 532 533 534 535 536 537 538 539 549 550 551 552 553 554 555 556 See 162 557 558 559 ' 570 571 572 573 See 1823 574 575 576 577 578 579 589 590 591 592 593 594 595 596 597 598 599 604 605 609 610 611 See 571 612 613 614 615 616 617 618 619 628 629 630 631 632 633 634 635 636 See 3054 637 638 639 649 650 651 652 653 654 655 See 150,1882 656 657 658 659 669 670 671 = 324 See 2042 672 = 322? 673 = 323? 674 See 77 675 676 677 See 1802 678 679 688 689 690 691 692 693 694 695 696 697 698 699 708 709 710 711 712 713 = 1802 714 715 716 717 See 439 718 719 729 730 = 1845 731 732 733 734 735 736 737 See 2020 738 739 CENTRAL AMERICAN PICTURE-WRITING. , PLATE LIV. soo 801 S02 S03 S04 S05 SOO S07 808 See 1882 809 ‘ 810 811 See 26 812 See 1940 813 See 1941,3011 900 901 902 903 904 905 906 907=1003 908 See 2020 909 910 See 1310 911 912 913 iooo 1001 1002 1003=907 1004 1005 1006 1007 100S 1009 See 2021 1010 See 3054 1011 See 1811-2 1012 1013 1100 1101 1102=717 1103 1104 See 1820 1105=2020 1106 See 2021 1107 See 1840 1108 See 1841? 1109 1110 = 1209 1113 1114 1115 1*200 1201 1202=1110 See 3054 1203 1204=1008 1205 1206 1207 See 1823 120S 1209=1110 1210 1211 1212 1213 1300 1301 1302 1303=1910 1304 1305 1306 1307 130S 1309 1310 See 910 1311 1312 1313 1400=1823 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1500 1501 1502=1010 1503 15°4={i02 1505 1506 1507 1508 1509 1510 1511 1512 1513 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609=1304 1610=1305 1611=1010 1612 1613 1700 1701 1702=1911 1703 1704 1705 1706 1707 1708 1709 1710 1711 = 15®? 1712 = 170S 1713 HOLDEN.] THE MAYA HIEROGLYPHS. 1S00 | 1 l 1S01 1S05 See 155 1S02 See 163,175 1S06 1803 1807 See 138 1504 1505 1961 1962 1963 1964 1965 1966 * 1S10 See 150 1811 See 139,179 1S12 See (1852) 1S13 See 131,146 1814 /See 126,127,176 1S15 1816 1967 1820 See 161 1821 1822 See 124 1S23 1S24 1825 1S26 See 122,160 196S 1S30 = 1S20 See 161 1S31 1S32 See 123,124 1S33 See 121 1834 See 163 1835 See 182 1836 See 123 1969 1S40 1S41 1842 See 1835 1843 See 124,1836 1S44 1845 = 1822 See 124 1846 See 179 1970 1S50 1851 1S52 1853 See 122 1854 = 1S06 1855 1S60 1861 1S62 See 126,127 1863 1S64 1865 = 2021 See 144 1866 See 136?, 184? 1870 = 1820 See 160,161 1S71 1872 = 1842? See 182 1873 = 1S03 1S74 1875 1S76 1880 1881 1882 See 150,162 1883 See 124 1884 = 1834 See 163,182 1885 See 132,144 1890 See 130,158 1891 See 131?, 147? 1S1V2 See 132? 1893 1894=1822 See 124 1895 See 144 1900 See 146 1901 1902 1903 See 157,182 1904 1905 = 1803 1971 See 1802 1910 See 174 1911 See 174 1912 See 141 & Co., Sc. Stiff alo. Fig. 48.—The Palenquean Group of the Cross. HOLDEN. 1 THE MAYA HIEROGLYPHS. 221 IY. Before any advance can be made in the deciphering of the hiero- glyphic inscriptions, it is necessary to know in what directions, along what lines or columns, the verbal sense proceeds. All the inscriptions that I know of are in rectangular figures. At Copan they are usually in squares. At Palenque the longest inscriptions are in rectangles. At Palenque again, there are some cases where there is a single horizontal line of hieroglyphs over a pictorial tablet. Here clearly the only question is, do the characters proceed from left to right, or from right to left ? In other cases as in the tablet of the cross, there are vertical columns. The question here is, shall we read up or down ? Now, the hieroglyphs must be phonetic or pictorial, or a mixture of the two. If they are phonetic, it will take more than one symbol to make a word, and we shall have groups of like characters when the same word is written in two places. If the signs are pictorial, the same thing will follow; that is, we shall have groups recurring when the same idea recurs. Further, we know that the subjects treated of in these tablets must be comparatively simple, and that names, as of gods, kings, etc., must necessarily recur. The names, then, will be the first words deciphered. At present no single name is known. These considerations, together with our system of nomenclature, will enable us to take some steps. Take, for example, the right-hand side of the Palenque cross tablet as given by Bau. See our figure 48, which is Plate LYI of Stephens (vol. ii, p. 345), with the addition of the part now in the National Mu- seum at Washington. IN WHAT ORDER ARE THE HIEROGLYPHICS READ? Our system of numbering is here 2020 2021 2022 2023 2024 2025 2030 2031 2032 2033 2034 2035 #*###* ****** ****** 3080 3081 3082 3083 3084 3085 Now pick out the duplicate hieroglyphs in this; that is, run through the tablet, and wherever 2020 occurs erase the number which fills the place and write in 2020. Do the same for 2021, 2022, etc., down to 3084. The result will be as follows: 222 CENTRAL AMERICAN PICTURE-WRITING. RIGHT-HAND SIDE OF PALENQUE CROSS TABLET (Rau). 2020 2021 2022 2023 2024 2025 V"‘ 2030 2031 2032 2033 2034 2035 2040 2041 2042 r 2025 2020 2021 2050 2051 2034 [ 2053 2054 2055 2053 2001 2002 2003 2004 2005 2070 2071 2020 2021 2022? 2024? j 2053 2020 2082 2083 2025 r 2053 J 2021 2091 2092 J 2025 2094 2095 3000 2023 2031 l 2053 2033 3005 3010 20S3 3012 2024 3014 2091 2053 3021 2023 2020 3024 2024 r 2024 2025 2021 3033 r 2025 2034* ?| 2053* 3021 3042 3043 1 2035 3045 3050 See 2082 2083 j' 2025 2034 3054 3055 2024 2020 l 2035 3003 2024 2025 2021 2031 2020 2021 2035 3045 3080 3081 2091 2093 2020 2021 14 cases of horizontal pairs; 4 cases of vertical pairs; 102 characters in all, of which 51 ap- pear more than once, so that there are hut 51 independent hieroglyphs. Hero the first two lines are unchanged. In the third line we find that 2043 is the same as 2025, 2044 = 2020, 2045 = 2021, and so on, and we write the smallest number in each case. After this is done, connect like pairs by braces whenever they are consecutive, either vertical or horizontal. Take the pair 2020 and 2021 for example; 2020 occurs eight times in the tablet, viz, as 2020, 2044, 2072, 2081, 3023, 3001, 3072, 3084. In five out of the eight cases, ib is followed by 2021, viz, as 2021, 2045, 2073, 3073, 3085. It is clear this is not the result of accident. The pair 2020 and 2021 means something, and when the two characters occur together they must be read together. There is no point of punctuation between them. We HOLDEN.] THE MAYA HIEROGLYPHS. 223 also learn that they are not inseparable. 2020 will make sense with 2082, 3024, and 3062. Here it looks as if the writing must be read in lines horizontally. We do not know yet in which direction. We must examine other cases. This is to be noticed: If the reading is in horizontal lines from left to right, then the progress is from top to bottom in columns, as the case of 3035 and 3040 shows. This occurs at the end of a line, and the corresponding chiffre required to make the pair is at the other end of the next line. I have marked this case with asterisks. If we must read in the lines from right to left we must neces- sarily read in columns from bottom to top. Thus the lines are connected. A similar process with all the other tablets in Stephens leads to the conclusion that the reading is in lines horizontally and in columns ver- tically. The cases 1835-’45, 1885-’95, 1914-’24, and 1936-’46 should, however, be examined. We have now to decide at which end of the lines to begin. The reasons given by Mr. Bancroft (Native Races, vol. ii, p. 782) appeared to me sufficient to decide the question before I was acquainted with his statement of them. Therefore, the sum total of our present data, examined by a rational method, leads to the conclusion, so far as we can know from these data, that the verbal sense proceeded in lines from left to right, in columns from top to bottom j just as the present page is written, in fact. For the present, the introduction of the method here indicated is the important step. It has, as yet, been applied only to the plates of Stephens’ work. The definite conclusion should be made to rest on all possible data, some of which is not at my disposition at present. Tab- lets exist in great numbers at other points besides Palenque, and for the final conclusion these must also be consulted. If each one is examined in the way I have indicated, it will yield a certain answer. The direction of reading for that plate can be thus determined. At Palenque the progress is in the order I have indicated. y. THE CARD-CATALOGUE OE HIEROGLYPHS. It lias already been explained liow a system of nomenclature was gradually formed. As I have said, tills is not perfect, but it is suffi- ciently simple and full for the purpose. By it, every plate in Stephens’ work receives a number and every hieroglyph in each plate is likewise numbered. This was first done in my private copy of the work. I then procured another copy and duplicated these numbers both for plates and single chiffres. The plates of this copy were then cut up into single hiero- 224 CENTRAL AMERICAN PICTURE-WRITING. glyphs and each single hieroglyph was mounted on a library card, as follows: No. 2020. Hieroglyph. Plate LV1. Same as Numbers. Similar to Num- bers. ' The cards were 6.5 by 4.5 inches. The chiffre was pasted on, in the center of the top space. Its number and the plate from which it came were placed as in the cut. The numbers of hieroglyphs which resembled the one in question could be written on the right half of the card, and the numbers corresponding to different recurrences of this hieroglyph occupied the left half. All this part of the work was most faithfully and intelligently per- formed for me by Miss Mary Lockwood, to whom I desire to express the full amount of my obligations. A mistake in any part would have been fatal. But no mistakes occurred. These cards could now be arranged in any way I saw fit. The simple chiffres, for example, could be placed so as to bring like ones together. A compound hieroglyph could be placed among simple ones agreeing with any one of its components, and so on. The expense of forming this card catalogue of about 1,500 single hie- roglyphs was borne by the Ethnological Bureau of the Smithsonian In- stitution, and the catalogue is the property of that bureau, forming only one of its many rich collections of American picture-writings. VI. COMPARISON OF PLATES I AND IT (COPAN). In examining the various statues at Copan, as given by Stephens, one naturally looks for points of striking resemblance or striking differ- ence. Where all is unknown, even the smallest sign is examined, in the hope that it may prove a clue. The Plate I, Fig. 49, has a twisted knot (the “square knot” of sailors) of cords over its head, and above this is a cliiffre composed of ellipses, and above this again a sign like a sea-shell. A natural suggestion was that these might be the signs for the name of the personage depicted in Plate I. If this is so and we should find the same sign elsewhere in connection with a figure, we should expect to find this second figure like the first in every particular. This would be Fig. 49.—Statue at Copan Fig. 50.—Statue at Copan. HOLDEN.] THE MAYA HIEROGLYPHS 225 a rigid test of the theory. After looking through the Palenque series, and finding no similar figure and sign, I examined the Copan series, and in Plate IV, our Pig. 50, I found the same signs exactly; i. e., the knot and the two chiffres. At first sight there is only the most general resemblance between the personages represented in the two plates; as Stephens says in his orig- inal account of them, they are u in many respects similar.” If he had known them to be the same, he would not have wasted his time in drawing them. The scale of the two drawings and of the two statues is different; but the two personages are the same identically. Figure for figure, or- nament for ornament, they correspond. It is unnecessary to give the minute comparison here in words. It can be made by any one from the two plates herewith. Take any part of Plate I, find the corresponding part of Plate IV, and whether it is human feature or sculptured orna- ment the two will be found to be the same. Take the middle face depending from the belt in each plate. The earrings are the same; the ornament below the chin, the knot above the head, the complicated beadwork on each side of this face, all are the same. The bracelets of the right arms of the main figures have each the forked serpent tongue, and the left arm bracelets are ornamented alike. The crosses with beads almost inclosed in the right hands are alike; the elliptic ornaments above each wrist, the knots and cliiffres over the serpent masks which surmount the faces, all are the same. In the steel plates given by Stephens there are even more coindences to be seen than in the excellent wood-cuts here given, which have been copied from them. Here, then, is an important fact. The theory that the chiffre over the forehead is characteristic, though it is not definitively proved, receives strong confirmation. The parts which have been lost by the effects of time on one statue can be supplied from the other. Better than all, we gain a test of the minuteness with which the sculptors worked, and an idea of how close the adherence to a type was required to be. Granting once that the two personages are the same (a fact about which I con- ceive there can be no possible doubt, since the chances in favor are lit- terally thousands to one), we learn what license was allowed, and what synonyms in stone might be employed. Thus, the ornament suspended from the neck in Plate IV is clearly a tiger’s skull. That from the neck of Plate I has been shown to be the derived form of a skull by Dr. Harrison Allen,* and we now know that this common form relates not to the human skull, as Dr. Allen has supposed, but to that of the tiger. We shall find this figure often repeated, and the identification is of importance. This is a case in regard to synonyms. The kind of symbolism so ably treated by Dr. Allen is well exemplified in the con- ventional sign for the crotalus jaw at the mouth of the mask over the head of each figure. This is again found on the body of the suake in *Tlie Life Form in Art, Trans. Amer. Phil. Soc., yoI. xv, 1873, p. 325. 226 CENTRAL AMERICAN PICTURE-WRITING. Plate LX, and in other places. Other important questions can bo settled by comparison of the two plates. For example, at Palenque we often find a sign conqiosed of a half ellipse, inside of which bars are drawn. I shall elsewhere show that there is reason to believe the ellipse is to represent the concave of the sky, its diameter to be the level earth, and in some cases at least the bars to be the descending and fertilizing rain. The bars are some- times two, three, and sometimes four in number. Are these variants of a single sign, or are they synonyms ? Before the discovery of the identity of the personages in these two plates, this question could not be answered. Xow we can say that they are not synonyms, or at least that they must be considered separately. To show this, examine the bands just above the wristlets of the two figures. Over the left hands of the figures the bars are two in number$ over the right hands there are four. This exact similarity is not accidental; there is a meaning in it, and we must search for its explanation elsewhere, but we now have a valuable test of what needs to be regarded, and of what, on the other hand, may be passed over as accidental or unimportant. One other case needs mentioning here, as it will be of future use. From the waist of each figure depend nine oval solids, six being hatched over like pine cones and the three central ones having two ovals, one within the other, engraved on them. In Plate IV the inner ovals are all on the right-hand side of the outer ovals. Would they mean the same if they were on the left-hand side H Plate I enables us to say that they would, since one of these inner ovals has been put by the artist on that side by accident or by an allowed caprice. It is by furnishing us with tests and criteria like these that the proof of the identity of these two plates is immediately important. In other ways, too, the proof is valuable and interesting, but we need not discuss them at this time. These statues, then, are to us a dictionary of synonyms in stone—a test of the degree of adherence to a prototype which was exacted, and a criterion of the kind of minor differences which must be noticed in any rigid study. I have not insisted more on the resemblances, since the accompanying figures present a demonstration. Let those who wish to verify these re- semblances compare minutely the ornaments above the knees of the two figures, those about the waists, above the heads,, and the square knots, etc., etc. HOLDEN.] THE MAYA HIEROGLYPHS. 227 VII. ARE THE HIEROGLYPHS OF COPAN AND PALENQUE IDENTICAL ? One of the first questions to be settled is whether the same system of writing was employed at Palenque and at Copan. Before any study of the meanings of the separate chiffres cam be made, we must have our material properly assorted, and must not include in the figures we are examining for the detection of a clue, any which may belong to a system possibly very different. The opinion of Stephens and of later writers is confirmed by my comparison of the Palenque and the Copan series; that is, it becomes evident that the latter series is far the older. In Nicaragua and Copan the statues of gods were placed at the foot of the pyramid; farther north, as at Palenque, they were placed in temples at the summit. Such differences show a marked change in customs, and must have required much time for their accomplishment. In this time did the picture-writing change, or, indeed, was it ever identical To settle the question whether they were written on the same system, I give here the results of a rapid survey of the card-catalogue of hiero- glyphs. A more minute examination is not necessary, as the present one is quite' sufficient to show that the system employed at the two places was the same in its general character and almost identical even in details. The practical result of this conclusion is that similar char- acters of the Copan and Palenque series may be used interchangeably. A detailed study of the undoubted synonyms of the two places will afford much light on the manner in which these characters were gradu- ally evolved. This is not the place for such a study, but it is interesting to remark how, even in un- mistakable synonyms, the Pa- lenque character is always the most conventional, the least pictorial; that is, the latest. Examples of this are No. 7, Plate Va, and No. 1969, Plate LYI. The mask in profile which forms the left-hand edge of No. 7 seems to have been conventionalized into the two hooks and the ball, which have the same jfiace in No. 1969. The larger of these two was cut on stone, the smaller in stucco. The mask has been changed into the ball and hooks; the angular nose ornament into a single ball, easier to make and quite as significant to the Maya priest. But to us the older (Copan) figure is infinitely more significant. The curious rows of little balls which are often placed at No. 7 No. 1969. Fig. 51.—Synonomous hieroglyphs from. Copan and Palenque. 228 CENTRAL AMERICAN PICTURE-WRITING. the left-hand edge of the various cliiffres are also conventions for older forms. It is to be noted that these balls always occur on the left hand of the hieroglyphs, except in one case, the cliiffre 1975 in the Palenque cross tablet, on which the left-hand acolyte stands. The conclusion that the two series are both written on the same sys- tem, and that like cliiffres occurring at the two places are synonyms, will, I think, be sufficiently evident to any one who will himself exam- ine the following cases. It is the nature of the agreements which proves the thesis, and not the number of cases here cited. The reader will re- member that the Copan series comprises Plates I to XXIII, inclusive; the Palenque series, Plate XXIY and higher numbers. The sign of the group of Mexican gods who relate to hell, i. e., a circle with a central dot, and with four small segments cut out at four equally distant points of its circumference, is found in Xo. 4291, Plate XXII, and in many of the Palenque plates, as Plate LVI, Xos. 2090, 2073, 2045, 2021, etc. In both places this sign is worn by human figures just below the ear. The same sign occurs as an important part of Xo. 4271, Plate XXII, and Xo. 4118, Plate XIII (Copan), and Xo. 20G4, Plate LYI (Palenque), etc. Xo. 7, Plate Ya, and Xo. 1969, Plate LYI, I regard as absolutely iden- tical. These are both human figures. Xo. 12, Plate Ya, and Xo. 637, Plate LIII, are probably the same. These probably rej)resent or relate to the long-nosed divinity, Yacatetjctli, the Mexican god of commerce, etc., or rather to his Maya representative. The sign of Tlaloc, or rather the family of Tlalocs, the gods of rain, floods, and waters, is an eye (or sometimes a mouth), around which there is a double line drawn. I take Xo. 26, Plate Ya, of the Copan series, and Xos. 154 and 165, Plate XXIY, to be corresponding references to members of this family. Xo. 4, Plate Ya, and Xo. 155 also correspond. Xo. 4242, Plate XXII, is probably related to Xo. 53, Plate XXIY and its congeners. Xos. 14 and 34, Plate Ya, are clearly related to Xo. 900, Plate LIY, Xos. 127 and 176, Plate XXIY, Xo. 3010, Plate LYI, and many others. Plate IIP of Copan is evidently identically the same as the Xo. 75 of the Palenque Plate Xo. XXIY. The right half of Xo. 27, Plate Ya, is the same as the right half of Xos. 3020, 3040, and many others of Plate LYI. Xo. 17, Plate Ya, is related to Xo. 2051, Plate LYI, and many others like it. The major part of Xo. 4105, Plate XIII, is the same as Xo. 124, Plate XXIY, etc. It is not necessary to add a greater number of examples here. The card- catalogue which I have mentioned enables me to at once pick out all the cases of which the above are specimens, taken just as they fell under my eye in rapidly turning over the cards. They therefore represent the Fig. 52.—Yucatec Stone. 229 HOLDEN.] THE MAYA HIEROGLYPHS. average agreement, neither more nor less. Taken together they show that the same signs were used at Copan and at Palenque. As the same symbols used at both places occur in like positions in regard to the human face, etc., I conclude that not only were the same signs used at both places, but that these signs had the same meaning; i. e., were truly synonyms. In future I shall regard this as demonstrated. VIII. IIUITZILOPOCHTLI (MEXICAN GOD OF WAR), TEOYAOMI- QUI (MEXICAN GODDESS OF DEATH), MICLANTECUTLI (MEXICAN GOD OF HEEL), AND TLALOC (MEXICAN RAIN- GOD), CONSIDERED IN REEATION TO CENTRAL AMERI- CAN DIVINITIES. Iii the Gongres des Americanistes, session de Luxembourg, vol. ii, p. 283, is a report of a memoir of Dr. Leemans, entitled u Description de quelques antiquites americaines conservees dans le Musee royal neer- landais d’antiquites & Leide.” On page 299 we find— M. G.-H.-Band, de Arnbeim, a eu la bout6 de me confier quelques antiquites prove- nanfc des anciens habitants du Yucatan et de PAncUSrique Centrale, avec autorisation d’en faire prendre des fac-similes pour le Musde, ce qui me permet de les faire connaitre aux membres du Congrbs. Elies ont 6t6 trouvdes enfouies h une grande profondeur dans le sol, lors de la construction d’un canal, vers la riviere Gracioza, prfes de San Filippo, sur la frontiere du Honduras britannique et de la republique de Guatemala par M. S.-A.-van Braam, ingdnieur nbeiiandais au service de la Guatemala-Company. From the maps given in Stieler’s Hand-Atlas and in Bancroft’s Native Races of the Pacific States I find that these relics were found 308 miles from Uxmal, 207 miles from Palenque, 92 miles from Copan, and 655 miles from the city of Mexico, the distances being in a straight line from place to place. The one of these objects with which we are now concerned is figured in Plate (63) of the work quoted, and is reproduced here as Fig. 52. Dr. Leemans refers to a similarity between this figure and others in Stephens’ Travels in Central America, but gives no general comparison. I wish to direct attention to some of the points of this cut. The chiffre or symbol of the principal figure is, perhaps, represented in his belt, and is a St. Andrew’s cross, with a circle at each end of it. Inside the large circle is a smaller one. It may be said, in passing, that the cross prob- ably relates to the air and the circle to the sun. The main figure has two hands folded against his breast. Two other arms are extended, one in front, the other behind, which carry two birds. Each arm has a bracelet. This second pair of hands is not described by Dr. Leemans. The two birds are exact duplicates, except that the eye of one is shut, of the other open. Just above the bill of each bird is something which might be taken as a second bill (which probably is not, 230 CENTRAL AMERICAN PICTURE-WRITING. however), and on this and on the back of each bird are five spines or claws. The corresponding claws are curved and shaped alike in the two sets. The birds are fastened to the neck of the person represented by two ornaments, which are alike, and which seem to be the usual hiero- glyph of the crotalus jaw. These jaws are placed similarly with respect to each bird. In Kingsboiiougii’s Mexican Antiquities, vol. I, Plate X, we find the parrot as the sign of Tonatihu, the sun, and in Plate XXY with Naolin, the sun. On a level with the nose of the principal figure are two symbols, one in front and one behind, each inclosing a St. An- drew’s cross, and surmounted by what seems to be a flaming fire. It is probably the cliiffre of the wind, as the cross is of the rain. Below the rear one of these is a head with protruding tongue (the sign of Quet- zalcoatl); below the other a hieroglyph (perhaps a bearded face). Each of these is upborne by a hand. It is to be noticed, also, that these last arms have bracelets different from the pair on the breast. In passing, it may be noted that the head in rear is under a cross, and has on its cheek the symbol U. These are the symbols of the left- hand figure in the Palenque cross tablet. The head hanging from the rear of the belt has an open eye (like that of the principal figure), and above it is a crotalus mask, with open eye, and teeth, and forked fangs. The principal figure wears over his head a mask, with open mouth, and with tusks, and above this mask is the eagle’s head. This eagle is a sign of Tlaloc, at least in Yucatan. In Mexico the eagle was part of the insignia of Tetzcatlipoca, “the devil,” who overthrew the good Quetzalcoatl and reintroduced human sacrifice. The characteristics of the principal figure, 63, are then briefly as fol- lows : I. His cliiffre is an air-cross with the sun-circle. II. He has four hands. III. He bears two birds as a symbol. IY. The claws or spikes on the backs of these are significant. Y. The mask with tusks over the head. YI. The head worn at the belt. YII. The captive trodden under foot. YIII. The chain from the belt attached to a kind of ornament or sym- bol. IX. The twisted flames (?) or winds (?) on each side of the figure. X. His association with Quetzalcoatl or Cuculkan, as shown by the mouth with protruding tongue, and with Tlaloc or Tetzcatlipoca, as shown by the eagle’s head. We may note here for reference the signification of one of the hiero- glyphs in the right-hand half of Fig. 52, i. e., in that half which contains only writing. The topmost cliiffre is undoubtedly the name, or part of the name, of the principal figure represented in the other half. It is in pure picture-writing; that is, it expresses the sum of his attributes. HOLDEN.J THE MAYA HIEROGLYPHS. 231 It lias the erotalus mask, with nose ornament, which he wears over his face; then the cross, with the “ five feathers” of Mexico, and the sun symbol. These are in the middle of the chiffre. Below these the oval may be, and probably is, heaven, with the rain descending and iiroduc- ing Irom the surface of the earth (the long axis of the ellipse), the seed, of which three grains are depicted. We know by the occurrence of the hieroglyphs on the reverse side of the stone that this is not of Aztec sculpture. These symbols are of the same sort as those at Copan, Palenque, etc., and I shall show later that some of them occur in the Palenque tablets. Hence, we know this en- graving to be Yucatec and not Aztec in its origin. If it had been sculpt- ured on one side only, and these hieroglyphs omitted, I am satisfied that the facts which I shall point out in the next paragraphs would have led to the conclusion that this stone was Mexican in its origin. Fortu- nately the native artist had the time to sculpture the Yucatec hiero- glyphs, which are the proof of its true origin. It was not dropped by a traveling Aztec; it was made by a Yucatec. In passing, it may be said that the upper left-hand, hieroglyph of Plate XIII most probably repeats this name. I collect from the third volume of Bancroft’s Native Eaces, chapter viii, such descriptions of Huitzilopochtli as he was represented among the Mexicans as will be of use to us in our comparisons. Xo display of learning in giving the references to the original works is necessary here, since Mr. Bancroft has placed all these in order and culled them for a use like the present. It will suffice once for all to refer the critical reader to this volume, and to express the highest sense of obligation to Mr. Bancroft’s compilation, which renders a survey of the characteristic features of the American divinities easy. In Mexico, then, this god had, among other symbols, “five balls of feathers arranged in the form of a cross.” This was in reference to the mysterious conception of his mother through the powers of the air. The upper hieroglyph in Fig. 52, and one of the lower ones, contain this sign: “ In his right hand he had an azured staff cutte in fashion of a waving snake.” (See Plate LXI of Stephens.) “ Joining to the tem- ple of this idol there was a piece of less work, where there was another idol they called Tlaloc. These two idolls were alwayes together, for that they held them as companions and of equal power.” To his temple “ there were foure gates,” in allusion to the form of the cross. The temple was surrounded by rows of skulls (as at Copan) and the temple itself was upon a high pyramid. Solis says the war god sat “ on a throne supported by a blue globe. From this, supposed to rep- resent the heavens, projected four staves with serpents’ heads. (See Plate XXIY, Stephens.) “ The image bore on its head a bird of wrought plumes,” “its right hand rested upon a crooked serpent.” “Upon the left arm was a buckler bearing five white plums arranged in form of a cross.” Saiiagun describes his device as a dragon’s head, “ frightful in the extreme, and casting fire out of his mouth.” 232 CENTRAL AMERICAN PICTURE-WRITING. Herrara describes Huitzilopochtli and Tezoatlipoca together, and says they were “ beset with pieces of gold wrought like birds, beasts, and fislies.v “For collars, they had ten hearts of men,” “and in their necks Death painted.” Torquemada derives the name of the war god in two ways. Ac- cording to some it is composed of two words, one signifying “a humming bird” and the other “ a sorcerer that spits fire.” Others say that the last word means “the left hand,” so that the whole name would mean “the shining feathered left hand.” “This god it was that led out the Mexicans from their own land and brought them into Anahuac.” Be- sides his regular statue, set up in Mexico, “ there was another renewed every year, made of different kinds of grains and seeds, moistened with the blood of children.” This was in allusion to the nature side of the god, as fully explained by Muller (Americanisclie TJrreligioneri). No description will give a better idea of the general features of this god than the following cuts from Bancroft’s Native Races, which are copied from Leon y Gama, Las Dos Piedras, etc. Figs. 53 and 54 are Fig. 53.—Huitzilopochtli (front). Fig. 54.—Huitzilopochtli (side). the war god himself; Fig. 55 is the hack of the former statue on a larger scale; Fig. 56 is the god of hell, and was engraved on the bottom of the block. These three were a trinity well nigh inseparable. It has been doubted whether they were not differ- ent attributes of the same personage. In the nat- ural course of things the primitive idea would become differentiated into its parts, and in process of time the most important of the parts would each receive a separate pictorial representation. By referring back a few pages the reader will find summarized the principal characteristics of the Cen- tral American figure represented in Fig. 52. He will also have noticed the remarkable agreement between the attributes of this figure, and Fig. 56.—Miclantecutli. Fig. 55.—Huitzilopoohtli (back). Fig. 57.—Adoratorio. 233 HOLDEN. J THE MAYA HIEROGLYPHS. those contained in the cuts or in the descriptions of the Mexican gods. Thus— I. The symbol of both was the cross. II. Fig. 52 and Fig. 55 each have four hands.* III. Both have birds as symbols. It is difficult to regard the bird of Fig. 52 as a humming bird, as it more resembles the parrot, which, as is well known, was a symbol of some of the Central American gods. Its occurrence here in connection with the four arms fixes it, however, as the bird symbol of Huitzilo- pochtli. In the Ms. Troano, plate xxxi (lower right-hand figure), we find this same personage with his two parrots, along with Tlaloo, the god of rain. IV. The claws of the Mexican statue may be symbolized by the spikes on the back of the birds in Fig. 52, but these latter appear to me to relate rather to the fangs and teeth of the various crotalus heads of the statues. V. The mask, with tusks, of Fig. 52, is the same as that at the top of Fig. 55, where we see that they represent the teeth of a serpent, and not the tusks of an animal. This is shown by the forked tongue beneath. The three groups of four dots each on Huitzilopochtli’s statue are references to his relationship with Tlaloc. With these main and striking duplications, and with other minor and corroborative resemblances, which the reader can see for himself, there is no doubt but that the two figures, Mexican and Yucatec, relate to the same personage. The Yucatec figure combines several of the attributes of the various members of the Mexican trinity named above, but we should not be surprised at this, for, as has been said, some writers con- sider that this trinity was one only of attributes and not of persons. What has been given above is sufficient to show that the personage represented in Fig. 52 is the Yucatec equivalent of Huitzilopochtli, and has relations to his trinity named at the head of this section, and also to the family of Tlaloc. I am not aware that the relationship of the Yucatec and Aztec gods has been so directly shown, on evidence almost purely pictorial, and therefore free from a certain kind of bias. If the conclusions above stated are true, there will be many corrobora- tions of them, and the most prominent of these I proceed to give, as it involves the explanation of one of the most important tablets of Palen- que, parts of which are shown *in Plates XXIV, LX, LXI, and LXII, vol. ii, of Stephens. Plate LXII, Fig. 57, represents the u Adoratorio or Alta Casa, Xo. 3” of Palenque. This is nothing else than the temple of the god Huit- zilopochtli and of his equal, Tlaloc. The god of war is shown on a larger scale in Plate LXI, Fig. 58, while Tlaloc is given in Plate LX, Fig. 59, and the tablet inside the temple in Plate XXIV, Fig. 60. The * From Kingsborough, vol. i, plate 48, it appears that Tlacli Tonatio may have had four bauds. His name meant (?) Let there he light. 234 CENTRAL AMERICAN PICTURE-WRITING. resemblances of Plate XXIV and of the Palenque cross tablet and their meanings will be considered further on. Returning to Plate LXII, the symbols of the roof and cornice refer to these two divinities. The faces at the ends of the cornice, with the double lines for eye and mouth, are unmistakable Tlaloc signs. The association of the two gods in one temple, as at Mexico, is a strong cor- roboration. Let us now take Plate LXI, Fig. 58, which represents Huitzilo- pochtli, or rather, the Yucatec equivalent of this Aztec god. I shall refer to him by the Aztec appelation, but I shall in future write it in italics; and in general the Yucatec equivalents of Aztec personages in italics, and the Aztec names in small capitals. Compare Fig. 52 and the Plate LXI (Fig. 58). As the two plates are before the reader, I need only point out the main resemblances, and, what is more important, the differences. The sandals, the belt, its front pendant, the bracelets, the neck orna- ment, the helmet, should be examined. The four hands of Fig. 52 are not in LXI, nor the parrots; but if we refer to Kingsborough, Yol. II, Plates 6 and 7 of the Laud manuscript, we shall find figures of Huit- zilopochtli with a parrot, and of Tlaloc with the stork with a fish in its mouth, as in the head-dress here. The prostrate figure of Fig. 52 is here led by a chain. At Labphak (Bancroft, Yol. iv., p. 251), he is held aloft in the air, and he is on what may be a sacrificial yoke. The Tlaloc eagle is in the head of the staff carried in the hand. This eagle is found in the second line from the bottom of Fig. 52, we may remark in pass- ing. Notice also the crescent moon in the ornament back of the shoul- ders of the personage of Fig. 58. The twisted cords which form the bottom of this ornament are in the hieroglyph No. 37, Plate XXIY (Fig. 60). Turning now to Plate LX (Fig. 59). This I take to be the sorcerer Tlaloc. He is blowing the wind from his mouth; he has the eagle in his head-dress, the jaw with grinders, the peculiar eye, the four Tlaloc dots over his ear and on it, the snake between his legs, curved in the form of a yoke (this is known to be a ser- pent by the conventional crotalus signs of jaw and rattles on it in nine places), the four Tlaloc dots again in his head-dress, etc. He has a leopard skin on his back (the tiger was the earth in Mexico) and his naked feet have peculiar anklets which*should be noticed. Although I am deferring the examination of the hieroglyphs to a later section, the chiffre 3201 should be noticed. It is the Tlaloc eye again, and 3203 is the cliiffre of the Mexican gods of hell. In passing I may just refer the reader to p. 164, Yol. ii, of Stephens’ book on Yucatan, where a figure occurring at Labphax is given. This I take to be the same as Huitzilopochtli of Plate LXI. Also in the MS. Troano, published by Brasseur de 'Bourbourg, a figure in Plate XXY and in other plates sits on a hieroglyph like 3201, and is Fig. 58.—Maya War God. Fig, 59,—Maya Rain God. Fig. 60.—Tablet at Palenquo. HOLDEN.] THE MAYA HIEROGLYPHS. 235 Tlaloc. This is known by the head-dress, the teeth, the air-trumpet, the serpent symbol, etc. In Plates XXVIII, XXXI, and XXXIII of the same work Huitzilopochtli and Tlaloc are represented together, in various adventures. In Plate LX (Fig. 59) notice also the chffre on the tassels before and behind the main personage. Xow turn to the Plate XXIY (Fig. 60), which is the main object in the u Adoratorio” (Fig. 57), where the human figures serve as flankers. First examine the caryatides who support the central structure. These are Tlalocs. Each has an eagle over his face, is clothed in leopard skin, has the characteristic eye and teeth, and the wristlets of Plate LX (Fig. 59). A vertical line through the center of Plate XXIV (Fig. 60) would sepa- rate the figures and ornaments into two groups. These groups are very similar, but never identical, and this holds good down to the minutest particulars and is not the result of accident. One side (the right-hand) belongs to Tlaloc, the other to Huitzilopochtli. The right-hand priest (let us call him, simply for a name and not to commit ourselves to a theory) has the sandals of Plate LXI; the left- hand priest the anklets of Plate LX. The beast on which the first stands and the man who supports the other are both marked with the tassel symbol of Plate LX. There is a cer- tain rude resemblance between the supplementary head of this beast and the pendant in front of the belt of Fig. 52. Four of these beasts sup- ply rain to the earth with Tlaloc in Plate XXVI of the MS. Troano. The infant offered by the right-hand priest has the two curls on his forehead which was a necessary mark of the victims for Tlaloc’s sacri- fices. The center of the whole plate is a horrid mask with an open mouth. Behind this are two staves with different ornaments crossed in the form of the air-cross. On either hand of this the ornaments are dif- ferent though similar. A curious resemblance may be traced between the positions, etc., of these two staves and those of the figure on p. 563, vol. iv, of Ban- croft’s Native Races, which is a Mexican stone. Again, this latter figure has at its upper right-hand corner a crouching animal (?) very similar to the gateway ornament given in the same volume, p. 321. This last is at Palenque. I quote these two examples in passing simply to reinforce the idea of similarity between the sacred sculptures of Yucatan and Mexico. I take it that the examination of which I have sketched the details will have left no doubt but that the personage of Fig. 52 is truly Huitzilo- pochtli, the Yucatec representative of Huitzilopochtli ; that Plate LXI (Fig. 58) is the same personage; that Plate LX (Fig. 59) represents Tlaloc ; and that Plate XXIV (Fig. 60) is a tablet relating to the serv- ice of these two gods. I have previously shown that the Palenque hieroglyphs are read in 236 CENTRAL AMERICAN PICTURE-WRITING. order from left to right. We should naturally expect, then, that the sign for Tlaloc or for H.uitzilopochtli would occupy the upper left-hand corner of Plate XXIY. In fact it does, and I was led to this discovery in the way I have indicated. Xo. 37 is the Palenque manner of writing the top sign of Fig. 52. I shall call the signs of Fig. 52 a, 5, c, etc., in order downwards. The crouching face in a occupies the lower central part of Xo. 37. Xotice also that this face occurs below the small cross in the detached ornament to the left of the central mask of Fig. CO. The crescent moon of Plate LXI (Fig. 58) is on its cheek; back of this is the sun-sign; the cross of a is just above its eye; the three signs for the celestial concave are at the top of 37, crossed with rain bands; the three seeds (?) are be- low these. The feathers are in the lower right-hand two-thirds. This is the sign or part of the sign for Huitzilopoclitli. If a Maya Indian had seen either of these signs a few centuries ago, he would have had the successive ideas—a war-god, with a feather-symbol, related to sun and moon, to fertilizing rain and influences, to clouds and seed; that is Huitzilopochtli, the companion of Tlaloc. Or if he had seen the upper left-hand symbol of the Palenque cross tablet (1800), he would have had related ideas, and so on. What I have previously said about the faithfulness with which the Yu- catec artist adhered to his prototypes in signs is perfectly true, although apparently partly contradicted by the identification I have just made. When a given attribute of a god (or other personage) was to be depicted, the cliiffres expressing this were marvellously alike. Witness the chiffres Xos. 2090, 2073, 2021,2045, 3085, 3073, 3070, 3032 of the Palenque cross tablet. But directly nfterwards some other attribute is to be brought out,and the chiffre changes; thus the hieroglyph 1009 of Plate LIV, or 265, Plate LII, has the same protruding tongue as 2021, etc., and is the same personage, but the style is quite changed. In Fig. 52, Huitzilo- pochtli is the war-god, in Plate XXIV he is the rain-god’s companion; and while every attribute is accounted for, prominence is given to the special ones worshipped or celebrated. Scores of instances of this have arisen in the course of my examination. Again, we must remember that this was no source of ambiguity to the Yucatecs, however much it may be to us. Each one of them, and spe- cially each officiating priest, was entirely familiar with every attribute of every god of the Yucatec pantheon. The sign of the attribute brought the idea of the power of the god in that special direction; the full idea of his divinity was the integral of all these special ideas. The limits were heaven and earth. This, then, is the first step. I consider that it is securely based, and that we may safely say that in proper names, at least, a kind of picture writing was used which was not phonetic. From this point we may go on. I must again remark that great famil- iarity with the literature of the Aztecs and Yucatecs is needed—a famil- HOLDEN.] THE MAYA HIEROGLYPHS. 237 iarity to which I personally cannot pertend—and that it is clear that the method to reach its full success must be applied by a true scholar in this special field. IX. TLALOC, OR HIS MAYA REPRESENTATIVE. Although there is no personage of all the Maya pantheon more easy to recognize in the form of a statue than Tlaloc, there is great difficulty in being certain of all the hieroglyphs which relate to him. There is every reason to believe that in Yucatan, as in Mexico, there was a family of rain-gods, Tlalocs, and the distinguishing signs of the several mem- bers are almost impossible of separation, so long as we know so little of the special functions of each member of this family. In Yucatan, as in Mexico, Tlalods main sign was a double line about the eye or mouth, or about both ; and further, some of the Tlalocs, at least, were bearded.* Cukulcan was also bearded, but we have separated out in the next section the chiffres, or certainly most of them, that relate to him. Those that are left remain to be distributed among the family of rain-gods; and this, as I have said, can only be done imperfectly, on account of our slight knowledge of the character of these gods. If we examine the plates given by Stephens, we shall find many pictorial allusions to Tlaloc. These are often used as mere ornaments or embellishments, as in borders, etc., and probably served only to notify, in a general way, the fact of the relationship of the personage repre- sented, to this family, and probably not to convey any specific meaning. Thus, in Plate XXXV of Stephens7 work the upper left-hand orna- 1 ment of the border is a head of Tlaloc with double lines about eye and mouth, and this ornament is repeated in a different form at the lower right-hand corner of the border just back of the right hand of the sit- ting figure, and also in the base of the border below the feet of the prin- cipal figure. Plate XLYIII (of Stephens7) is probably Ghalchihuitlicue (that is, the Yucatec equivalent of that goddess), who was the sister of Tlaloc. His sign occurs in the upper left-hand corner of the border, and in Plate XLIX the same sign occurs in a corresponding position. Plate iXIV (our Fig. 60) is full of Tlaloc signs. The bottom of the tablet has a hieroglyph, 93 (Huitzilopochtli), at one end and 185 (Tlaloc) at the other. The leopard skin, eagle, and the crouching tiger (?) under the feet of the priest of Tlaloc (the right-hand figure) are all given. The infant (?) offered by this priest has two locks of curled hair at its fore- head, as was prescribed for children offered to this god. * See Kingsborough, vol. ii, Plate I, of the Laud MS. 238 CENTRAL AMERICAN PICTURE-WRITING In Plate LVI (our Fig. 48) the mask at the foot of the cross is a human mask, and not a serpent mask, as has been ingeniously proved by Dr. Harrison Allen in his paper so often quoted. It is the mask of Tlaloc, as shown by the teeth and corroborated (not proved) by the way in which the eye is expressed. The curved hook within the eyeball here, as in 185, stands for the air—the wind—of which Tlaloc was also god. The Mexi- cans had a similar sign for breath, message. The chiffre 1975, on which HuitzilopochtWs priest is standing, I believe to be the synonym of 185 in Plate XXIV. Just in front of Tlaloc’s priest is a sacrificial yoke (?), at the top of which is a face, with the eye of the Tlalocs, and various decorations. This face is to be found also at the lower left-hand corner of Plate XLI (of Stephens’), and also (?) in the same position in Plate XLII (of Stephens’). These will serve as sub- jects for further study. Xotice in Plate LYI (our Fig. 48) how the ornaments in corresponding positions on either side of the central line are similar, yet never the same. A careful study of these pairs will show how the two gods celebrated, differed. A large part, at least, of the attributes of each god is re- corded in this way by antithesis. I have not made enough progress in this direction to make the very few conclusions of which I am certain worth recording. The general fact of such an antithesis is obvious when once it is pointed out, and it is in just such paths as this that ad- vances must be looked for. I have just mentioned, in this rapid survey of the plates of vol. ii of Stephens’ work, the principal pictorial signs relating to Tlaloc. There are a number almost equally well marked in vol. i, in Plates VII, IX, X, XIII, and XV, but they need not be described. Those who are especi- ally interested can find them for themselves. The following brief account and plate of a Tlaloc inscription at Kabah will be useful for future use, and is the more interesting as it is compara- tively unknown. INSCRIPTION AT KABAH {Yucatan). This hitherto unpublished inscription on a rock at Kabah is given in Archives paleographiques, vol. i, part ii, Plate 20. It deserves attention on account of its resemblances, but still more on account of its differ- ences, with certain other Yucatee glyphs. We may first compare it with the Plate LX of Stephens (our Fig. 59). The head-dress in Plate 20 is quite simple, and presents no resemblance to the elaborate gear of Plate LX, in which the ornament of a leaf (?), or more probably feather, cross-hatched at the end and divided sym- metrically by a stem (?) or quill about which four dots are placed, seems characteristic. Possibly, and only possibly, the square in the rear of the head of Plate 20, which has two cross-hatchings, may refer to the elaborate cross-hatcliings in Plate LX. The four dots are found twice, once in HOLDEN.] THE MAYA HIEROGLYPHS. 239 front and once in rear of the figure. The heads of the two figures have only one resemblance, but this is a very important one. The tusks be- long to Huitzilopochtli and to his trinity, and specially to Tlaloc, his companion. Both Plate 20 and LX have the serpent wand or yoke clearly ex- pressed. In LX the serpent is decorated with crotalus heads; in 20 by images of the sun (?), as in the Ferjavary MS. (Kingsborough). The front apron or ornament of Plate 20 is of snake skin, ornamented with sun-symbols. Comparing Plate 20 with Fig. 52 (ante), we find quite other resemblances. The head-dress of 20 is the same as the pro- jecting arm of the head-dress of Fig. 52; and the tusks are found in the helmet or mask of Fig. 52. These and other resemblances show the Kabah inscription to be a Tlaloc. It is interesting specially on account of its hieroglyphs, which I hope to examine subsequently. The style of this writing ap- pears to be late, and may serve as a connecting link between the stones and the manuscripts, and it is noteworthy that even the style of the drawing itself seems to be in the manner of the Mexican MS. of Laud, rather than in that of the Palenque stone tablets. From the card catalogue I select the following chiffres as appertaining to the family of the Tlalocs. As I have said, these must for the present remain in a group, unseparated. Future studies will be necessary to discriminate between the special signs which relate to special members of the family. The chiffres are Xos. 3200; 1864; 1403; 811; 11071?; 1943?; 4114??; 6?; 1893 (bearded faces, or faces with teeth very promi- nent); 166?; 4??; 807?; 62?; 155?; 26; 154?; 165?; 164?; 805; 4109; 1915?; 675??; 635?? (distinguished by the characteristic eye of the Tlalocs). Here, again, the writing is ideographic, and not phonetic. X. CUKULCAN OR QUETZALCOATL. The character 2021 occurs many times in Plate LYI (Fig. 48), and oc- casionally elsewhere. The personage represented is distinguished by having a protruding tongue, and was therefore at once suspected to be Quetzalcoatl. (See Bancroft’s Native Races, vol. iii, p. 280.) The protruding tongue is probably a reference to his introduction of the sac- rificial acts performed by wounding that member. The rest of the sign I suppose to be the rebus of his name, u Snake- plumage”; the part cross-hatched being 11 snake,” the feather-like orna- ment at the upper left-hand corner being “ plumage.” It is necessary, however, to prove this before accepting the theory. To do this I had recourse to Plates I and IV (Figs. 40, 50), my dictionary of synonyms. 240 CENTRAL AMERICAN PICTURE-WRITING. This Gross hatching occurs in Plate I. In the six tassels below the waist, where the cross-hatching might indicate the serpent skin, notice the ends of the tassels ; these are in a scroll-like form, and as if rolled or coiled up. In Plate IY they are the same, naturally. So far there is but little light. In Plate IV, just above each wrist, is a sign composed of ellipse and bars5 a little above each of these signs, among coils which may be ser- pent coils, and on the horizontal line through the top of the necklace pendaut, are two surfaces cross-hatched all over. What do these mean? Referring to Plate I, we find, in exactly the same relative situation, the forked tongue and the rattles of the crotalus. These are, then, synonyms, and the guess is confirmed.' The cross-hatching means serpent-skin. Is this always so ? We must examine other plates to decide. The same ornament is found in Plates IX, XIY, XVI, XVIII, XIX, XX, XXI, XXXV (of Stephens’), but its situation does not allow us to gain any additional light. In Plate XII (Stephens’) none of the ornaments below the belt will help us. At the level of the mouth are four patches of it. Take the upper right-hand one of these. Immediately to its right is a serpent’s head; below the curve and above the frog’s (?) head are the rattles. Here is another confirmation. In Plate XVIII I refer the cross-hatch- ing to the jaw of the crocodile. In Plate XXIII have numbered the chiffres as follows: 4201 4202 4203 4204. 4211 4212 4213 4214. * * * * * * * * * * * * 4311 4312 4313 4314. 4204 has the cross-hatching at its top, and to its left in 4203 is the ser. pent’s head. The same is true in 4233-4. In 4264 we have the same symbol that we are trying to interpret; it is in its perfect form here and in Xo. 1865 of the Palenque series. In the caryatides of Plate XXIV (Fig. 60") the cross-hatching is included in the spots of the leopard’s skin; in the ornaments at the base, in and near the masks which they are sup- porting, it is again serpent skin. Take the lower mask; its jaws, forked- tongue, and teeth j)rove it to be a serpent-mask, as well as the ornament just above it. In Plate LX (Fig. 59) is to be noticed that the leopard spots are not cross-hatched, but that this ornament is given at the lower end of the leopard robe, which ends moreover in a crotalus tongue marked with the sign of the jaw (near the top of this ornament) and of the rattles (near the bottom). This again confirms the theory of the rebus meaning of the cross-hatching. In Plate XXIV (Fig. 60) the cross- hatching on the leopard spots probably is meant to add the serpent attribute to the leopard symbol, and not simply to denote the latter. Thus an examination of the whole of the material available, shows that the preceding half of the hieroglyph 2021 and its congeners is nothing IIOLDEN.j THE MAYA HIEROGLYPHS. 241 but the rebus for Quetzalcoatl, or rather for Cukulcan, the Maya name for this god. Brasseur de Bourbourg, as quoted in Ban- croft’s Native Races, vol. ii, p. 699, foot note, says Cukulcan, comes from kuk or kukul, a bird, which appears to be the same as the quetzal, and from can, serpent; so that Cukulcan in Maya is the same as Quetzalcoatl in Aztec. It is to be noticed how checks on the ac- curacy of any deciphering of hieroglyphs occur at every point, if we will only use them. The Maya equivalents of Huitzilopochtli and Tlaloc are un- doubtedly buried in the chiffres already deciphered, but we have no means of getting their names in Maya from the rebus of the chiffres. In the cases of these two gods we got the chiffre, and the rebus is still to seek. In the case of Quetzalcoatl or Cukulcan, the rebus was the means of getting the name; and if the names of this divinity had not been equivalent in the two tongues, our results would have led us to the (almost absurd) conclusion that a god of certain attributes was called by his Aztec name in the Maya nations. Thus every correct conclusion confirms every former one and is a basis for subsequent progress. The results of this analysis are that the Maya god Cukulcan is named in each one of the following chiffres, viz: Nos. 1009, 265, 2090, 2073, 2021, 3085, 2045, 3073, 3070, 3032, 1865, 265, 268?, 4291? 73?? I give the numbers in the order in which they are arranged in the card-catalogue. There is, of course, a reason for this order. Bancroft, vol. iii, p. 268, says of Quetzalcoatl that u his symbols were the bird, the serpent, the cross, and the flint, representing the clouds, the lightning, the four winds, and the thunderbolt.” We shall find all of his titles except one, the bird, in what follows. We must notice here that in the chiffre 2021 and its congeners the bird appears directly over the head of Cukulcan. It is plainly shown in the heliotype which accompanies Professor Bau’s work on the Palenque cross, though not so well in our Fig. 48. In what has gone before, we have seen that the characters 2021, 2045, 2073, 3073, 3085, 265, etc., present the portrait and the rebus of Cukul- can. It will not be forgotten that in the examination of the question as to the order in which the stone inscriptions were read we found a number of pairs in Plate LVI, Fig. 48; the characters 2021, etc., being one member of each. The other members of the pairs in the Plate LYI were 2020, 2044, 2072, 3072, 3084, etc. 264-265 is another example of the same pair elsewhere. I hoped to find that the name Cukulcan, or 2021, was associated in these pairs with some adjective or verb, and therefore examined the other members of the pair. In a case like this the card-catalogue is of great assistance; for ex- ample, I wish to examine here the chiffres Nos. 2020, 2044, 2072, 3072, 3084, etc. In the catalogue their cards occur in the same compart- ment, arranged so that two cards that are exactly alike are contiguous. 242 CENTRAL AMERICAN PICTURE-WRITING. We can often know that two chiffres are alike wlien one is in a far bet- ter state of preservation than the other. Hence we may select for study that one in which the lines and figures are best preserved; or from sev- eral characters known to be alike, and of which no one is entirely perfect, we may construct with accuracy the type upon which they were founded. In this case the hieroglyph 2020 is well preserved (see the right-hand side of Plate LYI, Fig. 48, the upper left-hand glyph). It consists of a human hand, with the symbol of the sun in it; above this is a sign similar to that of the Maya day Ymix ; above this again, in miniature, is the rebus “snake ifiuuiage” or Culculcan; and to the left of the hiero- glyph are some curved lines not yet understood. Xo. 2003 of the same plate is also well preserved. It has the hand as in 2020, the rebus also, and the sign for Ymix is slightly different, being modified with a sign like the top of a cross, the symbol of the four winds. The symbol Ymix may be seen, by a reference to Plate XXYII (lower half) of the MS. Troano, to relate to the rain. The figure of that plate is pouring rain upon the earth from the orifices represented by Ymix. The cross of the four winds is still more plain in Xos. 2072, 3084, and 3072. The part of this symbol 2020 and its synonyms which consists of curved lines occupying the left hand one-third of the whole chiffre occurs only in this set of characters, and thus I cannot say certainly what this particu- lar part of the hieroglyph means; but if the reader will glance back over the last one hundred lines he will find that these chiffres contain the rebus Cukulcan, the sign of a human hand, of the sun, of the rain, and of the four winds. In Bancroft’s Native Races, vol. iii, chapter vii, we find that the titles of Quetzalcoatl (Cukulcan) were the air, the rattlesnake, the rumbler (in allusion to thunder), the strong hand, the lord of the four winds. The bird symbol exists in 2021, etc. Xow in 2020 and its congeners we have found every one-of these titles, save only that relating to the thunder. And we have found a meaning for every part of the hieroglyph 2020 save only one, viz, the left-hand one-third, consisting of concentric half ellipses or circles. It may be said to be quite proba- ble that the unexplained part of the sign (2020) corresponds to the un- used title, “ the rumbler.” But it is not rigorously proved, although very probable. The thunder would be well represented by repeating the sign for sky or heaven. This much seems to me certain. The sign is but another summing up of the attributes and titles of Cukulcan. 2021 gave his portrait, his bird symbol, made allusion to his institution of the sacrifice of wounding the tongue, and spelled out his name in rebus characters. 2020 repeats his name as a rebus and adds the titles of lord of the four winds, of the sun, of rain, of the strong hand, etc. It is his biography, as it were. In this connection, a passing reference to the characters 1810, etc., 1820, etc., 1830, etc., 1840, etc., 1850, etc., of the left-hand side of Plate LVI should be made. Among these, all the titles named above are to be found. These are suitable subjects for future study. HOLDEN.] THE MAYA HIEROGLYPHS. 243 We now see why the pair 2020,2021 occurs so many times in Plate LYI, and again as 2G4, 265, etc. The right-hand half of this tablet has much to say of Oukulcan, and whenever his name is mentioned a brief list of his titles accompanies it. Although it is disappointing to find loth mem- bers of this well-marked pair to be proper names, yet it is gratifying to see that the theory of pairs, on which the proof of the order in which the tablets are to be read must rest, has received such unexpected con- firmation. To conclude the search for the hieroglyphs of Cukulcan’s name, it will be necessary to collect all those faces with “ round beards ” (see Bancroft’s Native Races, vol. iii, p. 250). Tlaloc was also bearded, but all the historians refer to Quetzalcoatl as above cited. I refer hieroglyphs Nos. 658, 651650 ?, and 249 ? to this category. Perhaps also the sign No. 153 is the sign of Quetzalcoatl, as some- thing very similar to it is given as his sign in the Codex Telleriano Remen- sis, Kingsborough, vol. i, Plates I, II, and V (Plate I the best), where he wears it at his waist. In Plate LXIII of Stephens (vol. ii) is a small figure of Oukulcan which he calls “Bas Belief on Tablet.” Waldeck gives a much larger drawing (incorrect, however, in many details), in which the figure, the 11 Beau Belief,” is seen to wear bracelets high up on the arm. This was a distinguishing sign of Quetzalcoatl (see Bancroft’s Native Races, vol. iii, pp. 249 and 250), and this figure probably is a representation of the Maya divinity. He is on a stool with tigers for supports. The tiger belongs to the attributes which he had in common with Tlaloc, and we see again the intimate connection of these divinities—a connection often pointed out by Brasseur de Bourbourg. This is the third proper name which has been deciphered. All of them have been pure picture-writing, except in so far as their rebus character may make them in a sense phonetic. XI. COMPARISON OP THE SIGNS OP THE MAYA MONTHS (LANDA) WITH THE TABLETS. We have a set of signs for Maya months and days handed down to us by Land A along with his phonetic alphabet. A priori these are more likely to represent the primitive forms as carved in stone than are the alphabetic hieroglyphs, which may well have been invented by the Span- iards to assist the natives to memorize religious formulae.* * Since this was written I have seen a paper by Dr. Y alentini, ‘ ‘ The Land a alpha- bet a Spanish fabrication” (read before the American Antiquarian Society, April 28, 1880), and the conclusions of that paper seem to me to be undoubtedly correct. They are the same as those just given, but while my own were reached by a study of the stones and in the course of a general examination, Dr. Valentini has addressed himself successfully to the solution of a special problem. 244 CENTRAL AMERICAN PICTURE-WRITING. Brasseur de BouRBOURa has analyzed the signs for the day and month in his publication on the MS. Troano, and the strongest argu- ments which can be given for their phonetic origin are given by him. 1 have made a set of MS. copies of these signs and included them in my card-catalogue, and have carefully compared them with the tablets XXIY and LYI. My results are as follows: Plate XXIY (our Fig. 60). No. 42 is the Maya month Pop, beginning July 16. No. 54 is ZipVi, beginning August 25. No. 47 is Tzoz ? ?, beginning September 14. No. 57 is Tzec?, beginning October 4. No. 44-45 is Mol ?, beginning December 3. No. 39 is Yax, Zac, or Cell, beginning January 12, February 1, Feb- ruary 21, respectively. Plate LYI (our Fig. 48). No. 1804 is Ubf!ft No. 1807 is MoU No. 1901 is ZvpWW No. 1855 is Yax, Zac, or Cell. No. 1816 is Tzozn No. 1844 is Mac? No. 1814 is Tzec% Tlie only sign about which there is little or no doubt is No. 42, which seems pretty certainly to be the sign of the Maya month Pop, which began July 16. No. 39, just above it, seems also to be one of the months Yax, Zac, or Cell, which began on January 12, February 1, and February 21, respect- ively. Which one of these it corresponds to must be settled by other means than a direct comparison. The signs given by Landa for these three months all contain the same radical as No. 39, but it is impossible to decide with entire certainty to which it corresponds. It, however, most nearly resembles the sign for Zac (February 1); and it is note- worthy that it was precisely in this month that the greatest feast of Tlaloc took place,* and its presence in this tablet, which relates to Tlaloc, is especially interesting. In connection with the counting of time, a reference to the bottom part of the chiffre 3000 of the Palenque cross tablet should be made. This is a knot tied up in a string or scarf; and we know this to have been the method of expressing the expiration and completion of a cycle of years. It occurs just the symbol 3010, the chiffre for a metal. An examination of the original stone in the National Museum, Wash- ington, which is now in progress, has already convinced me that the methods which I have described in the preceding pages promise other interesting confirmations of the results I have reached. For the time, * See Brasseur de Bourbourg, Histoire du Mexique, yol. i, p. 328. 245 HOLDEN.] THE MAYA HIEROGLYPHS. I must leave the matter in its present state. I think I am justified in my confidence that suitable methods of procedure have been laid down, and that certain important results have already been reached. I do not believe that the conclusions stated will be changed, but I am confident that a rich reward will be found by any competent person who will continue the study of these stones. The proper names now known will serve as points of departure, and it is probable that some research will give us the signs for verbs or adjectives connected with them. It is an immense step to have rid ourselves of the phonetic or alpha- betic idea, and to have found the manner in which the Maya mind represented attributes and ideas. Their method was that of all nations at the origin of written language,* that is, pure picture-writing. At Copan this is found in its earliest state; at Palenque it was already highly conventionalized. The step from the Palenque character to that used in the Kabah inscription is apparently not greater than the step from the latter to the various manuscripts. An important research would be the application of the methods so ably applied by Dr. Allen to tracing the evolution of the latter characters from their earlier forms. In this way it will be possible to extend our present knowledge mate- rially.