SMITHSONIAN CONTRIBUTIONS TO KNOWLEDGE. ---—-------------159 -__________.______ RESEARCHES AMTOMY AND PHYSIOLOGY OF RESPIRATION CHELONI A. BY S. WEIR MITCHELL, M.D. AND GEORGE R. MOREHOUSE, M.D. [accki'tkd fok publication, march, 1863.] COMMISSION TO WHICH THIS PAPER HAS BEEN REFERRED. Prof. Jeffries Wym.vn, Prof. Joseph Leidy. Joseph Henry, Secretary S. I. COLL1N8, PRINTER, PUILADBLPHIA. PREFACE. With certain slight exceptions, which we haAre pointed out in the text, the folloAving essay is in the strictest possible sense the joint production of its tAvo authors, who are equally responsible for all of its statements. The woodcuts owe much of their accuracy to the skill of the engraver, Mr. Wilhelm, to whose experience as an anatomical draughtsman the authors are under obligation. They entertain the wish that the novel vieAvs of the present paper may induce comparative anatomists and physiologists to examine afresh the respiratory mechan- ism of other reptiles, and also of birds—a labor in which they indulge the hope of sharing. S. WEIR MITCHELL, No. 1226 Walnut Street, Philadelphia. GEO. R. MOREHOUSE, No. 227 South Ninth Street, Philadelphia. TABLE OF CONTENTS. Preface List of avood-cuts CHAPTER I. ANATOMY* OF THE RESPIRATORY APPARATUS OF CHELONIA Introductory remarks Anatomical history Dissertation on the respiration of the tortoise, by Roliert Townsou Bojanus, "Anatomia Testudinis Europaeae" Cuvier on respiration in turtles His criticism on Townson Dumeril and Bibron, "mechanism of respiration in chelonians' Anatomy of respiratory organs in chelydra serpentina . Hyoid bone, description of Laryngeal cartilages Trachea Relations of lungs Absence of striated muscular fibre in lungs Chelydra serpentina, the typical turtle of our researches General plan of respiratory apparatus . Respiratory myology ..... Peculiarities of, in different genera Chelydra serpentina, description of Respiratory act . Expiratory muscle ..... Dissection to expose the mnscle of expiration Appearance of expiratory muscle Origin and insertions of the anterior and posterior bellies of Tendon of ..... Relation of this muscle to the lungs Inspiratory muscle ..... Position of Dissection to expose it Description of . Its muscular fibres Origin and insertion of Central tendon . Homologue of Poupart's ligament Relation of inspiratory muscle to expiratory muscle Glottic muscles .... Crico-arytenoid .... M T A I! I. K O F CO N" T V \ TS. f'rico-livnid Origin and insertions of It■ spirutory muscles in Chelonioid: In Chclonia mydas Itc-spiratory muscles in Einrdoidiu In Nccteniydoidae. Ptychemys rugosa . Ptychemys mobiliensis Graptcinys geogra])hica Malacoclemmys palustris Chrysemvs picta In ClcmmvdoidaB. Xanemys guttata Call mys Miihlenbergii (Jlyptemys insculpta In Cistudinina. Cistudo Virginea . Table of origin of respiratory muscles of Emydoida' with dimen Peculiarities of origin of the expiratory muscles in connection w A^'ural apparatus of respiration in Chelonians Medulla oblongata Par vagum Origin and course of Superior laryngeal ncr\c Communicating branch of Branch to crico-arytenoid muscle Branch to crico-hyoid muscle Dissection to expose chiasm of superior laryngeal nerve Sensitive fibres of superior laryngeal nerve Inferior laryngeal nerve Origin and course of Distribution to crico-arytenoid muscle sions of species th the habits of E CHAPTER II. PHYSIOLOGY OF THE RESPIRATORY APPARATUS OP CHELONIA nistory of theories of Chelonian respiration R. Townson on Malpighi on Milne Edwards on T. Rymer Jones on Miiller, Carpenter, Agassiz, on . Perault on Tauvry, Haro on Townson's views most correct Views of one of the authors in previous paper Division of subject for study Externally visible phenomena of respirat Number of respirations per minute Elements of the respiratory act . Appearance of glottis Function of hyoid apparatus TABLE OF CONTENTS. Inspiration described Expiration described Pause ..... Absence of analogy between Batrachian and Chelonian respiration Obvious objection to the view usually entertained Previous experiments of one of the authors Direct observation of the inspiratory muscles during life. Galvanization of Function of, tested by their removal Expiratory muscle Direct observation of Function of, tested by other means Third period of respiratory movement Partial expiration Simplicity of plan of respiration in Cheloniaus Glottic muscles .... Function of Opening and closing muscles Movement of hyoid arches Necessity for firm closure of glottis Curious instinctive provision for holding air in the lung Physiology of pneumogastric nerve and its branches General remarks .... Experiments on laryngeal nerves Superior laryngeal nerve distributed to both glottic muscles Inferior laryngeal to the opening muscle only Their distribution explains the phenomena observed The superior laryngeal the nerve of sensation to the larynx and glottis Hypoglossal nerve Supplies the tongue only Experiments .... Result of experiments, discovery of a true chiasm of the superior laryngeal nerves No interlateral communication between the recurrent laryngeal nerves Experiments proving this ..... Discussion as to the reasons for this singular distribution of nerves Glottic function in turtles of great importance to their life Hence integrity of, guarded with great care All the nerves but one may be destroyed, and the glottis continue to act Importance of the chiasm in this connection Arrangements for protecting the neural apparatus of the glottis No chiasm in birds or mammals ..... Still to be looked for in Batrachians, and Ophidian and Saurian reptiles Section of pneumogastric nerves in turtles Sensibility of pneumogastric Its relation to cardiac movements Spinal respiratory nerves Respiratory centre Rhythmic repetition of habitual movements in the inspiratory muscles after section of the spine Continuance of glottic movements after section of cervical spine Cause of ...... Final conclusions ..... Appendix. Respiratory muscles in Trionychidae LIST OF WOOD-CUTS. Figure 1. Ilyoid bone ...... Figure 2. Laryngeal cartilages:— A. Cricoid cartilage with arytenoid cartilage superimposed. B. Cricoid cartilage Figure 3. Muscles of inspiration and expiration Figure 4. Glottic muscles and nerves . Figure 5. Diagram of origin of expiratory muscle I'igure 6. The intercommunicating nerve seen from below Figure 7. Appearance of the glottis when closed (A), when open in respiration (B) Figure 8. Diagram of the iutercounnuuicatiug nerve .... RESEARCHES UPON THB ANATOMY AND PHYSIOLOGY OF RESPIRATION IN THE CHELONIA. CHAPTER I. In the whole animal series there is scarcely a creature that would be less likely to suggest itself as a field for discovery than the Turtle. Its temptingly curious form, its Avorld-wide distribution, its limited means of escape and of defence, Avould seem to combine to render it an easy and early object of investigation to the naturalist. And yet the history of Chelonians is full of discordant observations; functions have been misinterpreted, and even important parts of structure have been asserted to exist by some, and again denied by others, until at the present day the uncertain record has forced opinion into error, and permitted the conduct of one of the most important processes of life, that of respiration, to remain misunderstood, and the means of its accomplishment neglected and in part unknown. The view noAV entertained by the leading authorities upon the subject, that Turtles inspire by an act of deglutition, as do the frogs, has prevailed from the first, and doubtless arose from the panting movements of the under part of the throat, common to both orders, and among turtles, especially observable in marine species. It Avill be the object of this paper to shoAV that this vieAV is incorrect, that turtles do not SAvalloAV the air in breathing, but that their respiratory act is effected by inspiratory and expiratory muscles situated within the trunk. The solid thorax clearly indicates that Chelonians do not enjoy the perfect respiratory mechanism of the highest vertebrates. The ordinary tranquil respiration of mammals, Avhen the ribs are at rest and the cavity of the thorax is enlarged by the descent of the diaphragm alone, is, hoAvever, very strikingly analogous to that of turtles, in Avhich the cavity of the shield is enlarged by the»contraction of the muscles of the flanks. In tracing the anatomical history of the organs of respiration in Chelonians, the earliest Avork to which Ave have had access is a " Dissertation on the Respiration of the Tortoise," by Robert ToAvnson, LL. D., AAiritten at Gottingen, May, 1795; and as Ave find in it a brief revieAV of all that Avas knoAvn previously upon the subject, Ave have taken the privilege of embodying this rare and- interesting paper in the present sketch. This avc do more cheerfully as an act of justice to the author; for, having conducted our mquiry with a full knoAvledge of the opinions of modern l o A N A T O M Y AND PHYSIOLOGY OF authorities, we Averc surprised, on aftorAvards learning the singularly truthful views of ToAvnson, to find they had fallen unappreciated, and that, in many instances, they had not even been honored by a notice, or, Avhen so noticed, had been men- tioned only to be condemned. Physiological Observations on tiie Amphibia. Dissertation t7te Third, on the Rr.^ti ration of the Tortoise. Robert Town son. The first inspection of the structure of the animals I haA-e lately treated of, the Ran.e and Salamandrae of Linnaeus, will shoAV that respiration cannot be performed in them as it is in man and animals similar to him; the absence of the osseous parts and diaphragm is sufficient to demonstrate this; and though, on the records of physiology, there are instances of the continuation of respiration after the mobility of the osseous parts had ceased, yet, as those Avore only instances of suffering nature, Avhere the accompanying assistant, the diaphragm, still continued in full energy, phy- siologists ought, likewise, in examining the structure of the animals I am iioav to treat of, the Tortoise-tribe, to have suspected that this function Avas not performed in them as it is in us. Yet these hints given by this anomalous structure have cither been neglected or made an improper use of, and the manner of their respiring remains in the greatest obscurity to the present hour. Before I proceed to shoAV the present state of our knoAvledge on this subject, by giving the opinions of the celebrated anatomists and physiologists AArho have Avritten upon it, I Avill just observe that, as the impossibility of respiration being performed in the frog-tribe, in the usual man- ner, consists in the absence of the ribs and diaphragm, so here the immobility of the Avhole bones of the trunk, and absence of the diaphragm, form the insuperable hindrance, and not a deficiency of solid parts as in the preceding; for a modifica- tion of the ribs and sternum here envelops the whole animal. The diaphragm, though said by some to exist, is really Avanting. Blasius, however, asserts its existence, and describes it thus: "Diaphragma insigne admodum, oblique a pectoris anteriore inferiorcsque parte sursum adscendet, lateribus primo, hinc dorso firmiter adlnercns; altiorem adeoque situm in posticis obtinet, quam in anticis, contra ac in hominc, canibus bobus aliisque animalibus obscrvamus, ubi anteriora sublimem majis locum habent posterioribus. Membranosum hoc totum notatur, similiter ac in avibus variis deprehendimus, nullis fibris carneis manifesto gaudens. Distinguit equidem thoracem a A'entre inferiore, ast non sit in animalibus aliis: Pulmoncs enim cum hie sesc in hoc magis, quam illo ventre exhibeant magne parte, diaphragmate haud includuntur, imo vix aliqua parte. Extendit se supra hepar partesque alias ipsi adsitas, usque ad Aesicam urinariam cui valide adeo unitur tota superficie supe- riore ut non nisi magno artificio separari queat. Superius pcricardio jungitur." But I am convinced he has taken the peritonaeum for it. I have sought for it in vain, as avcII as other zootomists; neither GotAvald nor Wallbaum has observed it, and the French academicians, Avho dissected one near five feet long, say, that " la tortue a non seulement son ecaille, qui lui sont lieu de thorax, absolument immobile, mais nous ne lui aA-ons trouve n'y de diaphragme, n'y d'autrcs parties qui puissent supplier a ce mouvement." This deficiency of the requisite mechanism for respira- RESPIRATION IN TnE CHELONIA. 3 tion has led some physiologists to explain this important function upon principles inconsistent with sound physiology, analogy, or experience. Perault attributes the expansion of the lungs, and consequent inspiration, to the elasticity of the mem- branes forming their cells; and the expiration to the compression of muscles, of Avhich, he says, these animals have plenty. " Apparement," he says, "il est neces- saire de supposer que l'inspiration se fait par le ressort des ligamens durs et fermes qui composent lcs mailles qui ont ete decrites: en sorte que lorsque les muscles qui peuvent comprimer le poumon viennent a se relacher, les ligamens s'etendent et elargissant les ouvertes de toutes les vessies augment la capacite de tout le poumon." Varnier boldly asserted that the whole process of respiration, both expiration as Avell as inspiration, was effected by the lungs themselves alone, by the means of their muscular texture, as a muscular netAvork surrounded them, by which means they could respire by the alternate dilatation and contraction of the vesicules Avith- out the aid of the other instruments of respiration. He says, " Je parvins a me demontrer a moimeme que le poumon de la tortue etoit entoure d'un reseau mus- culaire que par ce moyen ils etoit parfaitement irritable, qu'ils avoit une action propre, independente des autres agens de la respiration et qu'ils pouvoit inspirer par lui mOme;" and soon after adds, "le muscle du poumon de la tortue qui produit un mouvement convulsive," and then says that, "dans le tortue le poumon est cel- lulaire; les cellules se correspondent comme dans la grenouille; le muscle enveloppe toute la masse, et en se contractant la remue toute entiere;" and concludes his memoire by saying, "le poumon est un organ actif; qu'il est le premier et le prin- cipal agent de la respiration, et que cette fonction depend, comme dans les amphibies, de la dilatation et contraction alternative des vesicules qui determinent alternative- ment la contraction des muscles inspirateurs et expirateurs, et cela independamment de la volonte." Admitting the lungs to possess this muscular texture, Avhich could not be perceived by Haller and the best anatomists, they Avould still be ill adapted to inflate by their OAvn poAver. AVe learn, through the Transactions of the Royal Academy of Paris, that it Avas the opinion of Monsieur Tauvry that they breathed only in walking. " La tortue est enfirmee entre deux ecailles immobiles, et elles n'a d'ailleurs aucun diaphragme qui puisse servir a une compression alternative des poumons. Dans cette difficulte d'expliquer sa respiration" Monsieur Tauvry s'est avise d'en rapporter la cause au mouvement du marcher; quand la tortue est en rcpos, sa tete et ses pies sont retires sous l'ecaille superieure, et la peau qui l'enve- loppe entierement est plisse, mais quand l'animal marche, il pousse au dehors sa tete et ses pies; sa peau s'etend, puisqu'elle est tiree par ces parties, et par consequent elle forme interieurement un plus grand espace, et c'est dans cet espace vuide que l'air exterieur est oblige d'entrer." This explanation, which is very anomalous Avith everything Ave knoAV of this function in other animals, I put to the test of the following experiments, Avhich proved it erroneous. I took the Testudo orbicularis in its contracted state, and Avrapt it up in paper, binding it all round Avith bandages so fast, that the testa and sternum Avere brought so near before as not to admit the exit of the head. I then made an aperture in the paper opposite to its nose, and thus deprived of every motion, I placed it before the flame of a candle, yet I found not only that it bleAV the flame, but sometimes so strongly as nearly to extinguish it. 4 A NATO M Y A X D P II Y S I () L O (i Y () K This experiment, though conclusive against the opinion of Tauvry, T strengthen by another; in this 1 kept the legs out, binding them Aery firmly under the sternum, the head being contracted as before, yet I still observed that it breathed, and as in the former experiment, sometimes Avith great force. The respiration, therefore, of the tortoise has no more connection Avith its other motions than that of other ani- mals. But Morgagni, Avho Avas, as I have mentioned in the second dissertation, acquainted Avith the manner of respiring in frogs, Avhich I have giA-cn in detail, supposes that the tortoise respires in the same manner; for, speaking of the frog, he says: "Inspiratio autcm iis instrumentis per qua? inferior bucca? pars amplificata animal contracta aerem in pulmones compellit;" and then adds, "quin imo id ipsum, dum fluvialem quandam testudinem vivam inciderem, observavi invenique, totam cam partem qua? intra eavitatem mandibular inferioris est, multum posse extrorsum curvari ut hinc aer immitti posset, pulmonics vero fibrarum rite firinari, ut hinc aer vicissim posset remitti." XotAvithstanding the high reputation of Mor- gagni, I must dissent from the opinion of the tortoise respiring like the frog. I Avill not say that none of the genus do respire in this manner, as I have had no opportunity of examining any of the turtles. I Avish to be understood as speaking of the Tcstudo orbicularis, my observations having chiefly been confined to this species, though I think I may say the same of the grape a and palustris. Yet the opinion of this celebrated man is supported by Coitor and Yarnier saying that, after the sternum is taken off, and the lungs are laid bare, the animal can still inflate them. But if, after the sternum Avas taken off, the peritoneum cut through, and the lungs laid bare, these appeared to Coiter and Yarnier to inflate, this might not have pro- ceeded from any poAvcr residing in the lungs themselves, nor from any air being impelled into them by the muscles of the throat, but by the parts in contact Avith them, as the neck before, and the muscles behind (which I shall soon describe), shortening them, in which case they Avould appear more distended, though the quantity of air Avithin Avas not increased. The tortoises Avhich I opened I never observed to inflate their lungs as the frogs do; nor did the anatomists mentioned by Yalentini observe it, for they say, " Pulmones enim depressi remanebent, ncc inflabuntur ab ilia aeris attractione quod fieri potuisset tamen ab animali adhuc vivente licet capite truncato, quod ego subito, antequam aperiri, curarem, abscindi jusseram." Y'et adds, "Yitalis autem haec testudo actiones habuit horse spatio absque corde scd et absque capite; nam pedas movit ad factum nostrum et sine eodem quoque retraxit." These are the opinions of the older anatomists; and amongst the moderns I know of none who have said anything on this subject. Being dissatisfied Avith them, I entered into the investigation by actual observation, and opened one for this purpose. The sternum being taken off with great care, the periosteum presents itself as a strong white membrane like parchment; Avhen I had cut through this, I found many muscles inserted into it, particularly over the scapulae and os pubis, which, in the contracted state of the animal, are not far asunder. Just above the os pubis it is connected to the peritonaeum; by this means these bones, with their muscles, are enclosed as in a bag, having the peritonaeum beneath, and the periosteum above; the scapulae, and their connected bones and muscles, are enclosed in the same manner. The peritonaeum being cut through, RESPIRATION IN THE CHELONIA. 5 and the intestinal canal, liver, &c, removed, the lungs, consisting of tAvo lobes, are seen covering nearly the Avhole of the testae; they are cellular, as in the frog, and consist of tAvo lobes, one on each side of the spina dorsi, each of which is sub- divided into four or five indistinct lobules. The cellular texture of these is not uniform; the cells of the middle lobules being the smallest, and those of the last lobule the greatest; this lobule is likeAvise loose, not being tied doAvn on the sides nor beneath, the rest are tied doAvn to the spine. My attention was soon called to observe the structure and office of some muscles in the region of the flanks, which I observed often to be in motion, contracting and extending alternately, and though placed by the side of the hind legs, these were not moved by them. Further, they Avere placed at the end of the last lobule of the lungs, and they appeared to retain their irritability the longest. This was sufficient to lead me to conjecture that these might be the parts by which respiration in these animals Avas performed; and to see them act in their natural position I saAved off, in another tortoise, that part of the shell which covers them, and I then saAV them constantly Avorking. One was now placed nearly in a perpendicular direction, and another, or part of the same, was placed nearer the sternum, lying almost in a horizontal direction. The first in its contraction receded from the testa inwards, Avhilst the latter, in its contraction, observed a contrary direction. When I attributed to them the office of expirator and inspirator muscles, Avhich I supposed them to perform, I was embarrassed, because I could not conceive hoAV a muscle could be a constrictor with its convex side; yet when the expirator, by contracting, had receded from the shell inwards, it appeared, when vieAved from Avithout, to be concave. But this difficulty ceased as soon as I had opened the animal and dissected the parts, for I then found the following admirable contrivance of nature. This part is composed of two distinct muscles, Avith their risings and insertions quite different, yet firmly connected in the middle by cellular membrane. The first rises from the testa near the spina dorsi, and is inserted into the peritonaeum; this is the constrictor of the lungs, or the muscle of expiration. The other is nearly spread over the whole caA'ity betAveen the upper and under shell, Avhere the hind legs are draAvn in during the contracted state of the animal, being inserted into the margin of the testa above, and the mar- gin of the sternum beloAV. The places of insertion of these muscles, and their con- nection in the middle being known, there is then no difficulty in explaining why the muscle, while acting as a constrictor, appeared concave, as it was only the inspirator brought into that position by its antagonist; nor any difficulty in con- ceiving how they carry on the function of respiration; for the expirator being connected, as I have already said, to the testa below and to the peritonaeum above, envelops in a manner the last movable lobule of the lungs; when, therefore, it con- tracts, it compresses this part of the lungs, and by that means expels the air; then ceasing to act, the other contracts, and draAvs the former Avith it, thus a vacuum is formed, into Avhich the air rushes, as in the respiration of those animals which have a thorax. To prove that this explanation was well-founded, and that the motions of these muscles Avere really those of respiration, I made the folloAving experiment. I fastened on the nose of a tortoise a little valve made of Avhite paper, which covered 6 A X A T O M Y AND PHYSIOLOGY OF the nostrils, and with the assistance of a friend, I Avatclied the motions of the soft parts lying Avithin the hollow Avhere the hind legs came out, and I found that, these motions perfectly corresponded Avith the motions of the valve, which was put into motion by the expirations and inspirations of the animal. In this manner I con- ceive respiration to be carried on in the tortoise, Avithout, however, meaning to extend this explanation to the Avhole of the genus Testudo, some families of which 1 have never yet had an opportunity to examine. These animals Avill therefore materially differ from those of the two preceding families in the mode of respiring; the air in them being driven into the lungs by the muscles of the throat, Avhich act like a pair of bellows, whilst in these it is performed by the lungs folloAving the motions of their containing parts, and they will therefore differ from the animals having a thorax chiefly in the form and situation of the parts. Respiration is not, I think, the only office of the muscles Avhich I have just described; they are closely connected to the bladder, and to them, I think, this animal is indebted for the power it possesses of sucking in water by the anus, as I mentioned in my last dis- sertation ; but this investigation I must leave to another time. It Avill thus be seen, while this close observer fully realized that respiration in the turtle was not effected by deglutition, but by muscles of expiration and inspiration situated in the flank spaces, yet, failing to recognize the true office of the anterior muscles, his conception of the respiratory process Avas necessarily imperfect and insufficient, and to this, no doubt, must be ascribed the neglect into which his views have fallen. In LSI!) appeared the most important contribution to the literature of the subject, the monograph of Ludovicus Henricus Bojanus, on the Anatomy of the Testudo Europaea\ This Avork being purely anatomical, Ave have no means of judging as to the author's knoAvledge of the muscular apparatus concerned in respiration, except by the nomenclature he adopts, and some details of description. The in- spiratory muscle and the posterior belly of the expiratory muscle are grouped as abdominal muscles, and described as the obliquus and transversus-abdominis, Avhile the anterior belly of the expiratory muscles, under the name of diaphrag- maticus, is thus referred to: "A corpore vertebrae dorsi quartae et tcrtiae et a costa tertia oriundus, triplici fasciculo complanato, divergentibus eundo; quorum bini ad margincm internum pulmonis, sui lateris, descendunt eique agglutinantur; tertius vero supra pulmonis anterius extremum revolutes ad peritoneum (a pulmonum facie inferiore Arersus cardiam et hepar deflecteus) desinit." It is, no doubt, probable that these names have been determined by supposed homologies of the muscles, and yet we may reasonably conclude that Bojanus had not perceived any relationship betAveen the diaphragmaticus and the transversus abdominis, and did not realize that the broad fibrous membrane extending betAveen them was their common tendon. This conception is essential to the full realization of the respiratory process. G. De Cuvier, bearing in mind the type of batrachian respiration, regards the alternate contraction and dilatation of the throat as movements of deglutition of air, and thinks them a sufficient and the only means by Avhich inspiration is eff cted. The expulsion of the air from the lungs he refers to the agency of tAvo muscles in RESPIRATION IN TnE CHELONIA. 7 the flank, the obliquus and trans versus of Bojanus, at the same time calling attention to the fact that Townson has erroneously attributed to one of these (the obliquus) the function of an inspirator. He thinks also that the analogues of the quadratus lumborum and the rectus abdominis, by compressing the viscera, may assist in expiration. In his Lecons d'Anatomie Comparee, vol. vii. p. 216, Duvernoy's edition, 18-40, we find his opinion in detail. "Le meme mecanisme est mis en jeu dans les cheloniens. La deglutition de l'air est le seul moyen dont ils puissent se servir pour faire entrer ce fluide dans leurs poumons. Ils dilatent et contractent leur gorge alternativement, ayant la bouche fermee, absolument comme les batraciens et par les memes puissances. II est expulsee par deux pairs de muscles analogues a ceux du bas-ventre des animaux precedents. Ces muscles remplissent l'intervalle posterieur du sternum et de la carapace, dans lequel se replient les extremites pelviennes dans l'etat de repos, et c'cst a cet endroit qu'on apereoit dans les cheloniens les mouvements de contraction et de dilatation qui, dans les mammiLres, se voient dans toute l'etendue du ventre.(l) La premiere paire ou l'externe resemblent a l'oblique descendant, elle s'attache a tout le bord anterieur du bassin, a la carapace et au sternum, et s'etend dans tout l'intervalle posterieur de ces deux parties. L'interne ou 1'analogue du transverse est compose de fibres transversales s'attachent superieurement a la moitie posterieur de la carapace pres des Arertebres, descendent en dehors des visceres, les enveloppent et vierinent aboutir inferieurement a une aponevreuse moyenne. Celle-ci passe en partie sous la face inferieure de la vessie, et doit servir a la vider lorsque ces mus- cles se contractent. Ils ne comprennent immediatement qu'une petite portion des poumons; mais leur action s'exercant plus fortement sur les visceres du bas-ventre, ceux-ci pressent a leur tour les premiers organes et en expulsent l'air. Les mus- cles analogues du quarre des lombes et du droit abdominal qui ont ete decrits (t. i, pp. 488, 489) doivent aussi comprimer les visceres abdominaux, et par leur moyens les poumons. Les cheloniens qui ont leur cavite viscerale divise par le pleuro- peritoine a la maniere de celle des oiseaux, ont une de ces cloisons celle qui descend de la partie anterieure du bouclier dorsal, au devant du foie, constitute comme un diaphragme par des fibres musculaires et aponevretiques. Bojanus decrit un muscle diaphragmatique pair compose de fibres musculaires epanouies de chaque c6t6 sur cette cloison, que nous avons fait connaitre comme une sorte de diaphragme (t. iv, 2d partie, p. 651). Son action, quoique faible, peut contribuer a l'expiration en comprenant les poumons. " Peut etre que les poumons se contractent aussi par une force propre que reside dans le reseau tendineux qui entre dans leur composition (Art. 11, de cette Lecon, p. 130). "N. 1. Je crois l'avoir fait connaitre le premier (Bull, de la Soc. Phil. an. xiii, No. 97, p. 279) en demontrant, contrairement a l'opinion de ToAvnson, que les muscles du bas-ventre sont l'un et 1'autre des muscles expirateur. Et cependant e'est a cet auteur qu'on attribue l'explication que j'ai donnee en montrant F inexactitude de la sienne." Dumeril et Bibron, vol. i. p. 176, 1834, described Briefly the mechanism of respiration in chelonians thus: air enters the buccal cavity through the nose, then 8 A X A T O M Y A X D PHYSIOLOGY OF the fleshy tongue is applied to the posterior narcs so as to close them, and the mylo-hyoid floor of the mouth contracts, to force the imprisoned air into the lung. A succession of such acts fills it. Before entering upon the details of description, it may be Avell to premise, that this anatomical section of our paper is intended mainly as an exposition of the muscular and neural apparatus by means of which the movements of air to and from the lungs are effected in chelonians, and Avhilc, to render the subject more intelligible, we shall rehearse the general anatomy of the organs of respiration, avc shall avoid all questions of structure or function irrelative to the point of inquiry, referring the reader desirous of such knoAvledge to the more general Avorks on comparative anatomy. Underlying the floor of the mouth, and embracing Avith its cornua the sides of the pharynx posterior to the jaAV, is the hyoid apparatus, or the tongue-bone, Fig. 1, Fig. 1. Fig. 1. a, a\ lesser cornua; b, b\ greater cornua; c, c', cartilaginous processes; d, d', ossicles for attachment of suspensory ligaments; e, body of bone ;f, fenestrum, closed by cartilage; g, articulations of cornua with. body. an instrument conspicuous for the part it has evidently played in fixing upon its possessor the batrachian type of respiration. It consists of an elongated body, excavated for the lodgment of the larynx and, upper rings of the trachea, and of a cartilaginous process and tAvo bony cornua on each side, connected with it by mova- ble articulations. To the extremity of the anterior or major cornu is attached a knob or ossicle, for the reception of the suspensory ligament. This ligament arises from the mastoid process of the temporal bone of the cranium, and forms the fulcrum on Avhich the apparatus swings backAvards and forwards, moved by alternate contrac- tions of the genio-hyoid and oino-hyoid, and other muscles of the neck. The hyoid RESPIRATION IN THE CHELONIA. 9 bone, in its movements, carries with it the glottis, and removes it from obstructions during inspiration. The larynx, Fig. 2, A and B, consists of a largely-developed cricoid cartilage and two arytenoid cartilages. The cricoid rests in the boAvl of the hyoid bone, is somewhat helmet-shaped, and has on its under surface a visor-like oval fenestrum. This fenestrum is covered by membrane, and is traversed from side to side by the chiasm of the superior laryngeal nerves, of which we shall speak more fully hereafter. Superiorly the cricoid presents an oval opening, filled in by membrane, upon Avhich rest the arytenoid cartilages, one on either side, Avith the glottic slit A Fig. 2 6 b'. Cricoid cartilage; c, left arytenoid cartilage; a', cartilaginous tubercle capping the apex of the arytenoid cartilage; a, oval fenestrum of cricoid, filled in with membrane. betAveen them. The arytenoid cartilages, Fig. 2, A, c, are two irregularly triangular solids, opposing.flat surfaces to each other, their bases incorporated with the superior cricoid membrane, and their apices extending vertically, and terminating in a small cartilaginous tubercle. They are the framework upon Avhich the crico-hyoid and crico-arytenoid muscles act, in closing and opening the glottis. The trachea, smaller in diameter than the cricoid bulb, descends the neck, inclining to the left side, until opposite the margin of the shell it divides into tAvo bronchi, which, curving right and left, open free into the corresponding lungs, at the under and inner edge, a little behind the anterior extremity. The lungs are two wedge-shaped sacs, the base of the Avedge being anterior. They lie in contact with the vault of the dorsal shield, and are separated from each other by the large retractor mus- cles of the neck, the bloodvessels, and nerves. They are anterior and above the peritoneal sac, except the posterior pointed extremity, Avhich projects into that cavity, carrying Avith it a covering of peritoneum. The Avails of the lungs being elastic lend aid to the act of expiration, but as they give no evidence of muscular fibre- to mechanical or galvanic stimuli or to the microscope, it is impossible, for this and for other reasons, to suppose Avith Yarnier that they possess any intrinsic power to assist in the act of inspiration. The turtle Avhich has served us fpr most of our experiments, is Chelydra Ser- pentina, the Avell-knoAvn Snapping Turtle of the United States. Selected at first from the facility Avith Avhich we could procure fine specimens, Ave soon found that its well developed muscular system and its exposed flanks admirably fitted it for the study of respiratory myology, Avhile its middle rank among Testudinata led us to expect, in its organization, more striking ordinal characters than Ave Avould find in the extreme marine or terrestrial families. AVe have therefore adopted Chelydra 2 B b. Cricoid cartilage; a, superior opening; c, trachea. 10 A N A T 0 M Y A X D PHYSIOLOGY O F Serpentina as our typical turtle, and Avill describe in detail the apparatus of respiration as Ave find it in this species, noting, subsequently, the moditications of structure existing in the different genera that avc have had the opportunity to examine. In all turtles avc have found the general plan of the respirator) apparatus constant, an inspiratory muscle in each flank, and an expiratory muscle Avith four bellies, tAvo anterior and tAvo posterior, connected by a broad membranous tendon, inclosing the viscera and capable of compressing them against the under surface of the dorsal shield. The discrepancies characterizing different genera principally affect the origin of the anterior belly of the expiratory muscle; these may naturally be arranged in two groups, those in Avhich the origin is anterior (about the second rib), (see Fig. 5) and extends nearly across the width of the shield, and those in which the origin is posterior (about the third or fou'rth rib), and in extent more limited. The specimens Ave have had the opportunity to examine are too feAv to enable us to determine Avhethcr this structural diversity can be received as an element in determining generic rank. We Avill content ourselves, therefore, at present, Avith the description of each specimen, including a brief notice of its habits and shell- measurements, which may serve as a nucleus for future and more extended observations. Chelydra Serpentina is a carnivorous turtle living in the Avater, under bank-eaves, or at the bottom of streams, and yet capable of moving over the land Avith facility. The under surface of the body is much exposed, the plastron being small and cruciform, and connected AAith the carapace by a narroAV bridge, Avhich Avidcns to join the fourth, fifth, and sixth ribs. The flank spaces are large, flat, and unpro- tected, and the extremities incapable of complete retraction under the shell. The height of the trunk compared Avith the Avidth and length of the carapace is as one to three and three and a half. Carefully Avatching the animal AA'hile breathing, we notice synchronous move- ments of the trunk, of the throat, and of the glottis Avithin the mouth. With the first element of the respiratory act, expiration, the glottis opens, the hyoid apparatus descends and Avidens, the shoulders sink and the flanks become increasingly con- cave; then folloAvs immediately the inspiratory effort, the glottis remaining open, the throat narrows, the flanks become tense, and the shoulders are pushed fonvards as the act culminates; afterwards the muscles relax, the glottis closes, and the creature is at rest until again impelled to renew the air in its lungs, when the same sequence of expiration, inspiration and pause is repeated. AYe shall MIoav the order of the elements of the respiratory act in describing the apparatus by Avhich it is effected. And first, of the muscles of expiration. For the purpose of dissection, it is desirable to place the animal upon its back and fix it, by extending and securing its head, tail, and extremities. Separate Avith a saw the bony bridges connecting the plastron Avith the carapace, and SAvccping a knife close to the inner surface of the former, divide from before, backwards, the deltoid, pectoral, pelvic and flank muscles, the acromial articulations posterior to the first pair of sternal bones, and the loose cellular bands binding the visceral sac to the middle line. This permits the removal of the plastron. Drawing the shoulders for- Avard, cut the ligaments, holding the scapulae to the spine anterior to the first rib, RESPIRATION IN THE CHELONIA. 11 which loosens the entire muscular and bony mass, and facilitates its detachment. The section should be made Avith the lung partially inflated, to secure from injury the anterior belly of the expiratory muscle, Avhich lies in contact with the posterior surface of the serratus magnus. The further removal of the tissues of the flank, and their careful separation from the posterior belly of the expiratory muscle to which they are adherent, completes the exposure. Looking at the result of our dissection, Ave find a tendinous and muscular sac occupying the dorsal shield, filling its entire width in the middle and most of its length; its general form is cordate, the apex dipping into the pelvis, and its anterior notch giving place to the heart and the muscles and vessels of the neck. Much the larger portion of the sac visible is tendon (Fig. 3, g, g'), and has hitherto been regarded as peritoneum, but a closer scrutiny would have revealed its fibres gather- ing from all sides towards an oval centre, in Avhich they are inseparably interwoven. The tendon in many places can be lifted from the peritoneum, by which it is lined. Curving around its anterior and posterior borders are muscular fringes (Fig. 3, d, d! and e), the fibres running from the carapace in lines parallel to the long axis of the trunk. These are the anterior and posterior bellies of the expiratory muscle, the diaphragmaticus and transversus abdominis of Bojanus. These muscles are inserted into the common tendon, and in contracting compress the contained viscera against the shell and expel the air from the lungs. Dividing the tendon through its middle from side to side, and removing the abdominal organs and permitting the lungs to collapse, we are enabled to obtain a satisfactory view of the origin of these muscular bellies from within. The posterior belly arises from the pelvic fascia from a point opposite the anterior third of the ilium backwards to the spine, from the eighth vertebra, and by tendinous fibres from the carapace as far as the sixth rib, the line of origin sloAvly leaving the spinal column as it reaches forwards. Turning outAvards at an obtuse angle, after joining the sixth rib, the muscle follows its posterior edge until near its extremity, where it inclines fonvards and terminates at the fifth rib as it joins the marginal plates, a point corresponding very nearly with the pelvic end of the suture connecting the carapace and plastron. From this sigma-shaped origin the fibres curve backwards and downwards, embrac- ing the abdominal viscera, and unite with the tendon below, forming a regular and well-defined line, varying in position as the muscle is contracting or at rest. Fig. 3, c, represents the lungs distended and the muscle relaxed. This belly, considered by itself, is a strong membranous muscle, someAvhat triangular in shape, the apex being at the edge of the shell, and the base at the pelvis. In a turtle weighing sixteen pounds, the fibres at the apex measured one-half inch in length, while in the middle and at the base they measured respectively five and one-half and four inches. The anterior belly arises from the vertebral margins of the second and third intercostal spaces, from the second costal arch, from the second rib along two-thirds of its length, and across the carapace in a line curving backAvards and outAvards, from the third and fourth ribs, near their junction with the marginal plates. It will thus be seen that the outermost origins of the anterior and posterior bellies closely approxi- 12 A X A T O MY AND P II Y S I O L O G Y () F mate above the plastron where it meets the upper shield, while at the middle line of the body the origins arc separated by the distance of the eighth from the third vertebra. The fibres composing the anterior belly arc close and firm for the outer Fijr. 3. Fig. 3. Muscle? of inspiration and expiration.—a a' a" a'", margin of carapace; b b', portion of plastron in position; e, posterior belly of the expiratory muscle on the right side ; d d', anterior bellies of the expiratory muscle; e, reticulated portion showing the luug beneath; //', inspiratory muscle of the left side; f", central tendon ;/'", t.nlinous ligament; g F Fig. 4. Glottic muscles and nerves. —a a', crico-hyoid ; the muscle overlying it is the crico-aryte- noid ; 6, superior laryngeal nerve ; b\ communicating branch; b", branch to crico-arytenoid; b"\ branch to crico-hyoid ; c, recur- rent laryngeal; dd', glottic slit; t, point of hyoid bone; /, tongue. Fig. 4. is determined by tAvo pairs of muscles situated about the glottis, and controlling its movements. These are tin" crico-arytenoid and the crico-hyoid of Bojanus; to the former is intrusted the opening of the- glottic lips, while the latter, acting as a sphincter, servos to close them. The cricosirytenoid (Fig. 4) lies beneath the mucous membrane, superficial to the crico-hyoid, and crossing it nearly at right angles. It arises from the sides of the cricoid cartilage anteriorly, and is inserted into the body and vertical limb of the arytenoid as far as its extreme point. The crico-hyoid (Fig. 4, a a') arises from the body of the hyoid bone anterior to the depression for the larynx, its middle resting upon and exterior to the arytenoid cartilage. It is inserted into the cricoid car- tilage at its anterior raphe. The muscles of the two sides approximate each other at their origins, and in- terlace at their insertions, forming an elliptical muscle surrounding the rima glottidis. Our opportunities for studying the arrangement of the respiratory muscles in other turtles than Chelydra have been limited to the representatives of two families, Chelonioidae and Emydoidae. Among the Chelonians avc have examined but one species, Chelonia mydas, the Green Turtle of the Atlantic Ocean. Its habits are entirely aquatic, seeking the land only for the purpose of depositing its eggs. The body is flat, the under surface avcII covered by the plastron, leaving, however, naked flank spaces, as in the snapper. The union betAveen the plastron and carapace extends from the second to the seventh rib. The inspiratory muscles occupy the flanks, arising a half inch or more Avithin that part of their boundary Avhich is formed by the plastron. The central tendon exists, and is Avide and irregular, and extends the whole length of the muscle. The origin of the expiratory muscles is similar to that found in Chelydra; the muscular bellies are shorter, hoAvever, and the common tendon broader and longer in accordance with the shape of the turtle. The dimensions of the shell are—length, 38 inches; width, 28 inches; elevation, 13 inches. Among the Emydoidae we haA^e examined individuals from eight genera, and find them to present considerable variations in the origin of the anterior belly of the muscle of expiration. And as these differences seem to characterize groups in harmony with the subdivisions of Agassiz, founded on minor differences of form observed in this family, Ave shall folloAV his classification in their description. The first subdiA'ision suggested by this distinguished observer, and styled y<:c- tomydoidae. is thus characterized: " The body is rather flat. The bridge connect- ing the plastron and carapace is Avide, but flat. The hind legs are stouter than the RESPIRATION IN THE CHELONIA. 15 fore legs, and provided Avith a broad Aveb, extending beyond the articulation of the nail joint. The representatives of this group are the largest and most aquatic of the Avhole family." Of the genera included in this sub-family Ave have observed four: Ptychemys, Graptemys, Malacoclemmys and Chrysemys. Fig. 5. Diagram of carapace of turtle, showing with the dotted lines the two principal types of origin of trie expiratory muscle. The left side of the diagram shows the line of origin in the most aquatic species. The right side that of the most terrestrial. The numbers indicate the ribs. Ptychemys rugosa, Ag.—The inspiratory muscles are found in the flanks as usual, but they have no central tendon, a simple line or raphe marking the junction of the converging fibres. The anterior belly of the expiratory muscles arises from the vertebral margin of the fourth and fifth intercostal spaces and from the surface of the fourth rib near its posterior edge for a distance one-third its length. From this right-angular origin the fibres diverge, expanding over the upper and anterior surface of the lung, to join the common tendon at the anterior and inferior pulmonary margin. The fibres extending back on the under surface of the lung, as indicated by the dotted lines (Fig. 3, h), are numerous and large in this species, and seem almost to foreshadoAV the muscular separation betAveen the thoracic and abdominal viscera in higher vertebrates. The posterior belly in its origin presents no variation from that of the Snapper. Its outermost fibres, hoAvever, are much developed, forming a muscular band which reaches forwards nearly as far as the anterior junction of the carapace and plastron. The dimensions of this turtle are—length, 11 inches; Avidth, 8^ inches; and elevation, 5 inches. Ptychemys viobiJiensis.—Shell measurements. Length, 13{£ inches; Avidth, 9f4^ inches; elevation, 6^ inches. The general shape and appearance of this turtle resembles P. rugosa. The anterior and posterior extremities of the bridge con- 16 ANATOMY AND PHYSIOLOGY OF nocting the plastron and carapace are much more strongly involute than avc have observed in any other species. AY hen the shell is separated, they appear like tour projecting keels directed toAvards each other, the front ones looking imvards and backwards, and the posterior ones looking imvards and slightly fonvards. The con- cavities thus formed before and behind, external to the keels, lodge projecting por- tions of the lungs. The anterior and posterior keels projecting into the space usually occupied by the air sacs, deeply indent them, and cause them to present a tabulated appearance, Avhich they retain Avhen removed from the shell. Besides these four large indentations, there are smaller ones, in the edge of the lungs, one in front and tAvo or more between the keels. To the inner side and behind the posterior keel lies the large posterior lobe occupying chiefly the flank space immediately above and in front of the inspiratory muscle. The reticulations of its interior structure are much larger and more coarsely marked than those of other parts of the lung. The anterior belly of the expiratory muscle arises from the vertebral margins of the third and fourth intercostal spaces, and from the carapace in a line diverging at an angle of 30° from the spine for the space of tAvo inches, crossing the fourth rib. From this origin the fibres cover the front of the lung, the anterior and interior ones being distributed as in P. rugosa, and OAving to the intrusion of the anterior keel upon the lung, the external fibres are displaced, so to speak, Avith the portions of lung to which they belong, Avhich portions lie immediately back of the ridge or keel so often referred to. The largest band of those lateral fibres finds its Avay between the tAvo lobules of the lung which lie first and second in order of succession behind the ridge. The posterior belly arises from the pelvic fascia, from the eighth and seventh verte- brae, and from a curved line Avhose comexity looks fonvards, and Avhich terminates in front of the posterior projecting keel about tAvo and a half inches above the pos- terior angle of junction of the carapace and plastron. This line is rendered more sharply convex at its external third by the projection imvards of the keel alluded to. The muscular fibres curve around the posterior part of the lung. The inner ones for half the Avidth of the muscle are about tAvo and a quarter inches long; and from this point they increase gradually in length to the external edge, Avhcre they are longest, extending fonvard in a tongue-like band about five and a half inches. In the single specimen examined Ave foimd on the left side a feAV additional fibres reaching forwards and imvards at least tAvo inches beyond the main body of the muscle. The inspira- tory muscle arises as in P. rugosa. It has a linear central tendon, bifid at its pos- terior extremity, the shorter terminating arm being external. Into the tendon and its branches the muscular fibres are inserted as in other species. In Graptemys (jeograjJu'ra, Ag., the inspiratory muscle is, in its general features, the same as described in other species, and differs only in not having even a central raphe, the convergent fibres interlacing at the middle of the muscle in an imperfect netAVork AA'hich serves to replace the tendon usually found in this situation. The an- terior belly of the expiratory muscle arises from the vertebral margin of the third -and fourth intercostal spaces, and continuously from the costal margin of the third space nearly its entire circumference, and from the surface of the fourth and fifth ribs. The lines of origin diverge at an angle of 30° from the anterior margin of the third intercostal space, and in this specimen, the inner line bordering the spine measures RESPIRATION IN THE CHELONIA. 17 three-fourths of an inch, and the outer one, stretching along the intercostal space and across the shield, one inch and three-eighths; from this origin the fibres spread over the anterior part of the lung and are inserted into the common tendon and into the peritoneal covering of its under surface. The posterior belly is similar to that of P. rugosa, the muscular band underlying the bridge which joins the carapace and plastron being somewhat wider. The dimensions are—length, 8^ inches; width, 6 inches; elevation, 3^ inches. In Malacoclemmys palustris, Ag., the inspiratory muscle is the same as in Geo- graphica. The anterior belly of the expiratory muscle arises from the third and fourth spaces at their vertebral margins, and from a line running across the shield to the fourth rib, diverging at an angle of about 70°. The posterior belly is like that of geographica. The dimensions are—length, 7 inches; width, 41 inches; elevation, 3| inches. Chrysemys picta, Gray.—Inspiratory muscles as in E. terrapin. The anterior belly of the expiratory muscle arises from the vertebral margins of the third and fourth intercostal spaces and a slip from the fifth, and from across the carapace to the junction of the fourth and fifth ribs, the divergence being about 30°. The posterior belly as in geographica. Dimensions—Length, 4| inches; width, 3| inches; elevation, If inches. Of the second and third subdivisions Ave have examined no specimens. The fourth, Clemmydoidae, is characterized by " their more arched though elongated form, and the more compact structure of their feet, the front and hind pairs of which are more nearly equal, and their toes united by a smaller web; they are less aquatic and generally smaller than the preceding." Of these we have dissected representatives of three genera, Nanemys, Calemys and Glyptemys. In Nanemys guttata, Ag., the inspiratory muscle presents no peculiarities. The anterior belly of the expiratory muscle arises from the vertebral margins of the second and third intercostal spaces and from part of the fourth, and from the posterior edge of the second rib as far as its extremity; from this extensive origin the fibres descend over the lungs, covering the front and anterior part of their lateral walls. The posterior belly resembles that of the Snapper. , Dimensions are—length, 4JF inches; width, 2f inches; elevation, 1T9^ inches. In Calemys Muhlenbergii, Ag., the muscles are the same as in guttata. Dimen- sions—length, 31 inches; Avidth, 21 inches; elevation, 1| inches. In Glyptemys insculpta, Ag., the muscles are the same as in guttata. Dimen- sions—length, 4T9g inches; width, 3T7g inches; elevation, If inches. In the fifth subdivision, Cistudinina, "the body is remarkably short and high, slightly oblong, and almost round. The plastron, which is movable upon itself and upon the carapace, as in the Evemydoidae, is also connected with the carapace by a narrow bridge; but the feet are very different, the toes, as in the Testudinina, being nearly free of Aveb. Their habits are completely terrestrial." Of this sub- family we have examined one species, Cistudo Virginea, Ag. The flank spaces. in Avhich the inspiratory muscles play are extremely deep, owing to the high carapace. The amount of muscular fibre is relatively greater than- in the other turtles, and the central tendon is narroAv, and irregularly triangular in shape. The 3 18 A N A T O M Y AND PHYSIOLOGY O F anterior belly of the expiratory muscle arises from the Acrtcbral margins of the second and third intercostal spaces and from the second rib throughout its length. The posterior belly is like in origin to that of other Emydoidae; the muscular fibres are longer, however, and terminate squarely in the tendon, as docs also the anterior belly. For convenience of reference, we haAe throAvn into a tabular form the measure- ments and muscular origins of the above genera. Species. Mode of Life. Dimensions in inches. Origin of Respiratory Musoles. N E0TEMYDOID..E. L. W. £. Ptychemys rugosa . . 11 8TV 5 4th and 5th vertebra?. " mobiliensis 13jf q 4 6tV 3d and 4th Graptemys geographica Aquatic. 8* 6 H 3d and 4th " Malacoclemmys palustris 1 43 3§ 3d and 4th " Chrysemys picta . . . H »* If 3d and 4 th " Clemmydoid-e. Xnnemys guttata . . 4iV 03 -■I 1A 2d and 3d " Calemys miihlenbergii . Less aquatic. 3* 2~> 1 * 2d and 3d " Glyptemys iusculpta *i9s 'i 7 is 2d and 3d " ClSTUDIMN.K. Cistudo virginea . . . Terrestrial. 4f 8| 2! 2d and 3d " A glance at the table will show that in the most aquatic species of Emydoidae the origin of the anterior belly of the muscle of expiration is from nearly the middle of the shell; Avhile in the less aquatic and terrestrial genera it is from the fonvard part, and much more extensive. This arrangement is too uniform to be passed by unnoticed, although our facts are so few that we cannot form any con- clusions as to its generic meaning. Whether the same diversity of origin exists in the genera of other families, and bears a similar relation to their family rank, and also Avhether this origin is modified during the development of the turtle, we must leave for future inquiry. The neural apparatus of respiration in Chelonians, as in the Mammalia, consists essentially of the nervus vagus supplying the larynx, of spinal nerves distributed to the respiratory muscles of the trunk, and of the medulla oblongata, the common centre through Avhich the synchronous movements of the glottis and of the flanks are incited and controlled. Between the ganglionic enlargements supplying the upper and loAver extremities, the spinal cord is attenuated, the nerves coming from this region being restricted, by the existence of a bony thorax, almost entirely to those concerned in the moA-ements of respiration. The disposition of the trunks of these nerves closely resembles that of the intercostals in man. Escaping from the spinal canal at the intervertebral foramina they traverse the carapace in parallel lines betAveen the ribs, giAring off branches from time to time to their appropriate muscles. By dissection and by mechanical irritation of the peripheral end of the cut nerve, exciting contraction of different fibres, we have determined that the fila- ments finally distributed to the expiratory muscle are derived from the first, second and third dorsal nerves for the anterior belly, and from the fifth, sixth and seventh for the posterior belly. The sixth and seventh nerves are also the sources of supply to the muscles of inspiration, the seventh being distributed over the inner or pelvic side, RESPIRATION IN THE CHELONIA. 19 and the sixth to those fibres connecting the central tendon and carapace. Section of the medulla spinalis in the cervical region effectually intercepts communication between these nerves and their usual source of excitation. Under these circum- stances the muscles of the trunk remain at rest, although the movements of the glottis indicate that the creature feels the respiratory need. These glottic move- ments continue normal even after the further section of both pneumogastrics. The respiratory nerve of the larynx, the par vagum, emanating from the medulla oblongata, passes out of the cranium at the posterior jugular foramen, and courses down the neck within the sheath of the cervical vessels. Soon after leaving the skull, it gives off the superior laryngeals, and low down in the neck, opposite the aorta, the inferior laryngeal, the two branches that interest us at present. The superior laryngeal (Fig. 4, b), soon after separating from the parent nerve, approaches the major cornu of the hyoid bone, and under shelter of its pos- terior border, follows it closely to its junction with the body, then winding spirally forwards, it crosses the articulation, and runs along the margin of the excavation in close proximity with the larynx. In this position it gives off three principal branches. 1st. A communicating branch (Fig. 4, b'); 2d. A branch to the crico-arytenoid, or opening muscle of the glottis (Fig. 4, b"); and 3d. A branch to the crico-hyoid or glottic sphincter (Fig. 4, b'"). The communicating branch (Fig. 6) is a relatively large nerve, but has hitherto escaped observation; it is easily brought into view by dividing the trachea and lifting it forwards. It passes beneath the larynx directly from side to side, traversing the membrane of the cricoid fenestrum, about its middle. It is composed of fibres derived in part from each of the superior laryngeal nerves, which cross each other, to be distributed to the glottic muscles of the side opposite to that from which they originate. This re- markable nerve, we believe, furnishes the only known instance in nerve anatomy of an extracranial chiasm. Some few filaments penetrate the cricoid membrane, to be distributed to the mucous membrane of the larynx, and are doubtless sensitive fibres. At page 20 of the physiological section, will be found the experiments by which we have determined the function of this intercommunicating nerve. The second and third branches present no peculiarities; they penetrate the muscles and are lost to view. Sometimes, however, they can be seen to divide into three or more filaments before so doing. Fig. 4, c.—The pneumogastric, before reaching the aorta, gives off a branch, which, Avinding around the arch, changes its course upwards, and soon divides into two nerves; one crossing the neck enters the oesophageal tissue—the other, the recurrent laryngeal (Fig. 4, c), joins the trachea, and, in close contact with its side, follows it to the larynx, and enters the crico-arytenoid muscle. There are no fibres from the recurrent distributed to the crico-hyoid directly, or indirectly through communication with the superior laryngeal. Fig. 6. The intercommunicating nerve seen from below.—a a', su- perior laryngeal nerves; a", inter- communicating nerve crossing the fenestrum of the cricoid cartilage; 6, crico-arytenoid muscle; c, crico- hyoid muscle. A N A T O M Y AND 1'IIVSIOLO V, Y O F CHAPTER II. The preceding chapter has been altogether taken up with anatomical descriptions of the respiratory organs and their appendages. So much that Avas neAv Avas met Avith during our dissections, that it Avas thought better to separate the description of the anatomy from the physiological statements. AVe have thus the physiology of the respiratory organs still to describe, and this can noAV be done Avithout repeating any more of the anatomical detail than is necessary to enable the reader to comprehend the actions of the organs concerned. The history of the theories entertained as to the nature of the respiratory motions in turtles, appears to us one of the most extraordinary in the records of science. Totally misunderstood by the earlier naturalists and biologists, or con- founded as to type Avith the respiration of Batrachians, this function in turtles Avas first rightly comprehended, at least to some extent, by E. Townson in the latter part of the last century. How far he Avent, and Iioav far he Avas correct, we shall more fully point out in another place. The authority of more eminent naturalists, and an obstinate disposition to associate the turtle Avith the frog, and to insist on similarity as to the execution of their functions, gradually drew attention from Townson's statement, and more modern authors have paid it no deference Avhat- ever; yet, as Ave shall distinctly shoAV, all the later Avritcrs arc utterly wrong, and his opinions as to the facts in question are thus far the only ones which seem to be correct. In reading his very ingenious essay, which we have elsewhere quoted at length, p. 2, it is hard to see hoAV the statements and evidence could have failed of more respectful and permanent attention. A complete review of the theories enter- tained in regard to the respiratory function in Chelonian reptiles, will more fully illustrate the above remarks. As early as 1719, Malpighi1 described the respiration of turtles as similar to that of frogs. Both alike Avere supposed to distend the lungs by swallowing air, so that, in place of air being drawn into the lung-sacs, it was forced into them by the movements of parts above the trachea; but Avhile in the frog this was effected plainly through the aid of the bellows-like mouth, in the turtles their vast hyoid apparatus Avas by some supposed to constitute a forcing pump of similar purpose and nature. The authors of Malpighi's era shared these opinions, and with the one notable exception above mentioned, they have stood almost uncontradicted up to the date of a paper by one of the authors of the present essay. * Adversaria Anatomica, t. v Animadv, 29. RESPIRATION IN THE CHELONIA. 21 The latest and best Avork on comparative anatomy and physiology1 thus describes its author's conclusions as to this subject: "C'est aussi par des mouve- ments de deglutition que la majeure partie de l'air inspire est poussee dans les poumons chez les tortues; mais ici ce mode de respiration est necessite par une disposition organique inverse de celle que je viens de signaler chez les Batraciens." M. EdAvards then proceeds to point out the rigid form of the turtle's frame, the absence of mobile ribs, and the consequent necessity for the belief that the lungs in these animals cannot be dilated from without, as occurs in mammals. The same opinion is held by nearly all Avriters at the present time; but some, in place of describing the process as one of deglutition, effected alone by muscles on the floor of the mouth, regard the hyoid apparatus as the true forcing pump concerned in propelling air into the interior. Thus, T. Rymer Jones,2 after describing the fixity of the bones of the chest in turtles, adds, that "under these circumstances, as a compensation for the want of mobility in the chest, the os hyoides and the mus- cles of the throat are converted into a kind of bellows, by which the air is forced mechanically into the lungs, and they are thus distended at pleasure." In fact, the submaxillary space with the hyoid arches, are in continual motion in turtles, and this movement precisely resembles the like action in frogs; but while in these latter it is really a respiratory act, in turtles, as we shall shoAV, it has other purposes, and, while it has deceived observers, may be proved to have no influence of any moment in carrying on the breathing process. Muller3 gives a like account, and adds, that expiration is effected by means of muscles between the loAver shield or plastron, and the posterior extremities. Carpenter4 has a brief description of the respiration in chelonia, which corresponds to the general opinion already quoted above. Prof. Agassiz's description5 being one of the latest, and certainly one of the most authoritative statements, we quote in full, to complete our history of the generally received ideas as to the mechanism of chelonian respiration. " Here, again, we meet with a very striking ordinal character. The turtles swallow the air they breathe. The breast box, Avhich includes the lungs, being immovable, a respiration like that of the other reptiles, the birds, and mammalia, performed by the expansion and compression of the breast box, and consequently of the lungs, is impossible. Owing to the peculiar structure of their trunk, breath- ing is therefore only possible for turtles, by a pressure of the air from the mouth doAvn into the lungs; but though we are persuaded that this swallowing of the air constitutes the main act in the process of breathing, still Ave are inclined to believe, against the opinion of other anatomists, that the diaphragm, which in turtles is very much developed, and attached to the lungs, takes also its part in that act. Moreover, the muscles of the shoulder and of the pelvic region may assist in that 1 Milne Edwards, Lecons sur la Physiologie et l'Anatomie comparee de 1'Homme et des Animaux, t. deuxieme, deuxieme partie, p. 381. 1858. Paris. 8 The General Structure of the Animal Kingdom, p. 56T. s Physiology—London translation, p. 360, vol. ii. * Gen. and Comp. Phys., p. 493. 5 Contributions to the Natural History of the "United States, vol. i. p. 281. 22 A NATO M Y A N D P II Y S I O L O <; Y OP operation, either by immediately compressing the lungs, Avhich generally extend in turtles from one end of the trunk to the other, or by pressing the bowels against them. k' The act of swallowing the air is chiefly performed by the apparatus of the tongue-bone, and the tongue itself, which, by its large size, facilitates the operation. Being draAvn backAvards and upAvards, this organ shuts up the choannoe, and at the same time opens the slit of the windpipe, situated just at its base, thus giAfing to the air a passage into the windpipe, and at the same time preventing its entrance through the choannse into the nose. In this Avay, the tongue takes the place, in a certain sense, of the velum palatinum of the higher vertebrata, which is Avanting in turtles. After the air has passed into the windpipe, the tongue is draAvn for- Avards, and thus the longitudinal glottis is again closed, AA'hile now the choannaj arc again opened to a free communication with the cavity of the mouth." Professor Agassiz adds, in a folloAving note:— " We find the same mode of breathing in the class of Batrachians, but for an entirely different reason, namely, on account of the absence of ribs." Also. " The existence of a diaphragm is erroneously denied to turtles by Dumeril and Bibron, Erpetologie generale, 1, p. 175." In the above description, Prof. Agassiz exhibits some doubt as to the correctness of received vieAvs on this subject, and speaks of the musculus diaphragmaticus (Bojanus) as haAing something to do Avith the act of respiration, which he thinks may also be aided by other muscular parts, as those concerned in locomotion, and by certain pelvic muscles Avhich he does not specify by name. We shall show as Ave proceed that, although the muscle covering the lungs may be homologous with the diaphragm of mammals, it is really a muscle of cxjriration, and therefore not analogous to the diaphragm Avhen regarded from a physiological stand-point. Except for the purpose of completing this brief history of opinions held now or abandoned, it is only requisite to allude to the vieAvs of Perault, who attributed the inspiratory act to the elasticity of the lungs, and the expiratory motion to muscles of Avhich, he says naively, the turtle has an abundance". M. Tauvry, whose vieAvs Milne Edwards partially indorses, attributed the Avhole respiratory act to the changes in the capacity of the chest, caused during locomotion, by the advance of the head and limbs from and their retraction within the carapace. INI. Haro1 sup- ports the same A*ieAvs, but, although both are successful in showing that these movements may alter the capacity of the chest-box, and thus under some circum- stances modify respiration, neither has proved that respiration relies for its continued occurrence upon these motions, nor would such a supposition be entertained for a moment by any one who surveyed the mechanical conditions Avhich are effective in carrying on respiration in other animals. That the locomotive movements may, and perhaps do at times modify the respiratory process, may be taken for granted. That other agents are constantly employed in this function is not less clear, nor shall avc have any difficulty hi disproAing M. Haro's theory by unanswerable facts. 1 Mem. sur le respiration des Grenouilles, Ann. des Sc. Nat. 2 serie, t. xviii. p. 48. RESPIRATION IN THE CHELONIA. 23 The author to whom we have alluded as the only one who has approached to a clear comprehension of the true mechanism of respiration in turtles is Kobert Town- son, LL. D.1 The anatomy of the respiratory muscles of the breast-box is described by this author, as we have elsewhere shown, Avith much correctness. His statement as to the mechanism of the movements of the chest and belly muscles in breathing are, also, remarkably truthful, and are approached in this particular by those of no other or later authors. He came to the conclusion, as we have seen, p. 6, that the turtle and frog do not breathe alike, but that Avhile the latter forces air into the lungs, the former possesses a type of respiratory movement closely analogous to that of the mammal. He described an inspiratory muscle in the posterior flanks, and an expiratory muscle covering the back of each lung, and attached to a broad tendinous expan- sion, running forward, to be inserted in front on the carapace, above the lung. To do full justice to this most ingenious and neglected observer, we have quoted, in connection with the anatomy of our subject, the experiments, by means of which he proved that turtles do not force air into the lungs, p. 6, and by which he also showed that they draw the air into the chest, by muscles attached to the breast-box, and expel it through the aid of the expiratory muscle covering the posterior end of the lung. Considering the period at which he wrote, nothing could be clearer than the above statement, and we are amazed, that its obvious truth should have so long escaped recognition. In the summer of 1861, one of us, Dr. Weir Mitchell, while engaged in studying the blood-pressure in the snapping turtle, Chelydra serpentina, became convinced that the prevailing views as to the respiratory mechanism of Chelonian reptiles were totally incorrect. Accordingly he partially studied the subject, and incident- ally embodied his opinions in an essay upon the blood-pressure in the snapping turtle.2 At the time referred to, Dr. Mitchell was unacquainted with Townson's researches. The views of Dr. Mitchell, and the experiments by Avhich he supported them, Avill be found scattered through the text of the present essay, of which, indeed, they form the basis. In the summer of 1862, the present authors took up anew the study of the respiration in turtles, and have endeavored to render it as complete as possible. In so doing they have been fortunate enough to carry the subject far beyond the crude experiments of Townson, and to discover anatomical and physiolo- gical facts of the utmost interest and novelty, which have hitherto escaped attention. To facilitate the comprehension of the subject, we shall divide the physiological part of this essay in the following manner:— 1st. The externally visible phenomena of respiration. 2d. Physiology of the muscles of respiration. 3d. Physiology of the respiratory nerves. 1 Tracts and Observations in Natural History in Physiology. London, IT99. Cuvier's views and his criticism of Townson may be found appended to the full quotation of Townson's dissertation, at p. 6 of this essay. 8 American Phil. Trans., Phil. 1862. 24 A NATO M Y A N D PHYSIOLOGY O P When a turtle of any kind is observed with care, it Avill be seen that it breathes at Airy irregular intervals. These are much prolonged Avhen it is in the water, and half an hour or more may elapse before it rises to the top, to take two or three respirations, preparatory to a second plunge. When, during summer weather, the snapping turtle Avas placed on a table, and obscn'cd in air, its respiration averaged one to every tAvo minutes and a half, although certain individuals breathed more rarely, and all irregularly. The box turtle breathes still less frequently. A large1 snapper observed for some time, gave the folloAving record:—Ten respirations Avere noted Avith the intervals betAveen them, which were as follows:—1, 2, 1, \, 5, \, 1, 4, 3, 2, \ minutes respectively. In another the respirations during an hour avc re at almost pe-rfectly regular intervals of tAvo minutes. The size of the turtles did not seem to bear any notable relation to the number of respirations per minute. During the respiratory act in the snapping turtle, C. serpentina, the box turtle ('istudo Virginca, the green turtle Chel. mydas, and several Emydae, avc have noticed carefully tin* exact details of the motions of the various parts. The head and neck, the flank spaces in front of and behind the limbs, these themselves, and the mouth, glottis, and hyoid apparatus, have been scrutinized Avith care in hundreds of instances, and with these results. Turtles breathe easily Avith the mouth open or shut. This fact alone deprives their respiration of all resemblance to that of Batrachians. The respiratory process is threefold, and consists of— 1. Complete expiration. 2. Complete and very full inspiration. 3. An appearance of slight, or partial expiration, folloAved by a pause of greater or less duration. During the period which precedes this scries of movements, the turtle being at rest, the spaces betAveen the- posterior members and the plastron and carapace are nearly level, or only a little concave. The shoulders are pushed forward somcAvhat, the lungs being full at this time, Avhile the* large hyoid apparatus is usually dilated or draAvn backwards and doAvmvards. Sometimes it is in continual motion, like that of the frog when breathing, but in the turtle this rise and fall of the hyoid arches has no essential connection with that function. When, during the inter-respiratory pause, we open the jaws the same move- ments of the hyoid apparatus may still be seen, nor is it easy at these times to assign to them any very obvious purpose. The glottis may be seen at rest, as a linear slit, Fig. 7, A, in the centre of an ovoidal slightly elevated mound, just back of the tongue, on the floor of the mouth. The first respiratory act is one of expi- ration. Whether the mouth be opened for observation or not, the following move- ments occur: The hyoid apparatus descends and broadens laterally especially at its posterior part, carrying the glottis back and a little down. The object of this action avc suppose to be, the separation of the glottis from contact with the roof of the mouth, in order that the air may the more readily enter it after passing through the narcs. At the moment of beginning to expire the glottis opens Avide, so as to form a rhombic figure (Fig. 7, B.) It remains thus until the whole respiratory act is completed. Meamvhile, during expiration the limbs fall in towards the shell RESPIRATION IN THE CHELONIA. 25 quite passively, and the flank spaces in front of the posterior limbs sink so as to present deeply concave surfaces. Fig. 1. B Fig. 7, A. The glottis closed.—a a', the line formed by the glottic lips when the animal is not breathing; 6, the prominent central part of the glottic lips, indicating the summit of the arytenoid cartilage; c, tongue; d, lower jaw. Fig. 7, B. The appearance of the glottis during respiration.—a, right glottic lip; b, rima glottidis ; c, extremity of the hyoid bone. A full inspiration instantly follows. The flank spaces become flat and tense, rising to a level. The glottis remains open. The hyoid arches advance, and at the close of the inspiration the shoulders are pushed passively forward. As soon as the lungs are completely filled, a very slight expiration relieves them of the surplus air, the flank spaces sinking a little, the hyoid arch at rest, the glottis closing at the end of the expiration. The final action here described appears to be due to the cessation of activity on the part of the inspiratory muscles and to the passive falling in of the limbs displaced during their contraction. The lungs are thus left full of air, and ready for the next act of respiration. Whenever a turtle in air breathes, these triple actions occur, but when under water it occa- sionally expires air, and does not rise to renew the supply until some time has passed by. Type of respiration in Chelonia.—We are now prepared to examine the subject from another point of view. A superficial observer, or one who accepts the present belief, sees in the motions of the hyoid arches a movement in appearance corres- ponding to the respiratory play of the floor of the frog's mouth. Yet the slightest anatomical examination should have shown that, while in the frog the nostrils have valves essential to their mode of breathing, in the turtle there are none, while the form of the horny lips in the latter animal renders it impossible to make the mouth so air-tight as to act the part of a chamber in the supposed process of pumping air into the lungs. On the other hand, the laryngeal cavity is also too small to act as a chamber, nor does the hyoid arch, in its descent, enlarge the laryngeal area. When, at the beginning of this research, one of us observed the turtle (snapper) breathing with an open mouth, while watching a chance to bite, he Avas at once convinced that the agents of respiratory movement were beloAV the trachea, and the 4 2f, A N A T O M Y AND P II Y S I (> L 0 (J Y O F following verv simple experiments converted this conviction into the meist absolute certainty—a certainty which every future step served but to illustrate" from iicav points e>f a ie*w. On page 77 of the memoir of Dr. Weir Mitchell, previously cited, are to be found the experiments above alluded to. The trachea of a large snapping turtle was cut across, after Avhich breathing Avent on at the usual rate, or more often, oAving to causes presently to be mentioned. Next, a bent glass tube, tAvo millimetres in Avidth, Avas adapted to the upper or outer end of the divided trachea, and allowed to dip into Avater. If the breathing poAver resided in the hyoid arches, larynx, and mouth, the Avater in the tube should have been forced doAvmvards during inspiration, but, although respiration continued, the fluid moved at this time* only about one millimetre, and even this Avas plainly due to the motion of opening and closing the glottic lips, Avhich occurs synchronously with the respiratory movements in the breast-box. The same bent tube was next adapted securely to the loAver end of the divided trachea, and again dipped into water as before. At each subsequent inspiration the Avater Avas forcibly and largely draAvn up into the lung, and again rejected during expiration. After this no doubt could exist as to the locality in Avhieh arose the mechanical force productive of respiration. With this convincing proof the subject Avas laid aside for the future and more thorough investigation, of Avhich this essay is the record. Function of tlie respiratory muscles of the Turtle.—A large snapping turtle was secured on its back, and an incision made over the flank space, between the pos- terior limb and the plastron and carapace. The skin and superficial fascia were then carefully remove-el so as to expose the Avhole muscle which fills this space, and which has already been fully described. When inspiration took place, the muscle contracted, and as it is possessed of a central tendon from which radiate fibres in all directions, the result of their shortening was to comrert its previous deeply concave surface into one which Avas nearly level, while at the same time the air rushed through the open glottis into the lung. The analogy betAveen this muscle and the diaphragm of mammals was abso- lute*!}' perfect. The central tendon, the conAerging muscular fibres, and the form of movement resulting from this beautiful arrangement, all united to suggest the resemblance. The inspiratory functicn of this muscle was palpably evident, nor could any other office be possibly assigned to it, because it was attached to no movable bone or other parts susceptible of motion. Repeated gah-anization of this muscle served further to demonstrate its purpose. Finally, the muscles on both sides Avere removed, Avhen all inspiratory poAver was lost. The turtle could empty its lungs, but possessed no power to fill them anew. The muscles engaged in expiration Avere next made the subject of study. At first avc Avere led to believe*, that the elastic contractility of the lungs might alone suffice to empty them, but this Avas opposed to all physiological analogy, and the poAver Avith which expiration occurred Avas too great to allow us to suppose that no muscular force intenened for its production. To examine this part of the subject, a turtle (snapper) Avas secured, as usual, and RESPIRATION IN THE CHELONIA. 27 the plastron removed, Avith the exception of a rim at the back and on each side, to which remained attached the fibres of the inspiratory muscles. After a few minutes the turtle expired the* air in the lung. During this action, the fascia covering the lungs below, and lying betAveen the peritoneum and the plastron, was observed to be- come tense, owing to the contraction of the tAvo sheets of muscle, which terminate this tendon anteriorly and posteriorly, and find origin in the carapace. Recalling the full anatomical description already given, it will be remembered, that the lungs and abdominal viscera are covered outside of, and below the peri- toneal sac, by a white membranous tendon, which extends across the middle line, and is firmly attached to the pericardium, as well as by firm areolar tissue to the central line of the plastron or lower shell. The muscular bellies arising from this covering tendon, fold over the lung in front and behind. Opposite to the inspiratory muscles are also areolar fibres, binding its tendon to the fascia of the expiratory muscle above it. When the four bellies of this muscle, or muscles contract, the lungs are acted upon directly, or by being compressed through the medium of the other viscera which are, so to speak, grasped during this powerful movement. At the same time, the passive inspiratory muscles are draAvn up Avith the retreating lungs, OAving to the pressure of the external air, and to the close union betAveen the two sets of antagonistic muscles. Although the pericardium is also fastened to the expiratory tendon, this sac is so firmly bound to the plastron below it, that it does not appear to be disturbed during expiration, unless the connecting fibres are divided, in which case the heart sac and its contents are strongly drawn from the plastron, as the air is expired from the lung. As in the case of the inspiratory muscle, the expiratory muscle was also tested by observing its action when exposed in the living animal, and by galvanizing its fibres. The purpose of this singular sheet of muscle and connecting tendon admits then of no doubt. Aided by the elasticity of the lung, it empties that viscus of air, and no other muscle appears to lend it any aid. The third period of respiratory movement is marked by the closure of the glottis, and by the relaxation of the muscle of inspiration, the limbs then settling passively to their new positions. Hence the general appearance of a slight expiration at the end of the inspiratory act. It is impossible to review this account of the respiration in chelonians, without being struck with the simplicity of the plan. A box containing all the viscera of the chest and belly has an open space on each side, filled by a muscle of peculiar form, whose contraction increases the size of the visceral cavity, and thus causes air to rush into it. Within the breast-box, the lungs and visceral mass embraced by a single muscle, obey its contraction in effecting expiration, and as the visceral cavity thus becomes smaller, the inspiratory flank muscles curve in to fill the gap. After the most careful investigation, avc can discover no other respiratory muscles Avithin the breast-box. The muscular apparatus of the glottis is equally simple. There is a muscle to open it, and another muscle to close it. Here, as in the rest of this portion of our essay, we shall not commit ourselves by names, Avhich, although they may recognize homologies, confuse the reader, who has sometimes to bear in mind that their >N A X A T O M Y AND P IT Y S I <) I, O 0 Y () F functions may be exactly the reverse of those of the' human muscle wlmse* name they carry. The* tAvo glottic muse les have already been fully described; Avhen both are cut away e>r paraly/e*d, by seetiem of their nerves, the* glottis still closes, owing te> the elasticity of its cartilage's, but it docs not shut firmly, and if the* lungs be previously filled Avith air, a large part always escapes. Under ordinary circumstances, the* glottic lips are closely pressed together by the sphincter-like muscle which Ave* have* described and figured. The mass of its fibres lie beloAV the opening muscle, and are parallel to the direction of the glottic lip, while its connections are principally at the anterior and posterior end of the glottic line. When contracted, as it always is more or less strongly during the interval betAA'een tAvo respiratiems, it would tend te) pucker the glottis someAvhat, if it were not that the anterior and posterior insertion are firmly fixed, by the parts in front of and behind them respectively. Thus attached, the only influence it can exert, is to close the glottis Avhose lips stiffened by the arytenoid cartilages facilitate the process. The opening muscle lies outside of the closing muscle, nearly at right angles to it, and immediately under the mucous membrane of the glottic mound. At the moment Avhen expiration begins the respiratory act, this opening muscle contracts so as to draw the glottic lips Avide open and permit the air to escape. Then fol- lows a full inspiration, the glottis still open, and lastly it is closed by the constrictor muscle just after the great flank muscles of inspiration cease to act. The doAvnAvard movement of the hyoid arches is effected by the omo-hyoid and other muscles of the neck. It appears to be intended to remove the glottis from contact Avith the roof of the mouth during the act of respiration. The upAvard motion of the hyoid apparatus is produced by a thin sheet of muscular fibres spread transversely across it and over the Avhole upper part of the neck. The function of all of the above muscles was determined by simple observation, by stimulating them directly, and by irritating their nerves. The necessity for closing the glottis firmly in these animals becomes obvious, AArhen Ave reflect, that not only must they be enabled to retain the air, but Avhen under Avater be competent to exclude that fluid from the lungs. In fact, when we divide the trachea, or in any Avay paralyze the glottic muscles, the power of retain- ing air in the lungs is totally lost for a time. The moment the respiratory muscles cease to act, the elasticity of the lung asserts itself, and that viscus is immediately emptied. After a day or two, however, a curious change may be noticed; the turtle breathes as usual, but in place of allowing the air to escape through the open trachea, the animal holds the inspiratory muscle contracted, and thus retains the air in the lung a considerable time after each inspiration. There seems to be some urgent necessity for thus holding the air a long time in the lung, and perhaps for keeping the lung distended. The instinctive provision for these purposes when the usual means fail, is well Avorthy of note. As we proceed with the study of the laryngeal nerves, Ave shall have further occasion to observe the great importance of the glottis, and to Avonder at the singular means to which creative poAver has resorted, in order to secure the orifice from the ordinary chances of accident and disease. The physiology of the nerves of respiration in turtles has been the subject of RESPIRATION IN THE CHELONIA. 29 our most careful and complete study. So novel and surprising Avere some of its results that we have felt it right to surround ourselves with more than common precautions. For this purpose we have repeated our experiments and dissections on several species of turtles, and on numerous individuals of each species, until incessant repetition left no question unanswered, and no conclusion doubtful. We shall study, 1st. The physiology of the pneumogastric nerve and its branches, so far as they concern the respiratory function. 2d. The physiology of the nerves which supply the respiratory muscles of the breast-box. For all necessary details as to the anatomy of the vagus nerve and its branches we refer to the former part of this memoir. Here it will only be requisite to repeat that, as in most mammals, the larynx receives a superior laryngeal nerve, and an inferior or recurrent laryngeal trunk. The superior, Avhich in man is the nerve of sensation to the larynx, is in turtles distributed to the mucous membrane of that organ, and also to both of the glottic muscles. The recur- rent laryngeal, which in man is the principal motor nerve to the larynx and glottis, is in turtles also motor, but it sends branches only to the opening muscle. The remaining peculiarities will be better understood as we proceed to state in sequence the experiments which led to their discovery. Experiment.—A large turtle (snapper) was secured on its back, its mouth held open. It breathed well at intervals of two minutes or more. The recurrent nerves were exposed and galvanized at the middle third of the trachea. Irritation by this agent and by mechanical means, caused the lips of the glottis to open, although not very freely. The two nerves Avere then divided, and the trachea cut across. The glottic movements continued perfect, and Avere synchronous with the respiratory motions of the breast-box. The muscles of the right side over the hyoid apparatus were then removed, the covering fascia beneath them dissected off, and the superior laryngeal nerve discovered lying under the shelter of the superior hyoid wing. Irritation of this nerve or its felloAV on the opposite side caused the outer edge of the glottic lips to open, while the inner edge appeared to be forcibly closed at the same time. On cutting the nerves across, and stimulating the peripheral ends, like results were observed. The left superior laryngeal nerve being intact, galvanization of the centric end of the divided nerve on the right side caused first, closure of the inner lips and opening of the outer lips of the glottis; and second, violent and general muscular movements and winking, apparently expressive of acute pain. Finally the left superior laryngeal nerve was divided, when complete paralysis of the glottis ensued. Order of section, and results:— 1. Section of both inferior laryngeal nerves, causing glottis to open; glottic movements perfect after section. 2. Cut right superior laryngeal nerve, causing glottis to open superficially and to close below; galvanization of outer end of nerve caused same result; galvanization of centric end gave signs of sensibility and reflex closure of glottis, and opening of its outer lips. ;} move the gleittic lips on the* siele eut. AVe- next divided, in like manner, the* right superior laryngeal nerve*. The poAver to open the glottis remained but little impaired, but the air could no longer be retained in the lungs. Respiration went on as usual, but Avhen inspira- tion Avas complete and the muscles relaxed, the glottic lips fell together by virtue of their oaati elasticity, although this seemed insufficient to balance the contractile force of the expanded lung, AA'hose contents therefore escaped. Then folloAved re'iieAA'cd inspiratory efforts, necessitated by the loss of pemer to close the glottis, until the animal learned to hold the air in its lungs by keeping tense, for a time*, the flank muscles of inspiration. The left and right inferior laryngeal nerves having next been divided, entire paralysis of the glottis ensued, the flaccid lips falling together valve-like Avhen efforts Avere made to inhale air, Avhile, if air Avas bloAvn into the lungs, it escaped without difficulty. Order of section, and results: — ,, „. r , f. . , , f Glottic lips conAnlsed Sectum of left superior laryngeal. 1 \ [ by section. ., .. r . , , . , , f Loss of poAver to close Se'ctum of right superior larvngcal. 1 , . * ° l I glottis firmly. Section of both recurrent laryngeal. J Complete paralysis of glottis^ \ loss of poAver to open glottis. The aboArc experiments, repeated upAvards of tAvelve times on the Chrysemys pitta, the Cistuelo virginea, the Chelonia myelas, and the Chelydra serpentina, left no doubt in our minds as to the funetions of the tAvo laryngeal nerves in turtle's. Careful elisseetions enabled us moreover to trace these nerve's so as to show that, while the inferior laryngeal is distributed only to the opening muscle of the larynx, the* superior laryngeal sends branches to both the dilating and the constricting muscle's. This anatomical arrangement explained to us some of the difficulties which we had encountered while testing the function of the muscles by means of irritants applied to the nerves. Thus, Avhen the upper nerves Avere irritated, the glottis opened at the outer lip and closed Avithin, because the irritant necessarily acted both on the nerve fibres of the closing and of the opening muscles. Again, when the loAver nerve, inferior laryngeal, Avas galvanized, it caused the lips of the glottis to open, but not freely, because the motion of the lips seemed to act reflectively as a cause of irritation through the mucous branches of the superior laryngeal on to its nerve centres, and thence by its motor fibres upon the opponent closing muscles. When, however, the superior laryngeal nerA'es were cut, the closing poAver was abolished, and then, irritation of the inferior nerves produced more perfect dilata- tion of the glottic chink. We have thus determined by every nece*ssary means that the superior laryngeal nerA'es in turtles are the nerves of sensibility for the mucous membrane of the larynx and glottis. That they are the motor nerves of RESPIRATION IN THE CHELONIA. 31 all the true glottic muscles, and enjoy thus the ability to open and to close this orifice, and that the inferior laryngeal nerves are the motor nerves of the dilating muscles only, and have not sensibility or power to close the glottis. What then is the reason of this double distribution of tAvo nerves to one muscled Upon this question we shall presently return. It seems highly probable that both nerves usually act at once to open the glottis, since galvanization of either set of nerves does not fully effect this end, while, when both sets of nerves are stimulated, the glottis opens wide. The distribution and functions of the two laryngeal nerves in turtles are thus seen to be totally different from what we see in mammals. In them, as we need only to remind the reader, the superior laryngeal is a nerve of sensation chiefly, and although it possesses also a motor filament, this, in man at least, is distributed to a muscle, the crico-thyroid, which has neither homologue nor analogue in chelonian reptiles. In mammals the inferior laryngeal is, as in the turtle, a motor nerve, but it supplies alike the dilating and the closing muscles of the glottis. On reference to the anatomical part of this essay, it Avill be seen that the hypo- glossal nerve lies close to the track of the superior laryngeal neiwe, and might readily be confounded with it, Avhen the intention is to find and divide the latter alone. The nerve in question supplies muscular branches to the tongue only. Thus far the physiology of the glottic nerves in turtles, although determined for the first time, and shoAvn to present points of great interest and novelty, has not exhibited any peculiarity so exceptional as that to which we shall now direct attention. This was brought to our notice while further pursuing the study of the functions of the glottic nerves. The mode in which it was first suspected, then discovered, and finally set in clear light by every available means, will be best set forth in the following record of our experiments and inferences, in the order in Avhich they occurred. Experiment.—A small snapper, one and a half pounds in weight, Avas secured as usual. Its respiratory acts observed to be perfect, and the two inferior laryngeal nerves divided one after the other, causing twitching of the glottic lips. After this the glottis still opened and shut as before, and, indeed, equally as well. It was plain, as we have already seen, that the superior laryngeal nerves could open and shut the glottis without other aid. Next, the right superior laryngeal nerve Avas cut at the middle of the upper hyoid cornu, and the glottis was carefully obsenred. The section caused tAvitching of the glottic lip, and at the next respiration, to our great surprise, both sides of the glottis, the right as well as the left, opened equally well. In fact there was no difference. A close inspection satisfied us that the section of the nerve Avas complete. If now we recall the facts, that the glottis of both sides Avas moving despite the section of both recurrents and one superior laryngeal nerve, it Avill be seen how mvsterious this must have appeared to those Avho first observed it. We came to the conclusion e'ither that there existed some mechanical arrangement of the glottis and its muscles, which enabled one side, Avhile in motion, to communicate' that movement to the other, or, that there Avas a direct nerve communication betAveen 32 ANATOMY AND PHYSIOLOGY OF the right and left superior nerves e*f the larynx. The first hypothesis Avas unsup- ported by anything that Ave* knew of the parts. The second seemed unlikely, sinee on reflection avc could recall no instance of a true chiasm of any nerves except those* of sight. We hastened to examine the question by neAV experiments. Expkriment.—Snapper, Aveight tAAro pounds. We exposed and galvanized the* left inferior laryngeal nerve, thus causing both lips e>f the glottis te> open. The* same result Avas obtained Avith the right nerve. This fact, observed by us in other cases, was soon fenind to be due to the difficulty of insulating the current in one nerve*. Whein, hoAvever, Ave made use of mechanical irritants, stimulation of one nerve affected only the glottic lips of the same side. The right inferior laryngeal nerve Avas then cut, and immediately aftenvards the right superior laryngeal nerve. The glottis still moved as Avell as before these sections. Next, avc cut the left recurrent (inferior laryngeal nerve), thus leaving the left superior laryngeal the only nerve entire. Nevertheless, the glottic lips on both sides opened and shut, as Avell and as completely as ever. Lastly, avc cut this remaining nerve, causing total paralysis of the glottis, and the usual results as to respiration. Order of section, and results :— 1st. Cut right recurrent nerve (inferior laryngeal) and rig t superior laryngeal nerve; glottis continues to move perfectly on both sides. 2d. Cut left recurrent (left inferior laryngeal); glottic action perfect on both sides. 3d. Cut left superior laryngeal nerve; total paralysis of glottis. Experiment.—Snapping turtle, Aveight three and a half pounds. We dissected the' hyoid apparatus from its connection with the loAver jaAv, and held it back, thus freely exposing to vieAV the chink of the glottis. Up to this time avc had reached „ the conclusion, that someAvhere on the fenestrum in the cricoiel cartilage there might be a branch of communication betAveen the two superior laryngeal nerves of the larynx. Therefore, on the turtle prepared as above described, avc made an incision on to the fenestral membrane, betAveen the larynx and the hyoid bone, opposite to the junction of the superior cornu Avith this bone. The section made a little to the left of the median line caused slight tAvitching in the glottic muscles, but had no influence on the respiratory motions of the glottis. The tAvo inferior laryngeal nerves were next divided, and still the glottis moved as perfectly as before. The left superior laryngeal nerve Avas divided at the middle of the upper hyoid cornu, and immediately all motion of the left side of the glottis ceased, the right side moving during respiration as usual, although somewhat feebly, OAving perhaps to loss of blood during the first part of the experiment. Section of the right superior laryngeal nerve completed the paralysis of the glottis. Order of section, and results:— 1st. Section through supposed site of communicating nerve; no effect as to respiratory movements. 2d. Section of both inferior laryngeal nerves ; no further effect of any permanent nature. 3d. Section of left superior laryngeal nen-e; paralysis of left glottic lip. 4th. Section of right superior laryngeal nerve; complete paralysis of glottis RESPIRATION IN THE CHELONIA. ',]:} The above experiments led us, irresistibly, to the conclusion, that there must be a chiasm of the tAvo superior laryngeal nerves, and it only remained to prove, Avith the scalpel, the presence of this branch. A careful series of dissections on large turtles of various species and genera, satisfactorily proved that Ave Avere not mis- taken. In every case the nerve was readily found, and the physiological prediction as to its existence verified in the most absolute manner. The discovery of a new nerve in turtles, and upon ground over which the accurate knife of Bojanus had passed, called for a still more rigorous testing of our previous results. For this purpose the following experiments were made. The first of this second series is of unusual value, OAving to circumstances which arose incidentally. Experiment.—Snapping turtle, Aveight nineteen and three-quarter pounds. We cut down on the middle line of the hyoid bone and divided it throughout its length Avith a hair-saw and nippers. When this operation is done with care, it exposes to the operator enough of the cricoid fenestrum to enable him to cut the communicating nerve at its central part. Next, both recurrent nerves Avere divided at the middle of their course. The section, and after stimulation of the right nerve, had no effect on the glottis, Avhich we thought singular. Section of the right superior laryngeal nerve was satisfactorily made as usual, the nerve being readily exposed and divided. To our surprise, the right glottic lip became paralyzed almost totally, the left side moving in respiration as usual. This result was opposed to all our former experiments. After a rigid examination of the conditions of this latt experiment, and finding in them no explanation of the contradiction Avhich it offered, Ave dissected, with scrupulous care, the whole track of the pneumogastric nerve and its branches to the larynx, as well as that organ itself. The following appear- ances were noted: On the left mucous lip of the glottis, a small white patch of diseased tissue. The inner end of the right upper hyoid cornu Avas enlarged to double its normal size; thus of necessity stretching the right superior laryngeal nerve where it crosses the cornu at its inner end. On the left side the superior laryngeal nerve was perfect up to the point at which it gave off the interlateral communicating branch. This latter nerve, lying on the cricoid fenestrum, was involved in a mass of diseased tissue, which extended between the trachea and the body of the hyoid bone, from its lower part to a point about one-quarter of an inch above the fenestrum. This disease, doubtless, affected the communicating branch, so as to cause partial paralysis of the right glottic lip to follow section of the corresponding superior laryngeal nerve. Had the interlateral branch been completely destroyed, section of one laryngeal nerve must have produced entire paralysis of the glottic lip on the side operated upon. This observation, which at first promised to cast doubt upon those Avhich preceded it, thus proved at last the most conclusive evidence of the correctness of the view to which we had arrived. An accident of disease or injury had so altered the communication between the two superior nerves of the larynx, as to make unneces- sary the section, which Avould under ordinary circumstances have folloAved as the third step in the experiment, 5 ;it ANATOMY AND P II Y S I 0 L O (i Y OF Exjtrr intent.—This experiment AA'as designed to be a repetition of the plan of the* last one, but in dividing the* hyoid bone to reach the nerve* at the' middle line, the saw, accidentally carried too deep, touched the membrane on Avhich runs the nerve. Section of the recurrents folloAved with the usual negative result, Section of the right superior laryngeal nerve produced paralysis in the right glottic lip. If emr former vieAv be correct, then in the present case Ave must June cut the communi- cating branch Avith the saAV. In the above experiments, the sections and results may be thus stated : — 1. Section of interlateral communication betAveen the tAvo superior laryngeal nerves; glottic respiratory motions as usual. 2. Seetion of both inferior laryngeal nerves; glottic respiratory motions as usual. 3. Sectiem. of right superior laryngeal nen'c ; paralysis of right lip e>f glottis. Experiment.—Snapper, Aveight four pounds. We cut first the tAvo inferior laryn- geals; next Ave* divided the right superior laryngeal. The glottic- movements avc re still perfect. One nerve Avas sustaining unimpaired the Avholc ordinary moderns of the glottis in respiration. Indeed, the closest scrutiny failed to discover in its action any departure from the condition of health. Lastly, avc sawoel through the hyoiel bone, glottic acts still regular. Then Avith a hook avc lifte'd the nerve and divided it. Instantly a respiration folloAved, but the right glottic lip was iioav motionless. Ord< r of section, and results:— 1. Section of both inferior laryngeal nerves. 2. Section of right superior laryngeal nerve; after which the glottis moved in respiration as usual. 3. Seetion of median intercommunicating nerve; paralysis of right glottic lip. Ex per intent.—This turtle had been used for other purposes, and had undergone an hour before section of the middle cervical spine. The' respiratory motions of the breast-box had ceaseel, but at intervals the glottis opened and closed Avith normal regularity. The trachea was divided, and Avith it both recurrent laryngeal nerves. Next avc cut the interlateral communicating nerve*. The glottic acts still remained perfect. Lastly, we exposed the left superior laryngeal nerve, and divided it, causing instant paralysis of the left glottic lip. Order of section, and results:— 1. Section of both recurrent laryngeal nerves. 2. Section of communicating branch; glottic acts perfect. 3. Section of left superior laryngeal nerve; paralysis of left glottic lip. As further illustration, Ave give in brief the order of section and results in tAvo box-turtles. Experiment. 1. Section of both inferior laryngeal nerves; glottic motion perfect. 2. Seetion of right superior laryngeal nerve*; glottic motion perfect. 3. Section of communicating nerve ; paralysis of right lip of glottis. 4. Section of left superior laryngeal nerve ; total paralysis of glottis. Experiment. 1. Section of communicating nerve. 2. Section of right superior larynge>al; glottic acts perfeet. perhaps not closing RESPIRATION IX THE CHELONIA. 35 firmly on the right side; the right glottic lip noAV relied alone on the recurrent nerve for opening poAver. 3. Section of right recurrent (inferior laryngeal nerve); paralysis of right glottic lip. The above stated experiments Avere repeated very frequently, and always Avith the like results. If the evidence which Ave have given be reliable, we have now proved that in turtles there exists a communication between the right and left superior laryngeal nerves, of the nature of a true chiasm precisely like that of the optic nerves, and, so far as we know, the only instance thus far discovered of this anatomical peculiarity in nerves exterior to the great centres. Fig. 8. Fig. 8. Diagram of the chiasm of the superior laryngeal nerves.—a a', intercommunicating fibres of the right nerve; b b', similar fibres from the left nerve. The diagram, Fig. 8, illustrates our vieAvs in regard to the track of the nerve fibres. Part of each nerve probably proceeds directly to the two glottic muscles of its OAvn side, while another strand crosses over through the interlateral trunk to be similarly distributed to the two muscles of the opposite side. Keeping this in view, we can now see how one single superior laryngeal nerve may move the glottis on both sides, until the chiasm is divided, when it will be left in connection only Avith the muscles on its OAvn side of the glottis. Having thus established the fact of a chiasm betAveen the superior glottic nerves, it was requisite to ascertain whether the inferior or recurrent laryngeal nerves entered into communication with the superior nerves, or whether they possessed any similar interlateral connection of their own. Experiment.—Snapper, weight six pounds. We divided first the right and left superior laryngeal nerves. The glottis opened as usual, but had lost its power to close firmly. Section of the right recurrent which folloAved, as the next step, produced paraly- sis of the right glottic lip. Galvanization of one recurrent caused opening of only the corresponding lip of the glottis. Repetitions of the above experiment led to no different result. Order of section, and results:— 1. Section of both superior laryngeal nerves; loss of closing poAver. :u; A N A TO M Y A N P P HYSI (> L O G Y O F ■J. Section of right inferior laryngeal nerve; less of opening poAver in right lip of gle>ttis. We* inferred from the above stated experiment and the* repetitions of it, that no interlateral nerve* fibres connected the two inferior laryngeal nerves. Further- more, avc failed to discover any branch to which such a function could have been assigned. The object of the very extraordinary and really exceptional arrangements, which we- have* here pointed out, is not altogether clear. AVe arrive only at the general conclusion, that the integrity of the* glottic function in turtles, appears to have been guarded with unusual care. Why this should be the case in aquatic chelo- nians it is easy to understand, but the necessity for it in terrestrial species seems to us less obvious, yet it is as perfect in the box turtle as in the emydtc and ehelonura*. Perhaps the need for such precautions in all may be due to the fact that all re-tain the* inspired air during long periods, even A\he*n on land. Paraly- sis of the closing poAver of the glottis would allow the air to escape instantly, and would oblige the* animal to make repeated and therefore laborious inspiratory efforts. Paralysis of the opening power would insure death from apneca. Hence we have tAvo sets of nerves controlling the opening muscles. One entire sot may be* destroyed and yet respiration continue. Lven one of those remaining, if these be the upper nerves, may be lost, and still the* glottis fulfil its entire duty in the train of breathing movements. Thus, also, in regard to the closing poAver. The elasticity of the glottic lips is one agent, although but a subsidiary one. Then avc have the interlateral communication betAveen the two superior laryngeal nerves, which alone can forcibly close the chink of the glottis. By virtue of this true chiasm one of these nerves being injured, the other is ample to effect the normal purpose* of be>th. Nor is it less curious to obsen-e hoAV artfully the whole apparatus has been guarded against accident. The loAver or re'current laryngeal nerves lie alongside of the trachea, sheltered by its projecting form. The superior nerves are protected in their course by the supe- rior hvoid cornu, and the larynx and its singular nervous circle are deeply buried beneath, or rather above the' strong bony and cartilaginous body of the hyoid bone. Nature seems to have been lavish of expedients for securing the safety of these most important parts. Before leaving this portion of our subject, it may not be amiss to state that avc have made a number of experiments on birds and mammals, to ascertain whether anv such chiasm exists in the glottic nerves of these animals. But in all cases section of one motor nerve caused loss of movement in its OAvn side of the larynx, and Ave therefore conclude that this arrangement does not extend to the classes in epustion. Whether or not it is to be found in Batrachia and ophidian reptiles, we have not as yet ascertained. The remaining physiology of the pneumogastric nerve in turtles is not less obscure than in other animals. As in these latter, so in turtles, it sends branches to the trachea, lungs and heart. We have cut the nerve in a number of turtles, some of whom survived upwards RESPIRATION IN THE CHELONIA. 37 of a month and then exhibited no marked evidence of diseased lungs. In others, there Avas occasionally found an abscess at the base of the neck. This pathological occurrence is, hoAvever, a common one in turtles caught with the hook, and cannot, with any probability, be supposed to be due to the section of the pneumogastric. The only striking effect of this section was, the constant sensibility which the nerve then exhibited. At the moment of dividing or crushing it, the animal shoAved every possible evidence of acute pain. Irritation of the centric end of the cut nerve gave rise to like phenomena, while stimulation of the peripheral end caused no such results. A number of careful experiments were made to ascertain whether these irrita- tions of the nerve produced any instant effect, either upon the inspiratory or expiratory muscles of the breast-box. But in no case did the stimulation seem to influence them to movement. Galvanization of the pneumogastric nerve in turtles arrests the heart's move- ments. Gentle irritation of the trunk causes the heart to beat more rapidly. Sec- tion of one nerve causes the heart to quicken its pulsations. Division of both nerves induces still more rapid action, but in either case the heart, after a feAV hours, regains its original rate of pulsation. The nerves Avhich supply motor endowments to the internal respiratory muscles need no special illustration here. They are fully described in the anatomical sec- tion of this essay. It only remains to add, that their office and relation to the muscles was tested by stimulating them with galvanism and by dividing them, so as to cause paralysis of the muscles in question. The centre, to which proceed impressions, giving rise therein to respiratory impulses, appears to be, as in other animals, the medulla oblongata. The site of the respiratory ganglions would scarcely have attracted our attention, hoAvever, had it not been, that, in the following experiment, a fact was noticed which induced us to examine the question more fully. Experiment.—In a turtle previously used to examine the offices of the laryngeal nerves, and in AArhom the glottis could still open on one side, we divided the cer- vical spine at its upper third, and continued to watch the respiratory muscles. To our surprise the flank muscles acted at intervals for thirty minutes, but the tAvo sides no longer moved synchronously. At one moment the right muscle contracted, at another the left, and the movements of both were irregular and some- times incomplete. It appeared to us, that these motions after section of the spine might be merely the rhythmic repetition of habitual movements, such as, according to Brown- Sequard, appear sometimes in the diaphragms of mammals even. Long after these muscles in the turtle ceased to move, all the other reflex acts continued, and excepting these, almost every muscle beloAV the point of section could be excited easily to reflex motion; neither was there any longer a synchronism of action betAveen the respiratory muscles of the glottis and those of the breast-box. Experiment.—Turtle, weight six pounds. In this case, also, the cervical spine was divided, but although the reflex activity of most of the parts beloAV the section was remarkable, the respiratory muscles alone failed to respond to excitation of distant 38 A XATOMY AND PHYSIOLOGY OF parts. During the spasm caused by the section of the spine, the expiratory mus- cles, contracting, emptied the* lungs, which Avere not again filled Avith air. Exjxrimtut.— Turtle, Aveight 4 pounds. The svmpathetic nerves on both sieles. in this turtle, had been cut seAeral Ave-e-ks, and the Avounds in the neck Avere nearh healed. The animal seemed Avell and very actiAe. The cervical spine* Avas divieh'd Avith little loss of blood. General spasm ensued, the glottis opened, expiratiem folloAved, but no after inspiration, and the glottis closed. During an hour no inspi- ration occurred, although the glottis opened and shut at about the usual respiratory intervals. To make more sure of this, the trachea Avas cut across, the lung fully inflated, and a tube secured in the loAver end of the trachea. Through a slmrt caemtchouc tube the trachea Avas thus conne*cte'd with Poiscuillc's ha*maelynamomete*r, fille'el to its 0° Avith mercury; on turning a ste>pcock the column rose about tAvo milli- metres, the glottis continuing in repose. Then the glottis opened, but no synchronous contraction of the lung muscles took place; indeed, the slightest must have been indicated instantly by the mercurial column. During frequent repetitions of glottic motion, no correspondent activity AAas at any time exhibited by the respiratory muscles of the breast-box. It folloAvs, therefore, that Avhile the flank respiratory muscles may after separation from their nerve centres move for a time, as do other habitually rhythmical muscles like the heart, that these motions do not occur in all cases, and that they arc plainly not dependent on a respiratory centre below the line of spinal section. The re'gular move*ments of the glottis Ave*re, as Ave avc have shoAvn, uninterrupted by the section of the cervical spine. The e|ucstion arose as to the exeiting cause of these motions. That they Ave re not due to impulses propagated through the main trunks of the pneumogastric nerves, Avas shoAvn by their continuance after the successiA'e division of these tAvo nerves beloAV the origin of the glottic nerves. It thus became plain that the medulla must receive its excitations from the head alone, perhaps through the fifth pair of nerves, which acted as afferent trunks, the motor nerves of the larynx completing the nervous circle as efferent branches. Hence the continued action of the glottis after division of the cervical spine. The principal points in the foregoing paper to Avhich Ave desire to draAV attention as novelties are as folloAvs:— 1st. In Chelonians the superior laryngeal nerve is distributed both to the opening and closing muscles of the glottis. 2d. The inferior laryngeal nerve is distributed solely to the opening muscle of the glottis. 3d. A true chiasm exists betAveen the tAvo superior laryngeal nerves. 4th. The expiratory muscle lies Avithin the breast-box, and consists of anterior and posterior bellies connected by a strong tendon continuous across the middle line, and common to both sides of the animal. 5th. The inspiratory muscles occupy the flank spaces on either side. 6th. Inspiration is effected by the contraction of the flank muscles, Avhich in appearance strongly resemble the diaphragms of superior animals. 7 th. Expiration is effected by the consentaneous action of the four muscular bellies above described, Avhich thus compress the viscera against the lungs. The RESPIRATION IN THE CHELONIA. 39 act of respiration consists of an expiration and an inspiration, during Avhich the glottis remains open. 8th. The opening of the glottis is effected through the agency of the superior and inferior laryngeal nerves, both of which are distributed to the dilating muscle of the glottis. The superior laryngeal nerve presides over the closure of the glottis, being in part distributed to its sphincter muscle. The elastic contractility of the glottic cartilages aids in closing this orifice. After section of the superior laryngeal nerves, the glottis may still be opened by the agency of the inferior laryngeal nerves, its imperfect closure being then effected by means of the elasticity of its cartilaginous lips. The chiasm of the superior laryngeal nerves enables one of these nerves to open and shut the glottis after section or disease of the opposite nerve and of both inferior laryngeals. Physiologists have therefore been in error when describing the respiration of Chelonians as analogous to that of Batrachians, since it far more closely resembles the breathing of the higher vertebrates. A P T E N D 1 X. Since committing to the press the preceding paper, Ave have had the opportunity of examining the respiratory apparatus of one of the Trionychidse.1 The striking characters of this family and its border rank amongst fresh-Avater turtles, render a knowledge of its respiratory structure of peculiar interest, and constitute our apology for this appendix. Amyda mutica, Fitz. —The general plan of arrangement of the respiratory muscles is the same as heretofore described, the inspiratory muscles occupying the flank spaces, and the expiratory muscle being attached to the dorsal shield and inclosing the viscera. The origin of each inspiratory muscle, in detail, is as follows: From the bony edge of the plastron, from the carapace at the line of termination of the ribs, from the fascia lata, where the thigh bounds the flank space, from the spinous process of the ilium and thence to the place of beginning on the plastron. The central tendon into which these fibres are inserted, is a mere raphae posteriorly, but widens anteriorly into a lance-shaped extremity, one-fourth of an inch wide at its widest part. The whole muscle is relatively large compared with that of other species. The expiratory muscle presents greater variation in origin and form than the inspi- ratory muscle; the fibres are longer than in other turtles examined, and the ante- rior and posterior bellies broader. These bellies meet at their outer margins and leave no space under the bridge of the plastron Avhere the muscular fringe is absent Avhen the muscle is vieAved from beloAV, as exists in the snapper, Fig. 3. The effect of this widened muscular margin is to diminish the size of the common tendon, and reveal its true character more strikingly than in other families. On dividing the tendon and removing the viscera, the muscular fibres are observed spreading out across the dorsal shield in fan-like radii from each intercostal space, from the first to the sixth inclusive. The third intercostal space gives attachment to the strongest fibres, causing it to appear as the centre from which the whole muscle radiates, the fibres running forward and outward across the area of a quarter circle constituting the anterior belly, and those running backwards over a similar area, the posterior belly. More precisely, hoAvever, the origin of the anterior belly is from the first and second intercostal spaces and from the costal margin of the third space; 1 For living specimens of Amyda mutica, we are indebted to Mr. Robert L. Walker, of Allegheny County, Pennsylvania. 6 . ( 41 ) 12 APPENDIX. from these points the fibres extend in a prevailing direction forwards over the anterior quarter shell towards its periphery, and then, arching around the viscera, arc n.M-rted into the central tendon. At the line where the muscle leaves the shell it receives additional fibres. The posterior belly arises from the sixth fifth, and fourth intercostal spaces, and from the vertebral margm of the third smc J (overlying the fibres of the anterior beUy that come from this space), and stretches »vcr the posterior quarter circle of the shell, to be inserted into the common tendon It receives reinforcements of fibres Avhere it leaves the shell to inclose the viscera' as does the anterior belly. ' PUBLISHED BY THE SMITHSONIAN INSTITUTION, WASHINGTON, D. C. APRIL, 1 803.