lb87.1 O N THE LARYNX OF THE ANUROUS A M P H I M A . 391

teeth and the head two froutal tubercles, which, howelcr, are much sinaller than in any other species of the group, and folloaed behind by a broad hut shallow frontal cavity. The clypeus is pointed and recurved ; the mandibles without trace of tooth OII their outer edge. The thorax is as broad as the elytra, broadly rounded on the sides, slightly narroaed to the base, with acute hind angles ; the surface free from furrows and strongly and sparsely pnictured, the punc- tures on the disk larger and transverse-oval or oblong ; on each side of tlie auterior disk is R small rouud fovea. The rows of small cir- cular punctures on tlie elytra are obliterated 011 the sides posteriorly and on approaching the apical callus. The hind tibiae have no trace of spine on their outer edge (only a few short setse), but there is a iong spine a t their outer apex. The species tlierefore is intermediate betaeen Burmeister’s subgroups a and b . The basal joint of the hind tarsi is not prolonged into a spine. The pygidium is opaque and purictured ; tlie apical ventral segment very finely punctnred, the rest of‘ the abdomeu (except the extreme sides) being smootli. The metasternum is punctured on the sides and clothed with reddish hair.

4. 0 1 1 a hi ther to uiirccogtiized Fea turc iii tlie Larynx of tlic h iurous Ampliibia. By G. 13. Howus, P.Z.S., P.L.S., Assist. Prof. of Zoology, Noririal School of Scieuce and It. School of Miiics, S. Kensingtou.

[Receileil Jiuie 2, 1887.1

’l’hc general structure of the respiratory organs in the Amphibia was first carefully analyzed by Henle, nearly fifty years ago’. H e showed that well-developed laryngeal cartilages are nearly a1 ways preseiit, and that the trachea and bronchi, though subject to con- siderable variation, may be supported by cartilaginous elements also. Conspicuous among recent iuvestigators in the same field is Wie- dersheim. IIe has shown that in the Gymnophiona, as in some Urodeles (e. g. Siren and Bmphiirrna), trachea and bronchi are well differentiated, and that fully formed cartilaginous rings map be developed iii connection therewith. To him we are also iudebted for a description of the larynx of Rana errculenta in ail its details ’, and for a number of other observations upon the subject geiierally ; while he has summed up our knowledge of this in his Text-book of Vertebrate Snstomy. Dubois is now engaged upon a searching investigation into the rnorrhology of the larynx. H e calls attention in his preliniinary notes to the preseuce, among other things, of

Cy: also his Lehrbuch ‘ Vergleichendo nnatoniische Ceschreibung dea Kelilkopfs : ’ Leipzig, 1839.

a ‘ Uio Annt,omie der Gyrnnophioneii : ’ Jenn, 1879.

:I Origindly in his ‘ Lclirbacli,’ edit. i. vol. ii. 1882. 111). (i10-645. “ Ziu. M(11ylio1trgic dcs Lwyiix,” Atrat. dnzciger, vol. i. 1S8G.

d. vergleicli. Annturriie d Wirboltliiore.’

Seeelao Vau Ueiutrielen, Xoulog. Auzeiger, vol. x, 1887, p. 91.

492 PROF. G . B . HOWKS O N AN U X R E C O G N I Z E D FEATURE [June 7,

median procricoid (interarptenoid) eleriients in the Amphibia-an interesting feature of cornparison with the higher types.

The Amphibian lar.ynx is, like that of Reptiles, chiefly remarkable for the absence of a distiuct thyroid cartilage ; but the above r6sumh shows tliat with this exception there are represented in the respi- ratory organs of the Amphibin, as a group, the leading constituerits of those of the liiglier Amniota. The epiglottis would, however, appear to be absent ; and it is the object of this paper to inquire how far this is really the case.

I t is necessary to point out that IIenle wrote’ of ‘Engystonm’ that “ die Constrictoren des Stimmladetieingangs siiid bei clieseii Species nur sehr lose an die Stirriinlade befestigt, sie liege11 in einer Querfnlte der Schleirnhant hinter der Zunge und dem Eingang der Stininilade, und diese Falte bedeckt, wie eiue Epiglottis, der vorder- sten Theil des Eingaiigs der Stimrrilade.”

My attention was first dtawn to this subject two years ago, while dissecting a male of the Grass-Frog ( B . temporariu). In that speci- men (fig. 1 a)’ the frorit wall of the laryux was prolonged forwards

The larynx in 12uwn. 1 b. IZ. temporaria, 3 .

1. R. esculentn, young $.

~ g , epiglottis ; La, aditus laryirgis.

into two papillate folds (eg.), which were tumid and connected together by a thin film, the whole constituting a forward prolongation of the laryngeal mucous membrane apparently superadded to that whicli is customary. Wiedersheim, in describing the larynx of R. esculenta, points out tliat it lies immediately behiiid a deep depression of the ~ I U C O U S membrane wliicli corresponds with tlie

1 n. R. fernporariffi, $. 1 c. A’. te?iiporciria, ? . All maguiEed three timcs.

L. c. 11. 28. Origiiislly lig~ired in ‘ Atlas ol‘ Elc~n. Biology,’ 131. i. lig. 13 (16%). ‘ I ) iohu tumle dea Frodice,’ Eclicr auil Wiederdieitu, pt. 3, p. 8 (Eriuiswioli,

1 W ) .

1387.1 IN T H E LARYNX OF T H E ANUROUS AMPHIBIB. 493

hinder tiorder of the body of the hyoid. This was so in the specimen to which I have just referred, and the anterior prolongation described overhung this depression, in a rnanuer strikiiigly suggestive of the epiglottis.

€1011 and Royer and Bambecke a have most recently studied the anatomy of the mouth in the Anurous Amphibia : the first-named author deals chiefly with histological details in 22. temporaria ; tlie last-named deal with the subject in general; but 1 fail to find rrieution, in their writings, of those facts with which I am concerned. I am satkfied that the structures described abose may or may not be prcsent in individual examples of the conimoii Frog, and have found, to my surprise, that the f'ree anterior extremity of the larynx is siiljject to no incousiderable amount of variatiou iu it, to say nothing of the Anura as a group.

It' tlie l i l~s of the laryngeal aditus be exairiined with care in B. lemnpruria tliere will generally be found at its aiitrrior end folds itlciitical with those here figured, but more or less marked. They are sometimes so small that there is little wonder they should have been so long overlooked. They are well dirfereutiated from the rest of the larynx ; of a yellowish colour in life ai;d soft iind fleshy, pro- jecting fieely beyoud those parts which are supported iii cartilage (cf. figs. 1 , 1 (I, pg). There generiilly passes betweeii them a thin transverPe fold of mucous membrane, and occasionally, when very niiiiute, they are, together with the same, erected and closely applied to the front face of the larynx (figs. 1 6 and 3 ) . I was for some time disposed tb think that they might be peculiar to the males; but that this is not the case the larynx of an adult female, repre- sented iii fig. 1 c, will show. That specimen is furtlirr remarkable, among the larynges of a number of females which I have examined, for the fact that the folds were continued along the sides of the aditus, the anterior half of that being thus embraced by a hood- shaped lip.

On examining other members of the group, it early became obvious that tlie structure with which we here have to deal was by no means exceptional. I n Leptodactylus peiitadactylus and the Bull-Frog ( R . pipiens) , for example (figs. 2 and 3) , two papillate elevations were found to be present ; these were iu both cases small and erected, and united by a transverse fold as iu the first-named exmple. Compa- rison of figs. 1 b and 2 reveal8 an absolute identity hetween indivi- duals of R. temnyoraria and Leptoductylus pentadactylus.

111 two of the abo\e-named species 1 found, in addition to the fore- going, a couple of other folds which were related to the hind half of the aditus (ep, figs. 2 and 3). I n tlie Bull-Frog (fig. 3 ) they passed insensibly into the mucous membrane posterior to the larynx ; but in Leptodactylus (fig. 2) they united behind so as to forrii an insignificaiit lip which embraced the liiiid boundary of the aditus, milch as did the supposed epiglottis its front one. I have not seen

Sitziingcih. Wien. Akacl., Jan. 1887. " Sur 1es ear:act. l'ournis par 111 bouuhe des t5tards des Batraciens auoures

d'Europe," Bullel. SOC. 2001. d. France, 1881, p. 75. PKOC. ZOOL. Soc.-1887, No. X X x I I I . 3 3

494 PROF. G . B. HOWES ON AN UNRECOGNIZED FEATURE [June 7

these folds in either the Grass or Edible Frogs. I propose to term them, in accordance with their position, epi/aryngeaZ,folds.

Firstly, as to the epilaryngeal folds. Their condition, as seen in the Bull-Frog, might conceivably be such as would have resulted from puckering of a loose membrane under muscular contraction ; while their sym-

At this stage two difficulties presented themselves.

The larynx in Ram yz'pieics, 0 , top view. Magnifid twice. ey, epiglottis ; rp, epilsrgngeal folds ; I a, aditus larjngis ; t, tongue ; a?, ceso-

pbsgus.

metry and union in Leptoductylus pointed, no less distinctly, to the conclusion that they were definite and permanent structures. Doubt was dispelled on examination of other genera, for in a male of Cera- tophrys umericana (fig. 4 ) the two folds not only united behiud but gave rise to a clearly differentiated overhanging lip, ep., such as could only have been a permanent structure.

Secondly, as to the anterior folds, eg. As already stated, I at first took these to represent the epiglottis of the higher Amniota ; hut

1587.1 IN THE LARYNX OF THE ANUROUS AMPHIBIA. 495

their paired nature, which is most conspicuous throughout, remained for some tirne a stumbliiig-block in the way of that iiiterpretation. His, however, has shown that the human epiglottis and plicze ary- epiglottic= are formed from a primarily paired structure. H e describes them as arising within what he terms the ' mesobranchial area,' from the niodification of a couple of conjoined papillate fold3 or ' f u r c ~ l a . ~ I n other words, the human epiglottis is, according to him, a strictly bilateriilly symmetrical structure. Comparison of his desription and figures (see especially fig. 44, p. G6, 1. c.) with those here given shows that there is no important difference, either in position or origiii, between the developing hiiman epiglottis and the epiglottidean folds of the Anura; but i t also introduces a fresh difficulty, on ticcount of the presence in the latter of a couple of tumid

Fig. 4. The larynx in C ' m u ' n p h y s umcricai2a, 8. Pig. 5. The mine in Cal~ptocqhakis ynyi, 8.

Magnified three times. D1;tguiGed t,wice.

References as in figs. 2 and 3.

folds below those which I have thus far described (* figs. 1 c and 3 ) . These might conceivably reiresent tlie furcula of His. They are, however, exceedingly inconstant, and in numerous instances unrecog- nizalile ; fig. 3 represents their maximum development observed. I n one instance I noted ( R . temporaria, 9 , fig. 1 c) that they were asymmetrical, that of the left side being much the snialler of the two '; and this a t once suggested that they might be accidental and due to displacement. From cnrrfiil study of both living and preserved specimens, I am fully satisfied that this is the case. Careful dissec- tion has shown me that they are mere displacements of the mucous membrane, varying with the degree of contraction of the constrictor laryngis and petrohyoid muscles, and I regard them, moreover, as identical with that fold described by Heule in Engystoma, and likened by hiin to the epiglottis, as already skated.

'I'he facts before us go far to justify the view that the epiglottideau folds of the A.nura are homologous with the epiglottis of Mammals in its most typical form ; a striking difference, however, suggests itself when comparing the two more closely. The RIammalian epi- glottis is always chondrified, and iu direct relationship with the

1 . Anatonlie ~iienschlicher Erubryonen,' pt. 3, p. 60 et zp. Leipig, 1885. Insufficiently expressed in the wooclcut,, at **.

3 : 3 *

496 PROF. G. R . HOWES ON A N UNRECOGKIZED FEATURE [June 7,

thyroid cartilage ; in the Amphibia, on the other hand, the epiglot- tidean fold is entirely membranous and in direct connection with the arytenoids, the front faces of which it surmounts (fig. 9, ey.). This difficulty, however, vanishes wlien it is considered that all recent investigation goes to show that the epiglottis is a secondary structure, formed independently of the rest of the larynx, and suhject to the greatest variation. Dubois states’ that in hlammals it represents a chondrification of the submucous tissue of the glosso-laryngeal fold, and that it only secondarily comes into connection with the thyroid cartilage. Clearly, then, the absence of a thyroid cartilage in the Amphibia cannot militate against my view. The same autliw further states that the epiglottis is present only in Mammals ; but comparison of figs. 1, 1 a, 4, with His’s figures of the human embryo referred to, shows that that is certainly not the case. Consideratinn of the above facts, taken collectively, forces us to the conclusion that the epiglottis may be represented in Anurous Amphibians, and that in a form most nearly realizing, so far as our present knowledge carries us, the initial stage in its development in man himself.

The interest attaching itself to the discovery of the epiglottis in Amphibia is self-evident, and it opens lip questions of no little mor- phological importance. Future investigation must decide how f‘ar that which is customarily termed the epiglottis in some Reptiles2 and Birds may or may not correspond with that of Mammals.

A consideration of the functional significance of this orgaii, as here described, reveals some interesting facts. Examination of the figures shows at once that the Amphibian epiglottis cannot have much, if any- thing, to do with deglutition. The classical Johannes Muller WRS

one of the first to insist upon the functional importance of the Marn- malian epiglottis as a voice-organ ; and it is now clearly established’ that while that structure is not essential to deglutition, either of solids or liqnids, it is indispensable to the full erercke of the voice, playing a part in phonation of unexpected importance.

Turning to the Amphibia, it is satisfactory to note that nothing a t all comparable to that which I have described is forthcoming among the tailed forms. The epiglottis is clearly appearing among the Anura ; and it becomes a question of the highest interest to inquire, in the knowledge of the above physiological facts, whether its relative development is in any way associated with that of other accessories to the voice, so well known i n the males of certain genera.

Henle long ago called attention (1. c . ) to the existence of minor sexual differences in the laryngo-tracheal skeleton in Pipa and other Anura. I have already shonn that traces of both the epiglottis and epilaryngeal folds may be found in females of certain species (figs. 1 c and 3) ; and, on turning niy attention to the proposition above named, I was, a t first, awarded with unexpected success. I n a male of

L.C. p. 186. qj’. Henle, 1. c. pp. 61,52, and 60, GI. H andbucli d. Physiolog~e, 1840. ‘ Walton, “The Punction of the Epiglottis,” Journ. of Physiology, vol. i.

1SiS-9.

1857.1 IN THE LARYNX O F THE ANUKOUS AMPHIBIA. 497

Ceratophys au?ericana (fig. 4, p. 495) both epiglottis and epilaryngenl folds were found to be well developed ; but though conspicuous, they remain cornptiratively iiisignificant beside those of an adult male of

Chiroleptes australis, that being one of the genera possessed of the well-known gular sac (s.g, s g ' , figs. 6 and 9). This specimen presents the maximnm developmelit observed, and J have accordiiigly figured

498 PROF. G. R . HOWES ON A N U N R E C O G N I Z E D FEATURE [June 7,

it (figs. 6 lo 10) in all its iniportant aspects. The interest of it is vastly increased when it is stated that i i i a fentale which I have exa- mined the folds were absent, while in a second male they were insignificant '--facts which point to the conclusioii that their develop- ment takes place comparatively late in life, in all probability corre- latively with sexual maturation.

The sauie, in longitudinal section. References as in previolla fignres. c.n, right arytenoid : c It, cricoid (laryngo-

trac1ie;il) cartilage; hy, body of hyoicl ; vz 96, penio-liyoglossus muscles ; pc, pericardinm ; pl , entraiice to right lung ; 5.9, gular 5 m , uc, riglit vocal cord.

Figs. 6 to 9 will show how fully the two structures may be de- veloped. The epiglottis, ey., stands out prominently in front ; the epilaryngeal fold is hood-shaped, overlianging the hinder third of the aditus, and the t w o together emhrace the lateral walls of the adituj for nearly its whole extent-so much so, that I at first took them to be continuous. This, however, is not the case, and here, as in all other examples with which I have met, the t w o structures, although they may approximate, are perfectly distinct (cf. fig. 5). The epilaryngeal fold is entirely membranous, exhibiting on its upper part, which is

For t h o opportnnity of examining these two specimens, as for further assistance, I am indebted to tlie courtesy of Mr. Boulenger, F.X.S., of the National Museum at South Kensington.

Calyptocephabus gayi. 8. Ceratophrys americuna. 8 . Chiroleptes australis. Adult d .

Heleioporus albopunctatus. d . Leptodactylus pentadactylus.

d .

“Bombinator bornbinzcs. 6 . Chiroleptes alboguttatus. 6 . - australzs. d juv.

“Rana esculenta. d, Q . *- temporarta. d, Q .

500 PROF. G . R. HOWES ON AN UNRECOGNIZED F E A T U R E [Jiice 7 ,

d. Epiglottis and epilaryngeal folds both absent. Bufo typhonius. 9 . Cornujer uitianus. Q .

*Hpla meridionalis. - venutosn. d . “Hylodes martinicenais. 9 jiiv.

Nunnophrys cel/lonensis. . Phyllortierlzcsa Cicdor. juv.

d , 9 .

P h r j J J l i S C U S I a V i S . Q .

*Rann esculentn. 8 , Q . *- temporaria. 8, Q . - teirtporalis. d . Rnppin ninrwtorntn. $2 . Rhacophorus inaciiltrlus. 8, Q . Rhznodermn dnrwini. CS . Xenophrys inoat icda. Q .

Perusal of the above briirgs into proniinenre a uitle range of in- dividual and specific variations. The former are, however, less striking tira~r appeiirs a t first sight, aid little more reniailiahle tlian those of the Ampliibian matins and pep, or of tIie B;itoid intrstinal valve’ ; tlie latter tiud a near 1iarallel i i i the modifications of the 1:iiynx in certain fruit-eating Bats recorded by I)obsonZ. That obsen er has obtained a satisfactory exlilanation of the pheiioniena upon purely physiological grounds, and tlie ana1og.y suggests that an investiga- tion into tlie habits of the liting Anura iiiitler consideration niiplit furiiisli a *ilnilar clue. If, as the analogy to Chirolejites would suggest, the folds in questiou are associated mitli the development of vocal sacs, it is difficult to reconcile their absence in Hylo meri- dionaEis and 11. venutosa, with their presetice i n 11. doEichopsis and Bombinator ; while the combination met with in Heleioporus would appear to negative the suppsitioil. It rriay be held that the struc- tures are developed periodically or with sexual maturation ; but, if so, it becomes hard to account for their insignificance ill the male of Leptodaclylus (fig. 2) , in which the accessory spurs had attained R considerable developmeiit. The facts adduced in C h i r o l q t e s aus- tralis, while they appear to favour this b:Jlief, are, in themselves, capable of it different interpretation, for that the differences between the two n:alrs which I have examined are due to age is a surmise. The facts are very puzzling; but who i 3 to say that, in the nio& marked cases recorded, we may not be dealing with a character of‘ taxonomic value, fit at least to rank with those custoniarily relied

considerations, examination of tlie figures shows most conclusively fi iat, iii the slieciineirs which I have described,

Chiroleptes (fig. 6) and Rana (fig. 1 6) stand at oppode ends of a series, the interniedinte steps in wlrich are furnished by Cerntophrys, Lepiodgctylus, and Anna pipiens, in snccession. The question therefore arises as to which of the two first-nanied typifies the more primitive arrangement.

upona. Apart from the abov

gradational niodification can be traced.

’ T. J. Parker, T. Z . 8. 1880, pp. 4941. a P. Z . S. 1881, pp. (i85-93.

Cy: remarks by Husley on the “Taxonomy of the Caniclre” (P. Z. S. 18S0, p. 2W), and by Herdinan, “On the Sliecific unct 1,ocal Variations in tlic l’uniwta (First Report of the Liverpool Marine Biological Committee, 1886, pp. 355, 356).

1887.1 I N THE LARYNX OF THE ANUROUS AMPHIBIA. 50 1

If the apparatus be held to be of none but physiological significance, the clue to its meaning has yet to be found ; it is clearly i n no way associated with the developillelit of the vocal sac, nor is it confined to the males, as might be supposed. In the higher forms, however, it is vestigial and of little or no functional importance ; and there is nothiiig forthcoming i u ally one specimen which is not represented in tlie male of Chiroleptes. These facts, in view of the uniformity of development of the pnrts in the latter, would seem to me to suggest that the apparatus has an important morphological significance, and that in the admittedly lowly Australian type the primitive condition is most nearly exemplified. The problem is an interesting one, aild furtlier iiivestigation is necessary for its solution.

Turning, finally, to the question of general morphological import- ance, i t is clear that the discovery of the epiglottis in the Amphibia, in the form aiid under the conditions here recorded, carries back R stage further the initiation of one more structure pecuiiarly charac- teristic 0.f Mammals. The interest of this is increased when we reflect upon the identity of the Amphibian epiglottis, which is clearly bilaterally symmetiical, i f not actually paired, with the initial phase in development of that organ in the human subject as observed by His. The facts show that the origin of this typically Mamma- lian structure must be sought in animals lower than the living Lizards ’. I t would be wide of the mark to form data for diwu,sing the question of Mammalian affinities upon it. I n view, however, of the aiiticipation o f t h e caecum coli i n the Common Frog’, and of the excursions made by those Amphibia with suppressed larval meta- morphoses’, in which it is highly probable there may have been foreshadowed the fcetal membranes of the Amniota (cf. Huxley, P. Z . S. 1880, p. MU), the facts here recorded can best be regarded as in[ icative of’ similar excursions towards the elaboration of the voice-organ, anticipatory, as has been shown, of the characteristically Mamuialian condition.

It is of interest here to note the existence of an epiglottis-like fibro-carti- laginous plate in Profopterus. Attention was first drawn to .it by Hcnle (1. c. pp. 5, (i), and it was sliortly afterwards described in full and figured by Bischoff’ (“ ?escr. Anat. d u Lepidosi~aia pm-ado.Cci,” Ann. Sci. Nat. t. xiv. Zool. 1840 p. 136). Wiederslieirn has refigured it, (‘ Lebi-bnch’) and recorded (ibid.) the cliscoyery of an antdogous structure in LqJidostezhCs. ’ Huxlev, in Huxley and Martin’s Eleiri. Eiology, 1875, p. 1GG. ’ Espec;dly K~totrentn, Weiuland, Arcliiv f. Anat. und Yhys. 1854; Hyboh,

Petcrs and Quncllach, Monatsb. Berlin. Acad. 187G ; Ranclopisthodon, Boulenger, Y. Z. S. 701. xii. 18HO ; I’hyllonzedusn, v. Ilering and Boulenger, Ann. Sr Mag. Nat. Hist. vol. xvii. 1886.

&’or a r&tlizC( of tlie subject gcnerallg, with fnll references and list of species, see Boulenger on l’hglloiiiedusn, op. cit. p. 464. Cf. also Smith and Cope on Deiadrobntcs, Aruer. Naturalist, 1887, pp. 307-311.