Download - On the Carpus and Tarsus of the Anura

1888.1 ON THE CARPUS AND TARSUS OF TEE ANURA. 141

a. Diseoglossidm, p. 164. b. Pelobst,idae, p. 165. c. Hylidtc, p. 167. d. Dufonidse, p. 169.

e. Cystignathids, p. 170. f. Engystotnatidx, p, 172. g. I)euclrobatidrc, p. 173. h. Ranida, p. 173.

I. INTRODUCTION. a. General.-While, during the last few years, the researches of

PROC. ZOOL. Soc.-l888, No. XI. 11

142 [Mar, 6,

Wiedersheim and his pupil Kehrer, and of Baur (l), have added ronsiderably to our knowledge of the limb-skeleton of the Urodela, that of the Anura has received much less attention. The most important investigations recorded since the classic of Gegenbaur (IS), are those of Born (3 , 5). Briihl has published (11) figures of a somewhat extraordinary character, pertaining to some seven or eight genera ; but concerning these we have little to say, except that we fully endorse Born’s criticisms and corrections of them (6, pp. 48 e t seq.).

Born’s latest communication purports to he somewhat revolu- tiouary, so far a t least as the carpus is concerned (6, p. 62). It will he seen, however, in the sequel, that we are unable to accept his determinations ; and as the discovery of new elements and relationships has led us to differ from most of our predecessors, in our estimate of the morphological value of leading elements of both carpus and tarsus, it is binding upon us to preface this paper with

4 an account of those observations up011 which our dissensions are based. I n accordance, then, with the exigencies of the case, we shall divide the paper iiito two sections : viz. :-firstly, a geneval port, in which will be given R review of the morphology of the carpal and tarsal elements, together with a detailed account of those hitherto unrecognizecl structures, whose discovery has necessitated a redeter- mination of the value of any one or more constituents ; and, secondly, a specin2 part, in which the leading families will be taken successively, the characters of thc carpus and tarsus of each being noted in the light of the preceJing section, and special peculiarities dealt with.

b. nroinencZatiire.-Our investigations do not profess to be exhaustive ; they leave much to be settled, and we have accordingly refrained from introducing a nomenrlature which niiglit embody a premature expression of homological relationships with the limbs of other animals. In dealing with elements whose morphological value remains in the least degree doubtful, we have, for these reasons, adopted the nomenclature of Ecker (17 ) as being the more empirical and better suited to our purpose than the alternative one of Dug& (1 6).

In describing the pre-hallux (calcar), we have, in order to facilitate comparison with the works of our predecessors, adopted the older system of nomenclature, calling the outermost digit the fifth, the iouermost one the pre-hallux or calcar, and that next to it the hallux or first digit.

New terms have been introduced only where unavoidable; and some of these are, for reasons already given, purely empirical.

c. Material.-It will be seen, from the second part of the paper, that we hale examined the limbs of 37 genera and G O species- adequate representatives of all the leading families as defined by Boulenger (7), with the exception of the Uyscophidm, Dendrophq- niacide, Anzphynathodontidre, aiid HenaiphractidcE. For these specimens our grateful acknowledgments are due mainly to Prof. IIuxley, and to Mr. G. A. Boulenger of the Natural History Museum. T o the last-named gentleman we are indebted for further

PROF. G . B. HOWIS AND W. RIDEWOOD ON

1888.1 TEE CARPUS A N D TARSUS OF THE ANURA. 143

aid and advice always most cheerfully given; while we have t o tender thanks to Prof. W. K. Parker for useful specimens, to Prof. Wiedersheim o f Freiburg for the limbs and a larva of Pipa, to Prof. W. N. Parker for some exquisitely preserved tadpoles of the Common Frog, and to Dr. G . Baur, of New Haven, Conn., for suggestions and advice.

Some of our best results have been obtained from feet in which ossification was just commencing (ex. fig. 5 ) . A t this stage the outlines of the cartilages are more clearly defined than at the earlier ones (g. fig. 9), while there is a total absence of any distortion or displacement consequent upon complete ossification (~f. figs. 6 & G). Identification of this stage by external appearances is very difficult; absolute size, moreover, gives 110 clue to it, for while in a Xenophrys measuring 27 millim. from mouth to vent (fig. 14) all expectations were realized, in it Brachycephalus of 20 millim. ossification was found to be complete.

d. Methods.-We early discarded the section-cutting method employed by Born (3, 6 ) , chiefly because the results finally arrived at were less satisfactory than those obtainable from clarified preparations. Clarification in potash solution (kindly suggested to 11s by Prof. Wiedersheim) was tried, but relinquished, chiefly on account of the difficulties experienced in mounting the preparations. Better results were arrived at by clarifying in a mixture of glycerine and potash, and mounting permanently in glycerine jelly ; but our best results were all obtained by the use of clove oil, which offers additional advantages in allowing the preparations to be stained. Borax-carmine solution proved most serviceable RS a dye; less successful preparations were, however, obtained with picro-carmine, haematoxylin, and bismarck-brown.

Our method of procedure was generally as follows :-The limb having been carefully skinned, the larger muscles were dissected off, in order to facilitate staining. The preparation was next immersed for about ten minutes in borax-carmine, and then transferred to acid alcohol (75 parts alcohol, 3 parts HNO,, and 22 parts water), in order that superfluous stain might be removed ; after dehydration in absolute alcohol, it was allowed to remain in oil of cloves. The final process consisted in carefully picking away the soft parts, little by little, with two pairs of small forceps under a simple microscope, while the preparation was still immersed in a glass trough of clove oil. The smaller preparations were ultimately mounted in Canada balsam.

The above method admits of an examination of all parts in the undisturbed state, and consequently of an accurate determination of the relationships of' apposed surfaces, such as is not possible with any other. Moreover, in that ossific centres, lines of fusion, furrows, or outstanding processes are rendered eqcially obvious, it is manifestly not open to objections which might conceivably arise out of exclusive reliance upon the section-cutter's art.

11"

144 PROF. G . B. HOWES AND w. RIDEWOOD OM

11. GENERAL PART. HIND FOOT.

a. Metatarsals and Phalanges.-Setting the pre-hallus aside, the lst, 2ud, 3rd, 4th, and 5th digits bear, in most genera (cf. footnote on p. 178) respectively 2, 2, 3, 4, 3, phalanges, the 4th, or outermost digit but one, beiug the longest. Pipa is alone exceptional, for in it the 3rd exceeds the 4th in length.

11. Astrugalus (a.) and Calcaneus (c.).-‘P’hese two elements were already greatly elongated in the youngest specimens in which their presence could be detected. When fully formed, the two bones are generally uniform in length ; they are relatively longest in the Tree- Frogs, shortest in Pelobates. When of unequal length, the preasial bone, or astragalus, is the shorter.

Wiedersheim has shown I , in ICizna esculenta, that the arteria iiiterossea perforates the men1 brane passing between these two bones,

rdorso-ventrally, to reach the plantar surface of the foot. Pelodytes is alone exceptional, among a11 the gcnera which we have examined, in the fact that it,e astragalus and calcaneus have become greatly elongated subsequent to complete fusion (Plate VIII. fig. 12), in a uianiier strikingly suggestive of the tibia and fibula. The above-named artery, however, remains true to its original relationships, a small forameu (f .;’ . , fig. 12) being left for its transmissioii. III this, as in a11 other genera, neither the astragalus uor calcaneus (however much modified) ever undergo aiiy sort of rotation ; they lie side by side, invariably com planate with the tibia and fibula.

Wiederahrirn has sliown further ’ that in the Urodela (Ranodon, Salainaizdrella, CryptoLrancJius) a blood-vessel perforates the tarsal region apparently, a t first sight, in the mamier of the above- named artery of the Frog. Baur has more recently recorded the same fact for Necturus (1, pl. i. figs. 12 & 17). Hyrtl, describing the vascular systerii of Cryptobra7dms, says of the crural artery :- “ horum ossium big.,?, cui nulla articulatio intercedit, et yiiae potius textu fibroso in unum qunsi corpus coiijungitur, arteriae iiostrae commodtlm praebet occasionem, trajecta, syndesmosi intertarsea, ad domafern tarsi regionem ernicandi, quo territorio semel poti ta, illico in duas, panllo post iu quatuor arterias digitales communes dorsales dilahitur, binorum digitorum interstitiis deatinatas.” There can be little doubt but that this description applies to the vessel noted by Wiedersheim. I n that it passes ventro-domally, however, it ditfers in toto from that of the Frog, but iu this it agrees just as closely with tliearreria br.acliialis OF the fore limb of that aniinal (cf. p. 156). These facts tend to show tlliit the perforation in question (foramen iirtertnrsi, [auct.]) is probably uot homologons with either that of the Urodele liind foot, or that of‘ the fore h o t in the Frog itself; if

’ Anatomie clcs Froeches, Eclier and Wiedersheim, Pint II., Brunswick (1881), 1). 81;.

2 “ Il ia idtesten J3oruirn des C ~ U Y uiid T a m s d. hentigel1 Ampllibien,”

Vindobonae, 1865,

[Mar. 6 ,

Norph. J:hrb. vol. ii. ( lSTt i ) , pp. 481-..23,5.

p. 113. Urgptobrsnclius Japouicus, Schedisma aimtomicum.

1888.1 THE CARPUS AND TARSUS OF THE ANURA. 145

SO, the arteria interossea will not help us in the least in estimating the morphological value of the astragalus and calcaneus.

Wiedersheim asserts (36, p. 211) that in the astragalus we have " ein vereinigtes Tibiale und Intermedium." Gegenbaur (18) and Born, on the other hand, are inclined to regard the tibiale as its sole representative. Proof of the former statement is not forthcoming ; and in spite of the most diligent researching, we have been unable to discover, a t any stage, the remotest trace of anythiug at all com- parable to a third proximal tarsal. Future research may, perhaps, demonstrate its existence ; but, mindful of the elongation of the earliest rudiments of these elements, we incline to the belief that the missing one (1 intermedium) has disappeared beyoiid all recognition, even daring the embrponic period.

c. Tarsalia 4 and 5.-In no Anurnn foot hitherto described have there been recognized more than t h e e distal tarsal elements (tarsalia). Examination of the Plates will show that when these remain distiiict they are directly associated with the hallux and the two next digits (ex. Plate VII. figs. 8 and 10). The tarsalia of the two outermost digits (4th and 5th) would thus appear to be abseiit. Gegeiibaur first drew attention to the existence, in certain genera, of a ligament (fias. 10, 17, and 19) which passes between the tarsal of the third digit and the head of the fifth metatarsal. H e described its relations minutely ( 1 8, p. 6 I ) , and concluded that '' er repriisentirt in ligamen- tosem Zustaadc Theile, die nnter anderen Verhaltnissen nls Knorpel gebildet sind." This structure has been recorded in H y h , Rhino- derma (Gegenbaur), Pelobntes (Gegenbanr, 18 ; Bayer, 2 ) , Disco- glossus (Wiedersheim, 36, p. 211), and others, and to the list we are able to add (cf. Part 11.).

Born says of Phryne (Bufo ') uu7garis ( 3 , p. 434) " im Band- polster unter der Basis von Metatarsale IV. traf ich einmal einen deutlichen Rest von Knorpelgewebe, was die Gegenbaur'sche Deut- uiig dieses Bandes als Homologon eines Tarsale IV. durchaus bestiitigt." We find that in the adult of Alytes obstetricans this ligament (fig. 10) carries a well-marked nodule of cartilage, which lies between the applied heads of the 4th and 5th metatarsals. That this represents one of the missing tarsals is hardly to be doubted from its relationships, but, from its position, it is difficult to say which. Gegenbaur goes on to assert (p. Gl), " wenn man das auf ein blosses Ligament reducirte Tarsusstiick einem der bei den iibrigien Amphibien nachgewiesenen Elementarstiicke vergleichen will, so knnn man in ihm nur nach Massgabe der Betheiligiing des Metatarsus das Tarsale 4 und 3 erkennen, welches bei den Tritoiien schon durch ein einziges Stiick dargestellt war." I n Bufo, Hylodes, and Hyla (Plate VTII. fig. 19) the ligament enters into an extensive connection with the head of the fifth metatarsal, like that originally figured by Gegenbaur (18, pl. 4. fig. 13) for the first-named genus. I t fuses (fig. 23) with the postaxial articular border of this bone, and the head of the Eame is seen to be obviously enlarged as the result of its attachment '. It becomes a question of considerable interest, therefore,

See Boulenger (7, p. 303). Cf. & n O J J h ? p , Plate V I I L fig. 17.

14G PROF. G . B. HOWTES A N D w. HIDEWOOD ON [Mar. 6,

whether this enlargement may or may not represent a coalesced 5 th tarsal (as, indeed, Gegenbaur was inclined to believe, pp. 64-65). We have failed to detect any trace of segmentation of the cartilage at this point. I n the Hylitlce, however, a second ligament is present, which passes (fig. 19) from the postasial wall of the capsule of the tarso-metatarsal joint, npwards and inwards, for attachment to the epiphysial cartilage of the astragalus and cnlcaneus. This ligament carries at its outer end a wedge-shaped cartilage, and that we at first took to represent the fifth tarsal. Upon reconsideration, however, we are disposed to regard it as an ordinary intra-articular cartilage of none but physiological significance (for further details see p. 168). lteflection upon these facts has led us to the belief that while the ligament which passes between the third tarsal and the head of the 5th metatarsal is the degraded representative of the 4th and 5th tarsalin, the cartilaginous nod& occasioiially carried by i t is exclusively a vestige of the 4th one. Gegenbaur records the presence of this liganient in Rana esculenta, in addition to the other genera named, and we cau confirm his statement. I n R. temporaria, however, we find no trace of it in the adult, and but a feeble one in the larva. This fact, while lending additional support to Gegenbaur's original deduction, shows how completely the vestige is, as it were, disappearing under our eyes.

d. l'arsulza 2 and 3 (0s cuboideum).-We have already stated that three tarsal elementa may exist ; 011 the other hand, the outermost two of these may not unfreqaently be replaced in a long bone (cnboideum), represented in its typical condition at ill figs. 17, 24, 27, 29, 31, S: 33. I t will be observed in all these cases that this structure overlies, more or less completely, the metatarsalia of the 2nd aiid 3rd digits ; comparison of the same with those forms in which the three tarsalia are found (Bombinotor or Alytes, figs. 8 B 10) leaves little doubt but that the bone represents a confluence o f the tarsalia of the 2nd and 3rd digits, as asserted by all recent authorities.

I n some cases, however, evamin~tion o f the adult lirnb would appear to render this questionable. For example : in N y l a caridea (fig. 19), in \thich two tarsalia are alone present, the larger one I S little, if a t nll, related to the 2nd metatarsal-the head of which is in apposition with the secoiid and smaller element. Com- parison of this tarsus with that possessed of the three tarsalia (figs. 8 & 10) lenders it l iwl , indeed, to say which two elements are represented in the Tree-Frog. Born has already experienced this difficulty in dealing (6, p. 443) with the adults of Yhryne vulgaris and Bufo varzabilis, and he adinits his inability to settle the question, for want of' embryological material.

0 1 1 turning to the latiit of B y l a (fig. Sl), we find that the outer of the two elemelits has (unhl\e that ot the adult) more than twice the bulk of its fellow, wliilc it agrees in every detail witli the coboideurn iii its most typical tbrrn (cf. Xenophrys, fig. 17) . Examina- tion of this element side by side with that of the adult limb (fig. 19) shows that its growth is arrested early in development : it shows,

1888.1 THE CARPUS AND TARSUS OF THE A N U R A . 1.17

further, that the connection between the innermost tarsal ( 1 ) and the head of the second digit is purely secondnry, a d that the first- named is really the tarsal of the Iiallux, displaced, as it were, in sympathy with its dnarfed neighbour.

Ecker speaks (17, p. 61) of the cuboideum in Bufo as consistirig of two pieces. We find that in that genus, as in Leptodactylus, a similar shortening up of this hone takes place ( c f . fig. 23) ; there can be little doubt therefore that Eclcer was misled by that, precisely in the manner indicated above.

The cuboideurn, then, must be held to represent the confluent tarsalia 2 and 3. We have to record its presence in the Raniclce, Dendrobatida, EiLgystomntide, Cystignathidce, Bzcfonida, Hylidce, Dactylethridm, and Pipa.

e. Fursale 1 (IIalZuz tarsal) (I). --Tliis element is of very constant occurrence. I t s relations in the Discoglossidce are exceptional and, as will be shown below (figs. 8 S: lo), inteiestirig ; it lies in them under cover of‘ the unviculare (n’), interposed between it and the 1st metatarsal. Gegenbaur appears, in dealing with Bombinator, to have mistaken the naviculare for the first tarsal (18, pl. 6. fig. 1 I ) ; but this is not to be wondered at, in consideration of the small size of the latter. I t is wanting as a distinct element in old specimens of Nuiinophrys and Phryniscirs, and we have found that it fuses in both genera witli the naviculare. 130m has described aud figured it accurately in R a n a esculenta. Concerning R. temporaria he says (3 , p. 4-11) that it “imnier in Laufe der Ontogenese bis auf uner- hebliche Spuren scliwindet.” Strictly speaking! this is not the case, for we are in possessiou of old examples in which it had persisted as a distinct element (woodcut, p. 156, fig. C, I). We are i n c h e d moreover to believe that instead of undergoing supprrssioo, as Born imagines, it more prohably fuses with the naviculare as in tlie two above-named genera.

f. Naviculare and Pre-hallux tarsal.-Ecker regarded the naviculare of these animnls R S analogous (17, p. 61) with the elerrient so named (centrale) in the tarsi of the liigher animals. Gegenbaur, on the other hand, who, as before stated, confused it with the hallux- tarsal, sums u p his coi~clusioi~s in the words (18, p. 67), “ein Centrale fehlt.” Born adopts this view; but he differs from Gegenbaur in that he regards the naviculare as the basal srgment (tarsale) of the pre-hallus. H e urites (3 , p. 448), “vou diesen [segments of the pre-hallux] liegt drr erste iii eiiier Reihe niit den Tarbalknorpeln, aelche Metatnrsustriger siiid, ahmt dieselbeti in Form und Striictur nach und ist von den Autoren auch, wie erwiihnt, irnmerals ‘ typisclier ’ Tarsallrnorpel aufgefasst worden.” Born’s deteriniiiatioii has been adopted by all subsequeut writers. I t will thus be seen that the two views entertained as to the morphological value of the naviculare are diametrically opposed, and that a settleirient of their differences is indispensable to the full determinatiou of the value of‘ the distal preasial eleinents. We have already drawu attzution to the fact that the naviculare of the Discoylossida (PL Bombinator, fig. 6, Alytes, fig. 10) is wedgecl in

148 PROF. G. B. HOWES AND w. RIDEWOOD ON [Mar. 6,

between the first tarsal and the astragalus, being in fact absolutely central and in a position which cannot, by any stretch of imagina- tion, be said to be “ i n einer Reihe mit den Tarsalknorpeln.” I t would be more correct to say that the 1st metatarsal, 1st tarsal, and naviculare are here disposed lineally.

Examination of the Plates will show that the relative size of the naviculare is in no way proportionate to that of the pre-hallux, as might be expected were Born’s hypothesis tenable (cf. figs. 3, 10, 12, & 17) ; it is rather the reverse. Choice appears to lie between two alternatives : the naviculare has either grown in and displaced the hallux tarsal proximally (as the Discoglossii2e would suggest, figs. 8 & 10, in which case the condition of the latter would, in this family, be secondary) or it,has undergone a reduction proportionate with an increase in size of the same. Eeccnt obserrers are agreed as to the lowly position of- the Discoglosside’, and it therefore becomes a question of first irnportance as to how far the arrangement just described in them is primary.

I n the youngest Discoglossidce examined by us (cf . p. 163), no appreciable difference conld be detected in the relative proportions of the two elements under consideration as compared with the adult ; we have here, then, stron? reason for regarding the condition exemplified iri them as really primary. The matter, however, presses still more closely. Examination of those forms in which the pre-hallux is regularly srgmented (fig. 19), shows that that segment which Born regards as its metatarsal (ph. i.) is generally disposed in a line with the naviculare, n’ (cf. figs. 13 & 19). I n Pelodytes, hoa- ever (fig. 13) and less conspicuously in Alytes (fig. 10) and IIyZo, the first-named element is intimately connected with the adjacent hallnx-tarsal (1) by means of a well-defined ligament, which runs distally to the naviculare. Comparison of this ligament with that previously referred to as the representative of the 4th and 5th tarsalia, leaves little room for doubting, if argument from arialogy is worth anything, that Born’s metatarsal of the pre-hallux is, in reality, its tarsal. If this be admitted, there can no longer be any doubt of the homology of Ecker’s naviculare (a’) with the 0s centrale tarsi.

This determination brings the tarsus of the Snura into closer harmony with that of the other and, especially, the higher Vertebrata than is now generally admitted, while it sets at rest further doubt as to the value of the hallux-tarsal.

Finally, Ecker has shown (17, p. 128) that the tendon of the addoctor longus digiti primi muscle (m.a.2. of woodcut, fig. C, p.. 176) is inserted into the naviculare. We have found this to be Invari- ably the case whether the hallux-tarsal is present or absent. It follows, therefore, that all possible source of confusion between that element and the naviculare is at an end.

g. Pre-hullus (Cu7cnr).--The skeleton of the pre-hallox was first figured by Rosenhof ( 3 2 ) ; he described it (Pelobates, p. 83) in the words “tandem etiam in sceleto hoc notari meretur singularis quidam

CJ Cope (12, p. 104) and Boulenger (7, p. 444).

1888.1 THE CARPUS AND TARSUS OF THE AKURA. 149

spurisque unguis.” I t has been likened by some anatomists to a single element in the tarsus (ex. entocuneiform, Owen, 31, p. 184) ; it is, however, unnecessary to recapitnliite these comparisons in detail.

All recent writers are agreed as to the variable natare of the pre- hallux, but the range of this has not yet been fnlly recorded. As will be seen from an examination of our figures, it is very inconstant in size, shape, and detailed characters. For exnmple : in the adnlt Xeizophryls (fig. 1 T), i t consists of a single elongated cartilage which ossifies very lat,e in life, whereas in &la (fig. I!)) i t attains a much greater development, and is segmented into four pieces. 1)isparity in size is by no means confined to members of different familieq for in Pelobates-the type of the family to which Xennphrys belongs- the calcar, while consisting of a single piece, attains enormoils proportions and ossifies very early ’. I n this burrowing genus tlie calcar supports the well-known horny blade, by means of which the animal is enabled to shovel up the ewrth. This being so, its early ossification is distinctly associated with the functional requirements of the case.

Born at first ascribed (3 , p. 148) four srgments to this structure in R a m , Bilfo, and others. In his later researches, however, h e records coiisiderable variation in tlie same, stating (6, p. 61), “ wechselt doch die Stiickzahl v. Rnna escirleutu, zwischen zwei und vier, bei Bufo atwiabilis, zwischeii drei untl fCiiif.” Oiir figs. 19 & 19 a agree very well with his dcscril)tion, while they represeiit tlie inaiiinrirn segrneiitat,ion into lineally disposed elements observed in any specimen. W e have already shown reason for regarding Born’s basal pre-iiailiii segment as the centrale (‘nnvicu- Clare), wlieiice i t follows that in no known Anuran does the calcnr, as c l e f i d by us, ever consist of more than four true segments. This rnasirnum number is reached iii Bufo * (?), IIyict, Hylarann, Leptorlnctylzcs, hrototrenza, and nann-repiesentatives, tliat is to say, of f’our distinct families.

Incident iipoii the foregoing there arises the question as to whether tlie unsegmented coiidition of the pre-hallux is the more primitive w e , or vice versl. I f the latter be the case, traces of segmentatioii ought to be forthcoming in the larva of those forms possessed of the sirigle piece. Borii accords to Rann esculenta a maximum of three pieces. Iu old Tadpoles of R. temporaria, however, me have detected the presence of four distinct segments of hyaliiie cartilage : the terminal one of these early degeiierates and becomes erentually fibrous, ni:d not until then can that fusion of the parts which Born descrilws tnke place. The facts recorded by this author alone for Hyla and Bufo (Phryne) (6, pl. 1. figs. 2 & 3 ) are, in themsclves, sufficient. to s!iow that the liiieally segmented condition is most probably the primitive one.

We are doitbtf‘nl as to how far the one or two pieces, met with in

’ Borii Inn l ies tlie assertion (6, 13. 69) tllst it, here bears origiiidly two plrs-

2 Born ((i, p. 61). 3 p. 50.

lniigos: this we cannut confirin. Wc h a w not observed it.

We have not obacrved this.

i50 PROF. G . B. HOWES AND w. RIDEWOOD O N [Mar. 6 ,

adults of certain species, may or may not arise by confluence or suppression of individual segments (cj’. fig Sj ; future research must decide this question.

1 , I f the segmented condition be the more primitive, may not the whole pre-hallux represent a shortened-up digit ? and 2, May not this supposed sixth digit represent a lost ray of the iclithyopterygiurn 1

In Nannoplwys, Phyllotnedusu (fig. 19 u), Rhombophryne, and especially in some species of Elyla, t.he pre-haIlux has quite a digitiform aspect, comparing, a t first sight, very favourably with the reduced pollex of the fore limb, tlie homology of which with the other digits nobody disputes (cf. Plate VII I . figs. 16 u & 20).

That the pre-hallux may, in a sense, resemble the pollex is clear from the preceding; but it must not be forgotten that this resem- blance is most marked in tlie specialized Tree-Frogs. Its segments’ are, in the embryo, pore digitiform than in the adult ; and taking all facts into consideration, we incline to a belief in a 11reponderance of the resemblances over the differences between this structure and a normal digit.

That the pre- hallux takes on certain of the essential relationships of a digit is beyond dispnte.

. T h e tarsal of the pre-hallus is, in most CRSCS, i p definite articulation with the hear1 of the naviculare (cf. Pelodytp, fig. 13). In many forms its second segmeot is tlie*largest (ex. fig. 2 ; ) ; when this is the case, that may be disposed parallel with tlie nietatarsals, in a manner strikingly suggestive of a serial honiology (cf. Xeiiopgs, fig. 3, arid Pseudis, fig. 27). In Xenophrys, Ilylu (figs. 17 and 19), and other genera, this supposed metatarsal sell& forwards (back- wards in situj B small retral lobe (*); in Cerutophrys (fig. 23) this attains a considerable development, and in individuals of this genus i t may exceed in size the I J O ~ Y of tlie segment from which it takes its origin.

Meckel (29) and Cuvier (15) held, and Born (3), Kehrer (24), and others still hold the pre-hallux to be a sixth digit. Leydig (25) , attacking it from a totally different standpoint, argues to the contrary. Gegenbaur a t first took a similar view, holding (18) it to be a secondary structure peculiar to the Anura, and his words are echoed by Hofmann (21) ; subsequeiitly however to the publication of Born’s paper, Gegenbaur accepted that author’s position, con- fessing to the sanie in his well-known test-book. Finally, Born asserts, in his latest paper (6, p. 61) ‘‘ Ubrigens bin ich jetzt geneigt in der starken Variabilitat der Gebilde der sechsten Zehe . . . nicht bloss mehr eine Eigenthiimliclilceit zu sehen, die derselbem als rudimentiirern Orgirn anhaftet, sondern ich sehe in der haufigen Verschmelzurig ein Eestreben ein inimer festeres Skelettstiick als

There remain two most important questions.

The second difficulty is not easily to be met.

That it really represents one is another question.

Van Deen records (34) an instance in which, in Ilaiaffi csczibntffi, these lay in His specimen was, unfor- a line with the hallus and supported thc web.

tunately, a monstrous one, with four hind legs.

1888.1 THE CARPUS AND *ARSUS OB T E ~ E ANURA. 151

Unterlage fur den Fersenhocker, der vielfach als Grab-Instrument benutzt wird, zu gewinnen.”

We have already called attention (p. 150) to the development of a retral process from that which we regard as the liallux-metatarsal. In Nyla Zichenutu this is completely segmented off (* in woodcut, fig. A), forming one of’ R series of marginal elements,

Fig. A.

The pre-hallus of opposite sides in Hyla lichenata. Dorsal aspect. X 6.

which skirt the outer face of the calcar and are, moreover, rouglily synimetrical 011 opposite sides. Comparison of’ this figure with that of the more normal pie-hallux of IT. cerulea (Plate VIII. fig. 19) renders it tolerably certain that these nodules are dismemherments of the main mass.

It would he superfluous here to recapitulate the well-known specu- lations and discussions which turn on the supposed ancestral condition of the pentadactyle limb ’ ; it is more pertinent to point out that the evidence against the supposition that all living Amphihia and Arnniota have directly inherited a pentadactjle member, is, to no inconsiderable extent, based upon the discovery of fragmentary cartilages in no way indistinguishable froni these now under consideration. Such fragments have been discovered flanking both pre- and postaxial borders of the one member or the other in even Man himself’ ; and in the fertile imaginations of Bardelehen, Kehrer (24), and others, we find them exalted to the dignity of lost r q s . The last-named author writes (p. 14) “so hatten wir also bei der Beurtheilung des Hand- und Fuss-skeletes der Wirbeltliiere kiinf- tighin nicht mehr von einer pentadactylen, sondern von einer hepta- dactylen Urform auszugehen, und von diesem Gesichtspunkte aus

For details and rismt6 see Gegenbaur (19). and Wieclerslieii~~ (36), pp. 2U4-2UO’i and ”29-280; also “ Zur Urgesch. d. Gliedmassen d. Wlrbelth,’’ ‘ Hnm- boldt,’ vol. 5, 16%.

‘L Bardelebeu, ‘ Jenaiscbe Zeitschr. fur Naturaiss.’ Bd. 19, N.F. xii. Supp1em.- Heft 3, 18%. Cy. Wiedereheim, ‘ Lelirbuch,’ p. 924, and ‘ Humboldt,’ nf .


betrachtet, werden auch firderhin die ‘ iiberzahligen ’ Finger und Zehen . . . . sondern als atavistische Bildungen angesehen werden diirfen.”

Baur is, of all later writers, the one who has done most to combat this doctrine. H e advances ( 1 , p. 68 e t seq.) equally good arguments for regarding these so-called supernumerary digits as purely secondary and adaptive structures, laying, a t the same time, great stress upon their late appearance, especially in the case of the pre- hallux itsrlf. The advocates of the opposite belief’ seek shelter under the stronghold of the Ennliosnz6?*ia, but recent investigation at least syggests that the paddles of those beasts were specialized derivatives ot pentadactyle predecessors ’.

We trust to have already shown satisfactorily that the naviculare can no longer be regarded as the hallux-tarsal, as Born supposes ; and that admitted, it follows that the pre-hallux conforms to the structural requirements of a sixth digit. Nore than this cannot be said at prcsent, and further speculation would be useless until the connecting link between the cheiropteryginm and its piscine pre- decessor shall have been discovered. For this we look to the paleontologist.

Setting aside further discussion as to the exact significance of the pre-hallux itself, we cannot refrain from regarding that fragmentary dismemberment of its outer free border, above represented, as an additional argnmeiit for the views of Leydig and Baur. I f the converse one be justifiable, we should have ample ground for pleading the cause of an octodactyle “ Urform,” and this would be primd f a c i e no adrarrce at all.

There is a growing tendency to attach too much importance to gristly fragments such as those now nnder consideration, and it is binding on those who rnay yet deal with these supposed vestiges (cqecially as manifested in the higher Vertebrata), that they shall determine at the outset, with greater ~ccuracy than has hitherto been done, what piocise relationships they bear to the soft parts. If not, the question bids fair to be reduced ere long to the condition of a reductio iid absurdum.

FORE FOOT. a. Ililetncarpnls and Plialanyes.--The pollex of the Anura nnder-

’ Parker writes (P. R. X. vol. 48 (1887), 13. 57), “ I have frequelltly noticed that aborted parts, like olershndowcd plants, are late to appear, and soon wither, or are nrrestcd in their growth.”

Enur , .‘ Ileruwkungen dbcr Sanropterygia und IclitliJ.opterJai;I,’’ 2001. Airzeiger, 18S& pp. 245-252; also “ Die Ab :niiiiuuug d. Auiuiotrn Wirbelth,,” Eiolog. C‘eiitraltil. Ed. 7 (IS“7. pp. 43-19 Sce also “ On tlio P11ylogenetjc brmngciue~i t of t h o S:luiopsida,” ‘ Jourual O T M ~ ~ $ l ~ l o g y , ’ Boston, rol. i. no. 1 (1887), pp. !GI04 ; tlic discovrry by Gadow and Caur (herein recurcled) of S I I ~ ~ ~ ~ I I I I I I I ~ ~ : L P ~ plialanges in tlio 111nn1is 01’ Nc~ictlicui.~ and A’amfzts is most welcome a t this juiiciare. Ccbor den Ursprnng d. Entreini- tiiten der Ichtliyopterygin,” &rich1 II. d. XX. Vers:iininlung de8 Oberrlieiniscl~en Geolog. Vereins, 1887.

Bee also Banir,

1888.1 T H E CARPUS A N D TARSUS OF THE ANURA. 153

‘goes, ns will be shown later, a by no means insignificant, range of modification. I n no case, however, is it ever segmented into more than three pieces exclusive of its carpal ; it thus never bears more than two phalanges. The other four digits bear each, in order of succession from within outwards, 2, 2 , 3 , 3 phalanges, certain Itanidae excepted (cy. footnote, p. 178).

The inner border of the metacarpal of the second digit becomes, in the males of certain Anura, as is well known in the case of the Common Frog, variously crested or tuberculated in connection with the overlying c r thumb-pad ” (2 ’ . fig. 6), or for support of the horny clasper as in Leptodactylus (** fig. 25). Leydig has recently described these modifications with great accuracy (26), so far as they concern the indigenous European forms.

b. Distal Curyal Elements (Carpalia) .-In the more specialized forms certain of these unite with each other or with adjacent elements (ex. Hyla, fig. 20, Rhomlophryne, fig. 3 2 ) ; and for the present we shall exclude such from consideration, dealing only with those feet in which the elements remain permanently distinct,

All previous writers are agreed that in such forms fire carpalia are represented. While we admit that such nzay be the case, we doubt, as the sequel will show, the homology of that element which our predecessors, not excluding Born (6). iritzrpret as the carpal of the 5th digit.

Curpale 2 (Tmpbze of Dug& ; irrupezoides of Ecker).-This element ( ( 2 ) of our figures) is never more than insignificantly displaced. Examination of any one drawing will show that it lies in a line with the head of the 2nd metacarpal (2‘) ; and comparison of Aly fes tadpole (fie. 9) shows this to be its prirnitive positiou.

I n the Biivco- glossidce it undergoes a slight reduction (ligs. 6 & 7) , and this reaches its maximum, amounting to insiqnificaiice in the of Leptodactylus (fig. 2 5 ) . Comparison of this carpus with the adult D~scoylossus (fig. G ) aiid the larval Atytes (fig. 9) is suficient to show that this partial atrophy is to no small extent associated with the specialization and enlargement of its metacarpal.

We have found this element to be free in all but BrackycephaZus, Pipa, and Rhornbophryne.

CnrpuZia 3 und 4.-In certain forins these two elements become confluent, constituting the iiucleus of a large bone (3 , 4 A, figs. 25, 26, SS), called by Ecker (17, p. 53) the m capitato-hamatuni." In the Discoglorsida?, Yelobatida, and Aglossn, however, carpalia 3 and 4 are distinct. Considering the general affinities of these three families, and that there are cornbilled antong them all the lowest terms of stiuctural detail met with in the whole Order, this point of agreement is the more welcome and suggestive.

Examination of the figs. which we append will show that in the two first-named families there is a tendency towards an increase ill

size of the 4th carpal aiid consequent displacement of the 3rd one ( c j . figs. 5, 7, and 18). The two, however, lie, in all, practically in a line with the 3rd and 4th metacarpals-absolutely so in the

It is usually of fair size, becoming well ossified.

154 [Mar. 6,

Aylossa. The representatives of that family retain then, in this feature of their organization, a more primitive condition than do any other living Anura.

Carpale &-A11 writers have, up to the present, regarded the large element marked k as the fifth carpal. In the Discoglosside, Pelobafidre, and in Xenopus, this is large and distinct; and seeing that in the last-named genus the head of the 5th metacarpal (5’ , fig’ 4) is especially excavated to receive it, the above-named determination would appear to be unassailable.

In the two first-named families (Discoylosside and Pelobatida) this element has no connection with the metacarpal, for in them there runs, from the postaxial border of the 4th carpal to the epiphysial cartilave of the 5th metacarpal, a ligamentous band (” yigs. 5, 7, and 15). %his structure courses rentrallp to the distal face of the supposed 5th carpal (kj, excluding that from direct connection with the adjacent metacarpal.

The absolute identity of this band with the ligament in the hind foot, admitted by all to represent the missing tarsalia 4 (? 5), is irrerntable; and, if argument from analogy is to be trusted, this strncture may justly be looked upon as a 5th carpal. If so, the older determination falls to the ground, and the element k can no longer be regarded as distal. We are in a position to snbstantiate this, for the ligament in question carries, in Xenophrys, a well- defined nodule of hyaline cartilage which ossifies as nge advances (5, fig. 16). This ligament is early differentiated, and comparison of t h e higher forms suggests that it represents something which, in them, hRs heen lost.

W e trust thus to have shown that the 5th carpal is practically absent in the Anura as an order, bnt that a vestige of it exists in the adults of the Discogfossidcz and Pelobatidm (in precisely those forms, that is, in which we mig!it have expected to find it), becoming ossified and attaining its most fnll development, so far as is a t present known, in the genus Xenophrys.

c. Carpale 1 and Pollex.--The pollex is, in all known Anura, relatively small and unimportant. It attains its maximum length in certain American Tree-Frogs (cf. Cope, 14, and Boulenger, 7, pp. 338, 339), and among the Cystignathidce (cf. p. 170); in other forms, again, it is very short. Born says of it in Bombinator (6, p. 62), “einmal fehlte sie ganz, ein andermal waren sogar zwei Phalanpen vorhanden;” Gegenbaur omits it in his well-known figure of Bufo (18, pl. i. fig. l l ) , while Brocchi denies its existence in Hemiphractus (10, p. 16). W e strongly doubt the assertion of the first-named author, and we can only attribute it to the fart that he relied exclusively upon microscopical sections. Small the pollex may be, but absent rarely, so far as our experieuce goes.

I n all but some few forms the body of the pollex is i n articulation with a proximal cartilage, po i. (the “trapezium” of Ecker, 17, p. 53), of somewhat variable character. In its predominant shape, this element may be fitly compared with the bowl of a very thick spoon,

PROF. G. B. HOWES AND w. RIDEWOOD ON

Further comment is needless.


I n Pseudis tadpole and Ryla (figs. 26 & 20), less conspicuously i n Bilfo and others, it is elongated in the manner of a finger-joint. With the exception of Dugbs, who first regarded it (16) as a metacarpal, all investigators have looked upon it as the carpal of the pollex. Examination of its detailed relationships and consideration of its embryological condition (Blydes, fig. 9, PO. i . ) leave no room for doribtirig the accuracy of the latter determination.

Assume what form it may, the pollex usrially ossifies more early in the males than in the females. In the Ranidre, for example, it remains in the females for a considerable time simple and unossified (woodcut, fig. B, p. 174). This precocious ossification in the male sex is clearly the result of its connection with the well- known ‘‘ thumb-pad,” for it supports that more or less extensively in different genera and species. Iu some forms there is a great tendency for it to enter into direct connrction with the 2nd metacarpal, either by simple appositinn or by fusion with the warted crest of the same. In the male Discoglossus, however, the reverse is the case, for it there appears (fig. I;) to have undergone an increase in breadth directly proportionate to its diminution i n length. Ex- amination of the living auimal shows that this change is an accompaniment of one undergone by the ‘‘ thumb-pad ; ” tor that is, as it were, duplicated, its lesser representative being alone supported by the pollex.

In the youngest individuals esamined by us (e. g. Ranide and Alytes) the polleu had the form of one elongated unsegmented cartilage. It ossifies as a single element in all Ranida, Bnfonida, arid Discoglossidre which we have manipulated. On the other hand, the like is true only of certain genera in the Cystignathide, Pelohutidc, and rIylidre.

I n all forms eaainiiied by us, not cited above, it had become more or less thoroughly segmented into two or three pieces ( p a ii. to PO. iv. of figs.). These, in some of the Tree-Frogs especially, take on the characters of a metacarpal twice as long as broad arid two short phalanges. In them and certain other forms it undergoes an elongation a t the same time ; in yet others the reverse is the case. I t must not, however, be imagined that a coilverse line of modification is her? set up, for Leptodactytus (fig. 25), Pelodytes (fig. 1 l), Ceratophrys, and Callula d l show that, wliile shortened up, it is still segmented into distinct joints ; and these, moreorer, subsequently undergo complete and indeperident ossification.

When ossification sets in without previous segmentation, it does not spread from one or more definite centres, but i t is diffuse and irregular.

I n the males of Rhombophryne (fig. 32) and Lymnodynastes (fig. 28) the pollex is quite calcariform. The characters of the rest of the member are, in both genera, such as to show, beyond doubt, that we have here to deal with extremely specialized forms.

d. Ulnare (PyramicZale).-Wiedersheim has shown’ that in Rana

’ ‘ Anat. d. Froschee,’ Abtli. ii. p. 79.

1% PROF. G . B. HOWES AND w. RIDEWOOD ON [Mar. 6,

esculenta the arteria brachialis sends u p a branch, which perforates the carpus ventro-dorsally, for distribution to the integument and parts adjacent. In some forms (ex. Disco~Zossus, figs. 5 , 6 , f . i . ) the bones of the prGxima1 iow become definitely notched at this point, enclosing that which we would term ’ a foramen intercaypi for passage of the artery. Born has dcmonstrated the same fact for certain Lacertilia (4, pp. 2-3) ; while Wiedershiem ’, Uaur (1, pl. i. fig. 4), and others have clone so, less conspicuously, for the Urodela. Examination of the several descriptions atid figures shows that the perforation invariably takes place betileen the postaxial (ulnare) and central (intermcdiuni) elenieiits of the proximal series.

I n certain Urodeles R conflueiice takes place between these skeletal units ; and when this is the case the uliiare and intermedium are invariably those which fuse, and in some of these cases the artery in question perforates the cotifliieut mass. Taking this fact into consideration it is clear that the ulnare is the only element which, under all circumstances, cau lie postaxially to the blood- vessel. Seeing, moreover, that the artery agrees in all investigated cases in every known detail of distrihution, it is obvious that, in it, we have a convenient landmark by which to estimate the morphological value o f t h e proximal carpals. This said, further arguments are unnecessary in order to show that the so called “ pyramidale” of Dug& and Ecker is the ulnare, as Gegenbaur showed and as all subsequent writers admit.

e . Lunatunz (Z.)-The extraordinary position in which the fore foot of maup living Anura is carried lias resulted in a rotation of the parts of the same upon the coalesced r d i u s and ulna, of such a nature that the distal extremity of the ulna comes to be directed towards the palrnar surface of the fore foot itself. This rotation is riiost marked in Dactylethra, IIyla, Ye/oCates, Psezzdis, and Raria ; and it is instructite to fiiid that the tadpoles of these genera do not exhibit it. The distortion restilting from it is such, that we olten found it impossible (ex. l’elobates, fig. 18) to obtain a comprehensive view of the carpus without first disarticulating the radius. The importance of this observation is by no means slight, for there can be little doubt but that we are to seek the clue to the remarkable delineations of some authors, to displacemelit of these parts, consequent upon their having flattened them out.

Prom what we have said (sup&) concerning the course of the arteria brachialis, it will be clear that the lunaturn cnu only represent the radiale or the intermedium-one or both-of the less modified forms. Gegenbaur was unable to make up his mind (18, 1’. 13) as to whether this bone represents a coiifluence of these two elements or a persistelit rndiale-the intermedium having disappeared. He further suggested the possibility that the last-named may have been absorbed by the radius and ulna during coalescence.

Born puts forward, in his latest cominunication ( G , p. 62), the startling supposition that the lunatum is the intermedium. He bases

’ Uniformly with the f. iizfecrtcini of the hind foot (11. 144). Morph Jahrb. vol. 2. (187G).

1888.1 THE CARPUS AND TARSUS OF THE ANURA. 1.57

his deduction upon appearances presented to him in sections of two “vierbeinigen Alyteslarven.” On p. 1.5s will be found a tull statement of our reasons for refusing to accept the main observation upon which this view it based ; consideration thereof will show that even if that be reliable, there are still insufficient grounds for assuming e vestigio that the lunatum is the intermedium.

When, to the above, we add that all attempts to discover a t any stage, in its proper place, a trace of a third proximal carpal element, have failed us, there is uothing left us but to record our belief that the lunatum represents the radiale alone, and that all traces of the intermedium have vanished from eveii the ontogenetic record (cf. hind foot, p. 145).

(5th carpal of earlier observers.) (k).- We have discussed this element sufficiently uiider the head of “carpal 5 ” (p. 15.1) to reiider further details needless. I t is present as a distinct element only in the lower families ; its great size and comparative uniformity are its most atribing features. If tlie mor- pliological value claimed by 11s (p. 151) for that new element which we interpret as the 5th carpal be accepted, this one, looked at from all possible standpoints, can only represent a centrale. We propose therefore to term it, in accordance with its position, the p o s t a d centrale, by way of distinction from the element next to be dealt with.

g. Nuvicdare (“ scaphoide ” of Dugis) (a.).-‘l’liis element is by far the nrost troublesoine in the whole carpus, if not iu carpus and tarsus taken collectively. I t may fu,e with other3 in a maiiiier to be descril~ed Inter, but in all the lower representatites of the Order it is permanently distinct. In position it is extremely vaiiable, except for the faet that it is prenxinl : in Leptodactylus (fig. 2.;) and Certatophys, for example, it lies in the proxirnnl row and srti- culntes directly upon the intlius ; in the Discoylosszdcz and Aglossc~, on tlie other hand, it is invari:il)ly central, being dispowd, 111 the rorrner family, side by side with tlie postaxial centrale (figs. 4, 5 , and 7), with which it seems to vie in size.

Born states ( 6 , p. 6 2 ) that in the two ‘‘ vierbeiiligen Alyteslarven” previously dlluded to he found “ inniitteu des Carpus em treies, wohl abgegrenztes Ihorpelc!ien von Iial bmondfurmiger Gestalt ; ” he regards this as the centrale, and adds “ bei zwei auderen Carpel1 war dasselbe rnit carpale 5 verwacheen L I I I ~ bildete an deinselbeii einen deutlich abgesetzten, znugenfiirrnigen k’oi tsatz 1 ” IIe relied, as we have before remailred, upon tlie section-cutter’s method alone, and considering that he makes this “ wichtige Fund ” the basis of a revolu- tionary redeteriniliation of the niorpliological value of’ certain leading ronstitueuts of the carpus (corning to regard the naviculare as a displaced radiale, a id the lunatum as the intermedium), he ought, in justice to himself, to have given drawings of a complete series of sections, in place of the solitary and somewhat diagrammatic one proffered (pl. 1. fig. 5 ) .

Born’s figure shows that ossification of the metacarpals had set in, and, from the circumstances of the case, it is logical to expect that, if

PROC. ZOOL. Soc.-1888, No. XII.

f. Postaxial Centrale

12

158 PROF. G . B. ROWES AND W. RIDEWOOD O N [Afar. 6,

his observation be reliahle, this supposed centrale ought to be forthcoming iu youuger speciinens as a distinct element. W e have exainiiied a large series of specimens, mauy of’ which were niucli yoiuiger t!iau aiiy wliicli passed througli Born’s hands, and they sliowecl without exception that the positioii of his “ centrale ” was occu~)ietl by the preaxial third of that element regarded by us as the postnxinl centralc ( k , fig. 9). We can coufidently assert that in no carpus uircler our Iiaiids does this show a trace of either segnientirtiou iiito two, choiidrification from two centres, or conflaence of distiiict eleirieiits ’.

In one spccinien we I i t n ~ r been able to detect a linear depression on the undcr surface of the cartilage, iicar t he point a t which Borii’s clisiiiemberment appears to have orcurred ; and we are strongly’ of opinioii that this groove was present in his specimen, and that the section represented by Iiiui iiassed tiiroiigh it. . Indeed, his secoiid assertion above cited sceins, i u itself, to coufiriri this belief.

Iri the adults of certain t‘orrm tlir naviculare may lie, as aIready said, iriorc or lcss coiiipletely iii the proximal row, pmixidly to the lrriintum and in inorc or less definite ~~rtic:ulatioii with tlic radius (c#: figs. (i, 18, 30). I n others it may be as fully rernorcd fi-om the latter (figs. 1 , 6, 7) . It therefore bccomes a cluestion of vital ir i~portmce :is to jvliich of these two conditions is the more prinii- ti\.?.

In examiuing n hrge series oflnrvcr: of the Edible Frog, measuring each about 20 mrn. froni snout to relit, we were iiot a little surlirisecl to firid that tlie 1i;iviculare varied greatly iu tlie extent to wliich it thus e i i i b r ~ c ~ , as it were, tlie 1uii:itUal. In thc atlult it comes to lie in tlie i~rosii i id row, iirnrlp iiiretiiig the radius (n , woodcut fig. U , p. 174). In a young Uiij’b of nbout 20 I X I I T I . it mas already in direct a ~ i i ~ ~ s i t i o i i with t l i i v boiie. 111 PQxz (I’late VII. fig. 11, Xezzopus (l’ltite VI I . fig. 4), and riiost I Iy l ide (Plate VII I . fib’. 20) it is strictly ceiitrnl ; in oiie oieniber of‘ tlie inst-n:iuiecl f m i l y iiowever (Noto trema) it shows a teiidcncv to liccoine ~ircixirn;il iu tlic adult.

l u the ntiiilt of Llon~binutor (l’latc 1711. fig. 7), and still iiiore so in that of Uis~~r1gloss2~s (iiig. (i), it sends u p a spur wliicli ap~troaches but does not iicnrly reach the ridius. Ai i i iea l to clevclo1)lrleilt shows (figs. 5 tk !I) that this s p u r is it 1:ite go\$ tli.

‘l’lic, coirtlition of’ clie parts in tlie M l o l i u f i d ~ e is dcxerviiig of special note. It will be secii that io tliosc Anura iii alricli tlie iiaviculare is riiost ceutrd i n positioii the distal cnrl);ils arc relatively t;rnall and reduced ; in X e ~ r o p ? r r ~ y s (fig. 14) aiitl PeloDrctes (fig. 1 S) these are, oil tlic coiitriiry, liiigsr autl inore nearly uuiform i i i size with tile otlier eleiiients of tlir carpiis, so much so iu tlie latter geiirrs tlint tlie carpus lias quite n Salarnandriiie aspect. Here too the iiavicirlare is p x i r n a l

O s s i h i i o i i of tliis elcineitt invariably proceeds fwn i one centre. lliis becontau h l l more iiiwliod in certain forius in which our postuxial

:’ ‘l’liere are otlier esccptiotinl 1)eculiiLritim about tliis c q i u s , for wliich we

2 i,

reiitmle is conflucnt w i t h otlier oleiiieiits ( c j : p. li iU).

find no paml!el in our own apcciinms.

1888.1 T H E CARPUS AND TARSUS OF THE ANURA. 159

(not strictly so in Pelobates), and, at the first glmce, we appear to be dealing with tlie least modified term iri the series ; comparison of uliich with the opposite one (ex. fig. 7) nould appear to sliow that thc i i a ~ iculare tJas primalily proximal, a i d that its central position has been assumed as the accompaniment of a reduction of the distal caipds.

If we admit that the Pelobatoid fore foot is tile more primitive, we are forced to tlie conclusion that the n:iviculiire belongs to the prmimal series; if we liold the Discoglosrid foot to be tlie iiiore ancestral, we imply that the suppression of the cnrpalin is not a specialized feature ’. In all those cases in whiclr we I i a b e obtained sufficiently young Iarvx!, the naricularc is central to begin with and free of‘tlie radius, its ascending process being of secondary origiu (ci’. Biscoylossus, larva and adult, figs. 5 & G). We r e g d this as conclusive evideiice that the naviculare does not bclong niorphologirally to the prosinial series, and hold that the condition of tlie Pelobatoitl limb iy a modified oiie, its iiavicwlare haxing become displaced, in all lildihood at a very early stage. Nor must the possibility be overlooked that, in those forms in which the naviculare is most completely in articulation with the radius, it may have becn to a large degree displaced as the direct result of enlargement of the second metacarpal (Leptodactylus, fig. 25), or cariied u p with the eiilarging po1:e.i (Peloclytes, fig. 11) (cf. p. 171). Thrse are our reasons for regarding the iiaviculare as a centrnl clernent, and we proposc to terni it, in accordance with its position, the prensial c c n t d e .

1 he carpus of the Auiir: t woaltl appcar, fiom the foregoing, to possess two eiioi nioiis ccntrali,i. L)nplicatioii or trililication ot tlie centiale is well lLnown in many liiiiip Uro(le1es. \Vie Uaur ( I ) hate p rid most atttliitio!i to this questioii, and tlie latter has adduced wcighty argrments ( I , pp. GS-69) in sopport of the view that such duplication is purely secondary. Tlie existence of but a solitary centrale in the lowest representatives of the Class (Amblystoinn, Banr, p. 40), the great rariatioii in the characters and assumed relationships of the superatidrd one, the occasional absence ot the same (Nec/zirzis, Cryptobmncicus, and others, pp. 20-ZS), ail point to the conclusion that in this duplication we have to deal with extreme specialization. ‘hrning now to tlre Aiinra, me innke bold to assert that tlie liind font of these animals is, except for its great elongation and cliilngcs incident theieou, if a:ytliing, less considerably niodified tliaii the aiiteiior member. What, we ask, could be more extraordiiiary than the psitioil ill which tlie living Xenopus cairies its fore foot? and may not the appearance of the secoiicl ceiitrale have been originally associated with some sucli specialization ? Taking all facts into consideration, wc submit tliat, as an

It is iiilere\ting to ob,ei.ve here th L t , wlicrens 111 Alytcs, Bond~amctor, and Diaco,q/os,~u~, the osiilio,itioii 01 the iiavicnlarc yrcccdes that of the c , t rpa l~~, in A>iiiiphjs cp. 4 14 tlic first to omlfy.

A ‘‘ IJeber clie Vcriiiehi ling des 0 s eentralu iru C a r p s nud T a w s des Axolotls,” Morph. Jahrb. vol. vi. (1860) pp. 581-583.

, 7 l l ie problem is a very perplexinq oiie.

,,

For other references see Baur. 12”

160 PROP. G . B. HOWES AND w. RIDEWOOD ON [Mar. 6 ,

Order, the Anura would appear to be unique among all living animals in that they possess, without exception, a double centrale carpi.

11. Compound strzcctures resulting from the fusion of two or nzorc Carpal elements.-Under this head we have designated certain large bones, met with only in the adults of the more specialized families and in some few isolated genera.

There can be little doubt but that the great development of our postaxial centrale ( k ) is associated with the peculiar “ tread ” of the fore foot of the Anura and with the remarkable distortion which the parts of this limb have undcrgone. In the compound structures now to be dealt with, this postaxial centrale forms as it were a central nucleus, towards wliich one or more adjacent elements become drawn and with wYbich they amalgamate. The fact that the resulting compounds may occur independently in different families is sufficient, to show that they are, for the most part, of nouglit but physiological sigiiificance.

Capitutuwi.-\Ve suggest. this term for the simplest of the above- named structures. I t occurs in Yelodytes (fig. 11) and Pseudophryne (fig. 22j (1 kj, being formed, as shown, by the confluence of the pobtnxial ceiitrak a i d the 4th carpal.

Ca~itato-lLamccturn~.-This term was a p p l i d by E c l m (sw infk?) to a large bone which, in the Edible Frog ( 3 . 4. /c, fig. B, woodcnt, 11. l i - l ) , carries the 3rd, 4th, and 5th digits. I t is foruietl by the confluence of our capitatum with the third c a i p l .

We have observed it in tlie Raniclu?, Dendrobatide, Engystomatida, Cystigiiuthidc, Bufonide, IIylidcz, atid certain others ; it is not present in either the Discoglossidie, Pelobatidcz, or Aglossa. It often sends up, especially in old individuals, a process along the outer face of the ulna (ex. Leptodacfylus , fig. 25) for muscular attachment. Tlris lobe is absent iu tlic tadpole, and, from its function, it need hardly be said that it has iiotliing to clo with articulation upon the forearm 2 .

lhere sometimes runs along the under surface of this bone a longitutlinal groove, wliich terminates posteriorly, in a line wit.11 tlie interspice between metncarrials 3 and 4. 111 old specimens this lecorr:es conrertrcl into a tubular perfbration, which will admit a bristle and transmits tlic ramus later& of the uluar nerve ”.

On exarriiiiatiou of tlir adult carpus, doitbts might be reasoilably entertained as to whether both third and fonrtli cnrpalia were incor- porated in this piece. Apart, however, froni the f’act that in the young animal their boundary 1i:ies are tlefinable, there is a feature of some interest in its mode of’ attachment. In lryla ccerulea (fig. 20) and h‘ofotremn there still rernaiiis that ligament wliicli, iu the simpler forms, passes bctweeii the 4th cnrlml and the head of the 5th metacarpal ; true to its Ielationshi~is, it arises on a level with the head of metacarpal 4 (cf. A-enophrys, fig. 14j, and it differs from

Unifornily with Ecker’J cn~itnto-hninatuiu (17, p. 53). ’ AS might at first be irnagiiied, froiii Wieclersheiin’s fig. of;Lhfo (36, pl. 311.

t$ Eeksr and Wiederslieim, ‘Anat. cl. E’rosclies,’ part ii. p. 40 and fig. 14.

r .

fig. 177), in .rvhicli t,lris fbatnrc is grussly exaggerated.

1888.1 T H E CARPUS .4ND TARSUS OF T H E ANURA. 161

that of this genus only in that it holds the heads of metacp. 4 and 5 in apposition.

I n its mode of ossification the capitato-hamatum is variable ; in the Frog it ossifies in a diffuse manner, similarly to the unsegmented pollex ; in some Cystiplnathida ossification proceeds from one definite centre.

I n Rhombophryne (fig. 3 2 ) a still further fusion had taken place in both fore feet of the solitary specimen which we examined. This consisted, as will be seen, of a coalescence of all the central and distal elements, the two proximal ones alone remaining frer. I t will be shown in the sequel (p. 172) that wc have observed iii I’hr!~nisoi~s an unexpected range of individnal variation ill the fusion of adjacent elements, and we acccrdingly refrain from proposiiig a distinctive mane for this bone in Rhonibophryne, one individnnl only h*Lving been at our disposal.

Postaxiu2e.-We propose to institute this term for an anomalous compound observed only in P i p (fig. 1, p k ) . Its ascending process and all detailed relationships show that it can only hc the product of a fusion between the postaxial ccntrale and tlic ulnare. With respect to this fiisiorl Pka stands alone among all known Anura whose feet we have examined The Inuatiun c? radiale) (1 . ) has iindergone reduction consequent upon it, but w e do not lmow snficient of the habits of the animal to hazard a guess as to its functional significance.

111. SPECIAL PART. A . AGLGSSA.

U . PIl’ID,C. Examined :-

Pz$a americana : large ; feet of a second adult ; a young specimen 19 mm. from mouth to vent.

Fore Foot (Plate VII. figs. 1 & 2 ) . The radius and ulna are flat with knife-lihe cdges, and the dis-

position of these bones is exceptional, in that, unlike whst is scen in nearly all other Anura, the onter edge of the ulnais directed dorsally. As the result of tliis, the radius comes to lie in the plane of the extended hand, while the ulna lies above it.

The limb-skcletoii of this genns ans fiist described and figured

working (28, 11. 5 3 3 ) with wet speciinens, offered certain criticisms upon the olisrrvations of Rlrthel (2Y) and showed (11. 3 2 ) , w i t h perfect accuracy, thaL there are six clcments in the carpus. I h c later researches of -4. J. C. Mayer and others have added h i t little to his description. Carpalia 3 and 4 are p$”ctly dihtinct, and it is impossible to examine these and other elcuients withoiit being struck ’ Born figures (G, 111. 1. fig. 5) a. similar but mcomplete fii\ioii in the l n n a

of Aiyfca.

by fireyer (9). Later C. &lager, who confirmed Breyei’s obscrv a t‘ 1011s,

This we hnic iicvcr observed.


with the absence of that definiteness of articulation between them, met with in nearly all other Anura. P i p stands alone, as before stated (p. 161), with respect to the fusion of the postaxial carpal clemerits ; the resnlting bone (~7;) crrtainly includes the postaxial centrale and the ulnare, but we are unable to say whether our 5th carpale is or is not embodied. W e incline to the belief that it is not. Cornliarison of the young and adidt feet (fig. 2, h) shows that this compound undergoes a shortelling up during development. Its postaxial lobe (" of fig. 1 ) is of purely physiological significance and late in origiu.

Ventrally to the hinatum therc is present a large lenticular bone (indicated ljy a dotted line in fig. 1 ) . Baur has shown that, i n the Urodele hind font, an c+meiit of the tarsus may be thus visillle only on one surface, and hcnre we liave stnrlied with carc the relationships of this elernent. I t lies wholly ont of the plane of the trne carpal elcnients and is snriouncled by tendinons tissue, while i t is totally absent in the yo~ ing specimen at our disposal. We regard this combination of characters as prnnf that it is a sesamoid.

1 he elernent marked po. i. is somewhat interesting. I n the adult it is lenticiilar, and in articulatiou ait l i 110th tlie nericulare ( 1 2 ) and the hrad of the 2ncl metncilrpal ( 2 ' ) . It might thrrcfore appear, frorn its rclxtionships, to be tlic carpal of t1,e vcond di;it. In the young s~iccirnen, liowcr er, it is excrcdingly smid (fig. 2 ) and wliollv destitute of connection with tlic metacnrpnl ; on coni~~arison with "le~ioipus (fig. 4), in which the carpal of tlic 2nd digit is distinrt, it is clear t l i n t it call oiily r q m w n t the cnrpd of thc Imllex ', carpal 2 hakiug, in :dl ,proti,ibility, hrcoirie contlurnt with thc iiar icrilnrc. Eunmin:ition of the oiing ~ ~ ) e c i i i i r n (fig. 2 ) rwtler5 tliis tlonltfnl, for the Iirnil of tlic ?nil inet:tcnrpal b ~ a i s n:i ed:n gcnicnt, such as niiglrt concrit ably hare rrsultcbtl f i n m :I confluence with its carpal. Upon tlria poilit mc 1i:we been unnble to obtain dccibive evidence.

r ,

The most complete observations hitherto made upon the tarsus of this genus are those of Mayer (28). 'l'lir liincl foot is, like its fellow in front, wry attomalous. T h e astragalus arid calcaiie~~s are widely separated a i d fuaed a t thrir ends, leaiing an elliptical hole in the iniildle ; the coinpomd thus a i i ~ i n g is only twice as long as broad. Tlic onter edges of the t n o bows-i. e . the preaxial border of tlle astragalus and the posta.;inl one of the ca1caiicus-are ridged in such a manner tliat each hone is T-shaped in ti ansverse section.

The 4th and 5th 1i:etataisals articulate directly npou the epipl~y- sin1 elid ot the calrnncuq, while tlie retiiaining three are as it wete shut out fioin tliat of the :istrag:ilns by tliin lanieila of I)one. TIiese 1,iniftIlz :ire sc rn tc l ~ e contimatioui of two out of three Iiones IiicIi lie o i l tlie T enti ;il .itlc or tlie t a r w - i n e t n t a i d joint, niid \ + o d d nl)l'car

1888.1 THE: CARPUS AND TARSUS OF THE ANURA. 163

at first sight to be sesamoids (they were descrihed as such by Mayer 28). The postaxial one lies altogether free of the tarsus, and it is beyond doubt, as Gegenbaur has asserted (18, p. GG), a sesamoid. The lamina of bone derivative of the middle one has all the relations of a reduced cuboideum of such an animal as the Frog ; and wc can only interpret the bone which gives origin to it as that structure, thrust out of position hy the great development of the astralagus and calcaneus, probably as an accornpminient of the great elongation of the metatarsals.

The third or preaxial bone of the above-named series is clearly the iiaviculare. Firstly, it gives iiisertioii to the tendon of the add. longus digiti I. muscle, which feature we have slioivir (p. 1 4 H ) to be characteristic of that elemelit ; secondly, it carries a dimiiiutive calcar of two short but ossified segments.

The young specimen exanlined shows no trace of distinct halluu- tarsal, nor have we been able to obtaiu any clue to its whereabouts.

6. DACTYLETHRIDB:. Exnmiiicd :-

Xenopua lawis : three adults ; young tadpole. A7enopzis calcawtus : hind half ot‘ a late tadpole.

Fore Foot (Plate VII. fig. 4). This carpus presents fewer anomalies tlian that of P i p . The

ulnare ( p ) is quite distinct and in 110 sort of fusion \$it11 the postaxial celltiale (k). Our fifth carpal i5 wnutiiig, but those of the four remniiiing digits itre large atid distinct. The iiaviculare (n), is in its large size and in the possession of a preaxial sliur, quite d i 6 e that of Y i p .

As iii Z’ipn, the metacarpals are greatly eloiigntetl ; this geiiris is exceptional, however, in the great cxpansion of the head of tlie 4th one (4’) and the excavation of thilt of the 5th for the reception of a special peg of the postaxinl ceritiale (X).

The lenticular sesamoid (indicated in a dotted h ie ) lies, in this genus, in the liiie of junction between the radius and ulna, instead of being preaxial as in Pipa.

Hind Foot (Plate VII. fig. 3). The astragalus and cnlcaneiis are moch more normal than in

pipa, their extremities are very broad (fig. 3) , and the distal epiphysial cartilage ossifies from three independent ceiiti es (sce fig. 3 ) .

There overlivs the head of the second metatarsal a sninll hone (1) ah ich is wedged in between it antl the centiale (nariculare n’) ; we regard this as the tarsnl of the JLU/ZUT, for reiisoiis slready given (p. 14F) in dealing with the Fartie coiiditiou in Bifi and the Tree- Frog. we accordingly hold to represeut tlie fused tarsalia of the 2nd xid 3rd digits.

The metatarsals are much eloiipted antl the head of the third oiie (3’) is exceptional for its great breadth; seeing that the tarsals are com-

The bone marked -

164 PROF. G . B. HOWES AND w. RIDEWOOD ON [Mar. 6,

planate with these and not displaced ventrally, it follows that the hind foot is here, like its fellow in front, much more normal than that of Pipa.

The calcar consists of three slender elements, the proximal one of which can only represent a tarsal.

B. PIIANEROGLOSSA.. a. DISCOGLOSSIDB.

Examined :- Discoglossuspictus : adult 8, a young specimen, and a tadpole. Alytes obstetkans : Bo?nnbinator igneus : ] large series of adults and tadpoles.

Fore Ifhot (figs. 5 , 6, 7, 9). Born's latest commiinicatiou upon the fore foot in A2ytes (6)

would be little short of revolut,ionary could his deductioiis be sub- stantiated. We have already fully discussed these (p. 157), giving our reasons for rejecting tliern in toto.

The general plan of the carpus in these three genera is very similar ; in all, the largest clement of the series is our postasiiil centrale (A), a i d it is in this that ossification first commences. Gegenbaur has already called nttention to tlie fact that in Bortzbinator it is " iiiclit gar scliwwr " to see in the naviculare ( n ) the centrale of the Urodela ; wc have already given our reasons for regarding this element ;IS a preasial centrale (13. 159).

Gcgenbaur s a y that i n Bomhinritor (18, p. 17) all the carpal elcrneiits are distinct ; we, howevcr, interpret the 5th carpale of all our predecessors as a postaxial cctitrale (1). 157). Concerning the remniiiiiig cnrpalia we ard at variance with Born ; he figures an individual dlytes larva in which carpale J appears to have fused, while still cart.ilngiiious, with cp. 5 (our postaxiitl ccntra1e)-this we have never ol~aerred. I Ie also states that iii Uonihinator a similar con- flireuce may occur between carpalia 2 a i d 3 while still cartilaginous ; and he remarks that these variations " auf cine Neigung zum Ueber- gang in die fiir Rann , B~tjb u. E . f. characteristisclie Carpiisform scliliesseri lassen." We have searclled most carefully for traces of any such fusion, with negative results ; c:irlialia I, 2 , 3 , and 4 beiog throughout distinct iu all our sprcimens. These eleirients remain for n long time unossificcl in ilfiytes (fig. 9), they o d y in order 1, 4, 3, 2 ; in the iiiterval, howews, between the ossification of our postaxial ceiitrale ilnd these carpilia, tlie ~invicnlnre, ulnare, and lunatuin ossify in the order unnied. The followiiig is the foraiula for the whole series :-

h.n .p .1 . 1 , 4 , 3 , 2 . The carpal of the 5th digit is represented in all three genera by

the ligamentous band previously described (p. 154) ; in Alytes this cnrricj a toleral)ly large c:irtilagiiious nodule, which cve regard as the vestige of the 5th carpale of the Urodela.

1888.1 THE CARPUS AND TARSUS OF THE ANURA. I65

According to Born (6, p. 62) the rudimentary pollex may, in Bom- binator, either bear two phalanges or be wholly ahsent. I n all our specimens we find that it bears two segments (PO. ii. St iii.), irrespective of its carpal (po. i.). I n Discoglossus and A y t e s there is but one segment other than- the carpal present : in the male DiscogZossus this becomes fully ossified and expanded.

During ossification of tlie leading elements in Discoglossus, a central deposit takes place in each (Z. n. fig. 5 ) : soon, however, a differentiation of the ossific centre into a superficial spongy annulus and a central denser core takes place (plc, fig. 5), leading up to the curious condition o f the adult bones (fig. 6) already noted (p. 143).

.Diwd Foot (figs. 8 & 10). Gegenbaur was in error in asserting (18, p. 60) that in Bombinator

the separation between the astragalus and calcaneus is complete. I n even the youngest larva examined by us these two elements had already become coiiflueiit, bearing a cornmon expanded epiphysial cartilage.

Born claims for tlie same genus (ti) a range of individual variatious in the tarsalia far exceeding anything which we have observed. I n all our specimens the tarsalia of digits 1 , 2, & 3 are distinct, that of the third being invariably the largest. Tarsalia 4 St 5 are represeiited by R ligament, which in illptes alone carries a nodule of liyaline ctirtilage (J). 145).

The naviculare (ceiitrale, n') is, in all, large, and so placed as to separate the hallnx-tarsal ( I ) from the astragalus, the epiphysial end of which is excavated so receive it. This element is always the first to ossify, the caIcar follows, tarsalia 1, 2, 3 remaining for a considerable time nnossified.

Tlie calcar never consists of more than two segments, and is in all small. Born, in opposition to Leydip, regards the solitary piece present in Alytes (fig. 10, yh. i.) as tlie metatarsal (our tarsal) + the phalanx of' tlic sixth toe. We have been unable to detect the presence, a t any stage, of a second segment ; we hold, therefore, that the element in question is really the hallux-tarsal.

6. PELOBATIDAL Examiued :-

Xenophyys inonticola : 3 specimens measuring respectively

PeloJytes punctutus : 3 adults aiitl a tadpole. Pelobutesjuscus : an adult d and one very young specimen.

Fore Foot (figs. 11, 14, 15, 16, IS).

70, 40, a i d 26 mm.

That which most characterizes the carpus in tliis family is the relatively large size of the four carpalia (cf. p. 133). The 5th carpal is represented in R Iigamcnt, the detailed rehtionsliips of which have been already described (p. 1.54). This, as will be seen in fig. 15, becomes suddeitly constricted at its point of insertion into

166 [Mar. 6,

the head of the outermost metacarpal, especially in Xenophrys, in the adult of wliich i t carries a distinct bony element ( 5 , fig. IS), rcgarded by us as a 5 th carpal (p. 1.54). I t is worthy of remark that in its mode of ossification this element conforms in every detail to that observed for the other capalia.

Born asscrts that he found (6, p. G3) in a Pelobates larva a centrale carpi, idcritical with that figured by him in the young of Alytes ; and he ey1)ressc.s his doubts as to whether the same does or does not regularly fuse with the 5th rarpal (our postexin1 centrale, k) iu the adults of hoth genera. I t must snfice to point out that we have not liere obserced the presence of that cartilage wliich he calls the ceutrale, and the statements alreadj made by us for Alytes ( 1 ) . 1.58) apply equally to this genus.

This Iircvnial ccntrale (navicularc, 72) may or may not reach the radius in individuals ot' a species. In the Pelobates figured (fig. 18) it is represented as ncnrly touching that bone, but in the other carpus i t did not rxteiid inore than halfway towards it. W e havc alrcady stated (1). 15s) our reasons for regirding this ascending lobe of the nnviculnre as purely secondary.

Extremes of moJification of f l i e pollcx are met with within the limits of this stiinll family. I n Pelobntes and Xenoplwys it consists, excluding its carpal (1'0. i.), of a small stFliturm structuie which is quite iinieanientrd ; ill I'cludytes, on the other liand, it is very large and snollcn (fig. 1 1 , p . i.-iii.) niid defikiitcly segmented. In all three geucra the iotntion o f t h e Iiniitl. upon the fusecl radius and ulna is V C I y ma1 hcd ; aiid i n Pelodytes thi5 reac1:es its mnxirnum, having golie on tliioiqli more t i inn n right angle (fig. 11). Comparison with the 0 t h two peiier,r susgcbts tliat the great developmelit of tlie ~)ollex may bc in soiiic w:iy nssocintcd Irith this distoi tion ; aild we are iiicliiied to asrrlbr to the bame cniise n distinct ossificatioii of tlie distal cxtretnity of tlie uliia here iriet with (* of fig. 11) '.

111 I'eZu<qtes tlie c:irpus i5 esceptional ainorjg those of this family, owing to the fusion of the postaxial centrale and fourth c a r p 1 to form one bone, the capiratum (4. A, fiz. 1 1 j.

PROF. G . B. nowm AND w. RIDEWOOD ON

1I:lind I'uot (figs. 12, I,'%, 17) . I t is interesting, in tlie light of the facts concerning the fore foot,

to note tliat the estrenies of modification undcrgone by the astragalus and calcaneus in the Anurn are exemplified in this family. In Pelod!jtes (fig. 12 j they itre coritluent for their whole lengths, much eloirgateci nod slender in the iriiddlc ; i i i I'elobntes they are relativcly shorter than ill :cay Frog which has come under our notice. Xeno- p h ! y s is, in respect to TI latire lengtli, iiiternicdinte between the two.

l ' l i e ca1r:ir (pre-l!nlluxj attailis, as is m-el l liiiorcn, its n:asinium ~ e v t ~ l o ~ ~ m c u t in this i'~11iiIy. ltoseiiliof first clescrilietl and figured

1 Tliis is i loi rerrkwl by tlir: i i lmw. a s night :ippe:ir 011 esnini~intioii of t,llc ligiirr ; it. 1ic.s. ii i li"*, ~cn t i* ;~ i ly t o tlir line ol' jruiclion bclwcn tl1;it boiie :,,~rrl tlir liiiiiiiiiiii. 11 Iwc: ih : i x i y niili g r d rc:idine::s, so I I I I I C ~ I so that we o r i ~ i ~ i a l l y iiii:igiiird i i t o bc :I disiiirc! clement).

1888.1 THE CARPUS AND TARSUS O F T H E A N U R A . 167

its skeleton (32), and Gegenbaur, Born, and Bayer (2) hare more recently rerlesrribed it. Born states that in the larva of I’eZobates it possesses two plialangcs (6, 11. >I))-tliat is to say, there was origiiially a second segnient atldctl to that which we regard as its metatarsal. While we have not seen this ourselves we are able to record the presence of siirh a plinlanx in I’elodytes ( p h . iii. fig. 13), attached at right anglcs to the nwtatarsal (PA. ii.). In Xenophrys the pre-lialluu is relatir ely bniall, consistiiig of a single piew (fig. 17, p.h) which ossifies latp.

?’he calcar of Pelobates is set nn a t right angles to the plane of tlie fhot, and, in tlislhcing it for examination side by side with atljncent part$, most obseners ha1 e liitlrerto clislodgrd tarsalia 1 and 2 irom their natural co~ii~c\ioiis. Born first desri ibetl the hallux- tnrsnl i n this animal (3, p. 446) ; and a e have alreadj7 sho~vn (1). 148) that its relationsliips to tlic basal srgmeiits of [lie lire-hnllux in Yelodyfes neces~itate a re\ ival of the oltlcr view coiiceriiing the niorplrology of the naviculare. In Xeizophrys the liallux-tarsal is bery larqe and lozenge-shaped, being wtdged in (fig. 17, 1 ) between the hcads of the 1st and 2nd metatarsals.

With respect to the reinaining tarsnlia we fintl, i i i this family, as mith tlie carpali;i, an iiiconqtaiicy. W11:lc in I’e‘elodytes m i d PrZobtrtes tarbalia 2 a i d 3 arc tluite cliotiiict, in X‘cnophr?ys (fig. 17, .’) they nnite to form tlic “ cnboidr~nni ” so well hnonn 111 tlie Cornmoii Prog. In oilrig specimens t l i i z 1)oiie is in a] ticnlatioil with the outer 11:ilf

of the head of nietatarc:il 2 , I J r i t i n the ntlult it bcc,,nies shortroecl up i n a :iiaiiitier suggestive ot J7j/hr ( 1 1 . I4( i & fiq. 19).

Tlir liqainentoas iepreseiitntivc of tlir I t11 x i i t l T,t [I tarsals is wcll iiiiiiLe(1 1 1 1 d l tlii ee gciiei \Ire i l l r iiii:iIile to ~ccoiicilc tlic tlcacriii- tlon i t n i l hcuirs of tliiz aiitl t l i r tlriiJ tarsal giteli by Diiyt~r (?), cithei wtl: each otlier or witli t l i r fret a t our disliohiil.

c. HYLID~T. Examined :-

I ly la peronii ; IT. j?eycineti ; ZI. lichpnata ; 2 ZI. eiriinflbii ; 5 IT. ccerulra ; 11. ulr’,oplci7ctuto ; 5 13. nrbowa, tadpoles.

Notofrema mnrszipiatuna. Ph y Uoniedusu hypnch ondrinlis : sin all specimen. f’hylloinedusu dacnicolor : large specinieii.

Fore Foot (fig. ‘LO). Tlie c1i:mcters of both fore and hind feet are very constant

throughout this family. Tlie rotation of the forc foot npon the fujecl ratliiis a n d riliin is, in all, very irrarhed. The navicnlare is central i n position iu all bnt f’h!/lZonicdiisa and il’o/otiw)ia, i:i wliich two genela it sends up a raclinl s l i w wliicli we lii1re sllown to be

l’he carl)din 3 :mil 4 ha\e coalesced tzith the postaxinl ccntralc to oeColldal.y (p. I,>!)).

168 PROF. G . B. HOWES AND w. RIDEWOOD ON [Mar. 6,

form a capitato-hamatum as in the conimon Frog (3,4, k, fig. 20), upon which digits 3, 4, and 5 articulate. This family is uuiyue iu respect to the retention, subsequent to thc incorporation of the 4th carpal in this bone, of that IigRrnent (* fig. 20) which, in the simple forms, represents the .Xh carpal (cf. p. 160). In Xenophrys (fig. 15) it will be seen that the fourth carpal is wedge-sliaped postero-externally, extending over the inner half of the head of the fourth metacarpal : it is from this point that the ligament in question arises in both &no- p J q s and HI&, a i d we hold this as proof coiiclnsive that the 4th carpal is, in the latter, represented in the manner described.

Carpalia 1 and 2 are both tlistiiict, the carpal of the polles being elongated and slightly constricted in tlie middle (po . i. fig. 20).

The pollev itself varies in degree of development ; its topo- graphical value has been discussed by Cope and others (14, p. 200). Structurally it coiisists essentially of one piece (excluding its carpal) wliicli may show traces of differentiation or constriction into two srgnieuts (PO. ii., fig. 20). In the Anierican Tree-Frogs with “ex- teriial pollex ” (fI!yfu d b o p n c t c r t a ) the parts differ oiily in being very broad.

IIirzd l h t (figs. 19, 21). As previously pointed out (p. 1-16), exnniinatioii of the adult

tarsus (fig. 19) would lead to the supposition that the tarsal of the hallus is absent, and that the two bones which are nearest related to the heads of the metatarsalia 2 and 3 represent their tarsals alone. In fact Gegenbnur (18, pl. 6. fig. 9) was led into this belief. Coril- p r i s o n of the larva sliows that this is not the case, but that the outer of the two elemenls, when it first appears, has a11 the characters and relationships of the “ cuboideum ” previously described in the Pe/obatirZm. This bone is present in all the EyZidm which we liarc examiner!, and it iz cliaiactcristic of this and all the succeediug families.

21, 111. 14. fig. 10) in all, and their epiphysial cartilage is generally enlarged for special articulation of the 4th metatarsal (cf. fig. 19); as the result of this, these boces come to be inclined at a considerable angle to tlie metatarsals.

The central portion of the epiphysis is ossified near the astragalar border (fig. IS) a t the point of attachment of a powerful ligament, which is inberted at its outer end into the wall of’ the articular capsule and cnrries a t its point of insertion a cartilaginoiis nodule. I n old specimens this may become ossified, and we believe, for reasons previously given (1). 1-16), that it is an adaptive structure of no morphologiczl inipoitance, liaiiiig nothing nliatever to do with the missing 5th tarsal.

‘l’lie navicalare is preseiit as a distinct element in all (d , figs. 19, 19 a), ljiiig in a line with the pie-lialluu taisal ( p h . i.). 111 I’hylfomdusa it is unusunlly small, being Inrely larger than tIie tarsal of the hallus.

,, 1 fie abtiagalus aiid calcaiieus are greatly elongated (cf. I-Iofmann,

There is no trace of it in the tadpole.

Cj: 11. 1 4 .

1888.1 THE CARPUS A N D TARSUS OF THE ANURA. 169

The pre-hallux is present iii all specinleiis exaniiiied by us I, and it consists in them of four segments-a proximal larger one or tarsal (ph . i.), a sinaller metatarsal, and t w o diminutive phalanges '. "lie retral prucess of the srcond or met:itarsal segment (" of fig. 19) is abscnt iii l'llyllonaeclusa (fig. 19 a). In an old specimen of Wyla Zichenatn, on the other hand, it was completely scgmentctl off on either side, forrniiio (woodcut, fig. A, 11. 151) the largest of' a series 9 of liodules which triiigcd the free border of the pre-hallus fold. We have a l r e d y discussed tbe probable significance of these.

d. BUFONIDX. Examined :-

B u f o vulgaris. Bufo v i d i s : large specimeii. Bufo calamita : H young. Bufo variegatus : young. 8 l'seua'ophryne h ' h o x i i .

Fore Foot (fig. 2 2 ) . The limb-skelelon of l h j o has been deicribecl by various authors ;

it was first figured by Urnndt : i d Ratzeburg (8),. and subsequently by Mayer (27). Gegenbaur, Owen, Uorli, and Wiedersheim have in turn dealt with this shcleton, and upon some of their observations we have already conimentcd.

I'seudophryne (fig. '22) is exceptional among the members of tliis family in tlie cliaracters of its carpus : in Bnfo the aaviculare ( n ) articulates directly with the radius, and lics in the proximal row, here it is comparatively small and central : in Bufo the 3rd and 4th carpalia and our postaxial centrale coxlescc to forni a capitnto- Iiamatum, as in the Couimoii Frog. 111 Z'seudophryne there are three carpals 1) ing to tlie inner side of the large cnpitatuin (-1 h), whence it fbllows that sp. 3 is distinct as in Pelodytes. The carpiis of Pseudophryne approairnntes more nearly towards that of Pe'elodytes than ally other genus yet examined; and considering that the Bufonidm are, with respcct to the coiifluence of their distal arid postaxial central eIemeiits, a stage iu advance of the Pelodatide, Pseudophryne is to them what Peelobates is to the latter.

Bind Foot (fig. 23) . W e can confirm Born's statements concerning the taraalia in this

family (3, p. 443) regarding the hoi~iologp of the element marlied (2, 3 ) with the tarsalia of the second and third digits, and we find that in old specimens this may beconie shortened-up in a manlier suggestive of the Tree-Frogs. We liave notliiiig to add to what we have already stated in coilfirmation of Gegenbaur's observations cow cerning the ligamentous 4th tarsal (cf. p. 145).

Leydig states that he wa\ unable to find its cartilages in H. n~6owu (Xi, '' Born has described and figured the details of ossification of these, subsequent to fusion, in a wry old speciineii of H. a~borea (6, pl. 1. figs. 1 and 2).

181).

I i 0 PROF. a. B. HOWES AND w. RIDPWOOD ON [Mar. 6,

Born records ( 6 , p. 5.5) individnal variations in the constitution of the pre-Idlux; we agree with him only so far as concerns tlie presence of three segments irree1)ectivc of oiir tarsal oiie (q f i I?. l'l!,). T h e terminal one of these is never recognizable iu tlie adult (fig. 2 3 ) , and we have failed to find it in I'sezidophryne.

e. CYSTIGNATHIDX. Examined :-

5 Limnorh~nast~s fnsrrianieirsis ; I;. clomnlis. L(yttoclacfyfus peiittrclnc~'!yliis . 140 m i r i .

L ~ p t o d a c t y l u s olhlribris : 2 3 mm. C'eratophiys ornata : 1 10 niin. Cera t opA rys urn eric, I ~ N . Chiroleptes aitstrnlzs. Crinia varia. 3 Pseudza pcii*ndo.m, tadpoles. 2 IJyfodes vut rt iii icensr s. Tefwiatobzus jelsLii, tadpole. Puludicofa bibronii. 4 Ileleioporus nlbopictrrtus.

Fore Foot (figs. 2.5, 26, arid 28). Gegenbaur (18) and IIofrriaiiii (21) haxe :illiidccl casually to the

linib- sheletoil of one or two gewra of tllis family ; but the rnost complete accoullt J e t publislied is that of Alayer (27).

The sheletoil of the fore fbot resembles, in 11s general aspect, that of the F ~ o g ; :mrl i n a fannly where three of t he eight carpal elements ate inwriably fuscd titgetlier (c:t~)it:ito-liniiiaturn, 3, 1 h) tliei P is little iooin for niot1itic:ition of any inor ~)hological illilmrtnncc. ! h e cai piis of ~iiiitiot!,/nftstes (fi;. 2 8 ) may be coiitenicutlv tal,en as it t jpe o l the whole fmiily. T h e nnricrilnre ( n ) lips nell towards tlie radius, heiiig prolonged foi naidr iiito :I tliiii edge. Tlte pollei c ~ i p a l is free and little inodified, c, irqing the pollex (PO. ii.). Tlns coiisists of :I single element which i.1, in the mde, shor cl-shal)ed and lieset by a Iioriiy inrestmelit, riincli iu tlie m:uiiler of the ralcar iii Pelo6ate.s or lleleiopoi-us.

T h e tollowinq are the niore iinpoi t n n t d r p i tiires from tlie above that a e hale ol)ser\ccl i i i the geiiera recoltled :-

Pseurlis (fig.. 2(j).--'l'Iie secoiid digit i i rcmnr1,ablc for its slender- iiess, while i t is closely related ti) tlie pullcbx, aud, with it, opposable. Tlris change is accoiiipaiiicd by :L coi iespoiiding iiicreakc iii c rlib19 of the head of the :ird nietacirp;il (i3'). 'i'h cai pal of the polley is greatly elongated, in cwess even of that which is seen in tlie IfyZida.

C P ~ A ~ ys.-'l'he nnviciilare liere lies in the proximal row, a i d o m broad surface for articnlation nit11 the radius. The carpal of the polles is stouter tliaii that of the second digit, and it bears two short, but well-defined and f'uily o-sifietl, segments '.

Leptodacty1u.s.-The naviculaie is here in full articulation with Tliis is nlio tlic OASC in PducZxolu.

1888.1 THE CARPUS AND TARSUS O F THE ANURA. 171

the racliris (fig. 2 5 ) . The pollex consists of two short segments, borne upou an equally insigificant carpal (PO. i.-iii.), tile wliole standing out a t a right angle to tlic long axis of the fore foot. 111 the male the head of tlie second metac:irpal is, :IS 3Iayer Iias sliown ( 2 i ) , inimerisely enlarged for support of the horripclaqier fig. 25).

I t would appear, a t first si,:lit, that tlie displacernciit of tlie naviculare might 1i;ive been the direct ontconie of' this enlargemcnt ; but if so, the reduction of the 2nd caqml to the condition of an : h o - lutely iusigiiiticaut iiodule ( 2 ) Imoiries unintelligible, for it certainly would appear that that. iiirist have resultcd froiii the saiiie ciiusc. We tiiid, iii the j-ouiig siicciincu at our disi'osal, that this is really tlie case, for in tlic :ibseiice of tlie clasper the 31111 mrt:icnrp:il is relativc4y larger. I t is iiitcrestiiig to note fnrtlicr, that while, in this speciiiieii, the iiaviculare does liot yuitc rcncli the radiiis, the Imllex is propor- tioiiately inuch s l outer tiinii in t h e adult. On consideration of' tlicje facts we iucline to tlie belief that the displaceineiit 0 1 ' the nxvicu1:ire was originally associated with t h e eii1:trgerneiit of the pollex, in R

riiiiiiiier siiirilar to what is seen i n I'eiodytes (tig. 1 I ), aud that with the redaction of the pollex its conuectioii with the radius 1)ec:inic lost, ouly to be resunled again on tlie great cnlargenient of the 2rid mr tacarpal.

In the male of Lini?iodynastes t h e prensi:il borticr of tlic secoud metacarpal bears an irregiiliir boiiy crest, like tlint of the Coniniuii Frog. We f i i id , Iiorver-r,r, that tlw pollcx newr liere fuses with this, as it may do in the last-iiamed geuus.

H i i z d Foot (figs. 2-1, 2 7 , and 29). 'l'he typical coiidition of the C'jc.tigri:itliid Iiiiicl foot is escniplitied

iir C~ys/zgnuthus niid I;iinnodyi,cistes (fig. ?!I), aiitl, as tlie iigurc sliows, it is, in g~weral, Frog-IikcA-i. e. tarsiili:~ 2 :iiicl 3 are fnsctl to lorin a single cuboidvuin (' ,"). III Leptochcfylus b i l k spliut is sliortciicd 1111, so that tlie i ~ i i l l u ~ - t a r ~ a l iiearly ecln:Lls it in siw.

In Ceiatophys the hrtlius-tarsal is :tbsent, Iiaviiig eitlicr i'usetl with tile naviculare or disappeared, as beliered by Born ( 3 , 1). 4-11> for Rm1u.

Tile ciilcar is, like tlic pollos, su1)jcet to no iiicmsidernble v:iriatioii, as might indeed be expected of :i i'iiniily wliosc ~irenibers lead sricli diverse riiodes of' life. Iri C'ei.iclophi,ys (fig. 24), C y ~ t i y ~ L h u s , arid Limiodyriustcs (fig. 29), we hiire ouly beeii nl)le to recogiiize two segnieiits, tlie b:isal oue of mliicli represeuts tlie t:irs:rl (ph . i.), while in Leptodnctyius the full complement of four pieces is reached.

I'seudis (fig. 2 1 ) bridges over the iiiterviii bcslweeii tliese two conditious, iu that a small first plialaux is present ( p , h . iii.). 111 this animal t h e niidclle or metatarsal segrnent greatly exceeds tlie basal one in leng~li, slid its proxiinn1 outer border is en!arged ; in Limnodymstes this enlargement becomes more obvious, leading up to the coiiditioii seen in Ceratophrys orizatn (fig. 24), where i t assutiies the form of a retral spur'.

1 This is vory fceble iii a silialler fipccimen of C'. aijleiin>/n exainiiicd.

172 PROF. G . B. HOWES AND w. RIDEWOOD O N [Mt-.r. 6,

In Cheiroleptes the proximal segment is by fw the stronger; while in Heleiopoi*us the second one is robust and forcibly reminds one of the calcar in Pclobates.

f. ENGYSTOMATIDX.. Examined :-

Rhinoderma darwini. Phr?jniscus lawis, varians, and cruci!jer, several. Brachycephalus ephippium. Callulu p ic ta ; C. buleata. Illicrohyla, ? sp., 2. Rhombophrpe testudo.

Fore Foot (figs. 30, 32) . The carpus in this family stands boldly out from that of all other

Anura, with respect to the extraordinary tendency towards fusion of more or fewer of its elements with each other or the adjacent metacarpals.

The naviculare shows a tendency to become proximal i n all ; in Phryniscvs (fig. 30) its ascendin; spur is strongly marked, nearly meeting the radius'. This bone is iiiiplicated in the simplest fusion obseryed (Xirrohyla) , becoming coilfluent with the liinatiim. It' our dcterminations of thc morphological value of the carpal elements are sound, it follows that in this almost unparalleled feature we have the precise converse of the unique phenomenon encountered in P i p (p. Ili2)-there, our postaxial ceutrale is in confluence with the proximal postaxial element ; here, tlie Imaxial one eilters into similar relatiorisliip uith the proximal prenaiiil bone.

Brtrc7~ycephahs stancls alone, anioiig all f o r m examined by us, in that carpale 2 early unites a ith its corresponding nirtacarpal.

The above simple cases fade into insignificaiice beside that of Zhonzbophryne ' (fig. 32). IIere, all four carpalia and our twe centralia have become welded into one great mass, interposed, as it were, between the luiiatum a i d ulnaie and the heads of the metacarpals. The boundary lilies of its originally distinct constituents may be, for the most part, folloaed.

Remarkable indeed is the condition met with in I'hryniscus, for here the variations are not even generically constant. In the two specimens of P. levis examined, all the parts were free (fig. SO), while in P. vrwians carpale 2 had fused ulith the iiaticulare. I n P. cruciger we found a unique state ofaffairs, carpals 1 & 2, metacarpals 1 & 2 , xiid the naviculare all being firmly anchylosed together. I t will be observed that here, as in Microhylu and less conspicuously in Brachycephalus, the whole tendency is towards imparting rigidity to the preaxial limb-border ; it is not surprising, therefore, to find the naviculare sending up a radial spur, and, in doing this, appearing to become proximal.

As already pointed ou t by Gegenbnur (18, 11. 17). One specimen only exaiiiinecl.


Attention has already been called by one of us ' to the existence of unexpected specific and even individual variations in the larynx of certain Ariura : and we anticipate that a similar range will sooner or later be demolistrated for those in the carpiis of this family, if of none other.

.The pollex consists, in all specimens with which we have dealt, of one piece, irrespective of its carpal. In Phiyniscus levis this structure recalls ( p a . ii. fig. SO), as does the ' haiirl ' generally, the condition of the Common Frog. 10 RhonzCophryne (6g. 3 2 ) it is broad and calcarlf'ortn.

Hind Foot (figs. 31 S: 33). I n this we meet with varititions little less surprising than

those of the fore foot. The astragialus and calcaneiis are in T'hryniscus Zrevis widely separated (a., c., fig. 31) ; in 1'. cruciger and P. vnrians they lie close togrther as i n the Tree-Frogs.

1 he hallnx-tarsal is variable in the evtrerne in its relations1iil)s. In Brachycephafus, Microhyla, and Rhoinhophryne (fig. 33 ') it is distinct and of fair size ; in Phi-yniscus cruciger it liad fused with the naviculare ; while in the solitary spxciineii ot 1'. variaiis examined it was conflueut with the os coboideurn (tarsalia 2 and 8).

The calcar consists, in its simplest condition (Brachycephnlus, Jlicrohylu, Phryniscus variuns, R hinoderma, Iz horn boldwyiw, fig. 3 3 ) of two elements. In I-'lw,t/niscus cruciyer these had anchylused, but in the young of P . lrevis figured (fig. 3 I ) four distinct eleriiciits werc cliscernible.

I t is worthy of remark tliat the anterior of the two ligaments which we record for the II!yZidm (1). l(i8, Plate VIII. fig. 19) re- appears here with its contained sesamoid cartilage.

r .

y. DEN DRO B ATIDB.

One specimen of Dendrobntes tinctorius W A S e m m i n d . but i ts carpus and tarsus correspond s3 closely with those of the Frog that i t is iieedless to describe them separately.

h. RANID& Examined :-

Rappia marmoratu. Nanrtophrys ceylonicu. Rhacophorus maculatus ; 1:. eques. Ixulus leucorfiiiius. Cornufer vitianus. Megalz.xaZus madagnscrrriensis. Rana Nlticola ; R. cyanophlyctis ; R . nrvrrlis. R n n n esculentu ; 12. teinporuria : a large series of specimens

There is a marked constancy in the carpus and tarsus of this I?. 2. S. 1887, pp. 401-601.

PROC. ZOOI,. s O C . - l 8 8 8 , NO. XIII.

a t all stages.

13

174 PROF. G. B. HOWES AND W. RIDEWOOD O N [ lk l r . 6,

family, and with t h e exception of a broadening of t he pollex in Cortzufer arid Rltncophorus, and an increase in size of t he proximal end of the pre-liallux metatarsal segment in the first-named genus, we have met with no imliortaut departure from what is seen in the Common Frog. As this animal is so largely in tlernantl for ordinary class use’, we append a somewhat detailed description of i ts limb- skeleton, in the l ight of our preceding observatioris.

.Fig. B.

Rairn tmporurirc, 9 , 54 rum. in leugtb. Lcft fore foot, tlorinl view, x 12. Tlp conlesred radius and ulna ( r . 16) are

For referenres see 11. 182. reprcsciitcd in ~ l r e n;itural position.

Fore Foot (Rana temporaria). Tn the adult carpus of this animal, six bones may be recognized,

all of which a re in thcir original positions \\it11 the exception of t he nat+ilare (n). This has, as already remarkecl (p. 158), iindergone clonpation a1111 disp1;icernent radially. W e were not a little sur- prised to find tha t there is considerable variation in the position of this rlemeiit in young Frogs, whose carpus and tarsus were still unossified. In a large series of specimens examined, measnriiig at

The drsrriptions given in the etrrilents’ manuals (TInuley & Mwtin’s ‘ E1t-m. J3ioIogy.’ hlnrsliirll’a ‘ Frog,’ n i i t l Mivnrt,’a ‘ Comnion Frog ’) are fhi- the uiiist pr~,i*t, ccinipilatioiis, nncl we cleein it unnecessary to criticiee tliem in clet,ail.

1888.1 THE CARPUS A N D TARSUS O F T H E ANURA. 175

most 18 to 20 mm. in length, it flanked from one-third to three- fourths of the preasial border of the Innaturn. In these, as in the adnlt (fig. 2), it is four-cornered as seen frorn the dorsal aspect ; in old Frogs i t becomes triangular and i t may fuse with the louatuin, in t he manlier already described (p. 172) fhr Uicrohy/a .

T h e Zunatrim (/.) articulates proximally with tlie epiphysial border of the radius, preasially with the iiavicnlare by a concave or flattened surtace, arid postaxially with the prosimal half of the inner border of the uluare (I)3rrarnidale)-l)etweeii i t and tlie distal half of this boiie the foramen iiitercarpi is euclosetl (cf. p. 156).

The ulnare (pyrarnidnZe) (p . ) is in articulation proximally with the ulna and distally with the large capitato-hamaturn ( 3 , 4 k) of the distal row ; the hiiider half of its preaxial border is free, as above stated. T h e articulations of this element are of a loose order with the exception of that associated with the cal)it~ato-linmatnm ; this bears antero-ventrally a projwting socket, which is excavated to receive a corresponding and prominent d0wngroat.h of the ulnare.

Looked at in situ, with the fore foot. in the esteiicled position, the Inuatnm is most dorsally extended, and the ulnare most ventrally S O . Examination of the parts thus shows that, while the rotation of the fore foot upon the radius and ulna has affected the articulation between these boues aiid the proximal carpals as a series, i t has also brought about a less coiispicuous rotatiou of the proxirnal upon the distal cHrpal eledents.

C'rc/,ifato-ltamatzinz ( 3 4 k).-It is needless t o recapitulate the grounds upon which we have come to regard this as a cornpoi~iitl of the 3rd and 4 th carpalia of previous workers with our postaxial centriile. Postaxially i t send3 u p a lobe which embraces the free horder of'ttie ulnare, much in the manner of t he secondary relation- shil) assumed between the naviculare and limnturn. I'roximo-inter- naIIy i t articulates with the naviciilare, lullaturn, arid uliiare, distally so with the metacarpals 3, 4, and 5 , by special contlylcs. This corn- pound transniits ventrally, in the adult, the lateral ranius of the tilnar nerve ; the foramen through wliich that passes is, in young animals, represented by an open groove (cf. p. 160).

Carpal of 2nd digit.-This element (*) varies considerably in shape, beiiig sometimes rectangular, a t others convex on all sides

his is more fully represented than might be imagined, on a knowledge of the great reduction ot' the pollex itself. I t is strongly convex externa!ly, and concave internally, tor articulation upou the carpal of the second digit and the naviculare respectively.

pdex.--This is invariably in a vestigiiil condition, consisting in the adlilt (see woodcut, fig. U j o t a single picce, excluding its carpal, with wliieh it is in defiiiite art,iculittioii by a facet. Applyiiig arguments previously adduced (p. Ids ) , we regard this structure as representative of the entire first digit, rather tlrari of i ts metacarpal aloue as is more geuerally aiitl empirically asserted.

iCletacarpnls.-In the male the mctacarpal of the second digit 13*

176 PROF. G . B. HOWES AND w. R I D E W O O D O N [Mar. 6,

becomes modified preaxially in connection with the thumb-pad. We have nothing to add to Leydig’s account of this feature (26). The heads of the Znd, 3rd, and 4th metacarpals are expanded and more or less cousiderahly excdvated, for articulation upon tlie carpaiia and the capitato-hamaturn. The head of the 5th metacarpal is, like its shaft, comparatively very slender, while it is so modified as to embrace the postero-external angle of the capitato-hamatum in the nianner of a pincers apparatus.

Hind Foot (Rana temporaria). The astragnlus and calcaneus (a, c.) are much elongate I , confluent

at their extreinities, coneave internally, and either circulur or el11ptic:d iu section. Their internal borders I)ound the limits of 01 igiii of the adductor loiigu8 digiti prinii muscle, the tendon of wliicli ( m . c ~ L . )

Fig. C.

/

Rciiza tpqmraria, 0 , 51 mm. in length. Left hind font, d o r d vicw, x IS. n u d . tendon of the adductor longas

digiti primi iiiiiscle. Other references at p. l@3.

is inserted into the naviculare (n’). With the full development of this there arises from the epiphysial carlilage of the astragalus a proinitierice wliich overlies the tendon I , and which ossifies with age ; i i i old individuals there passes between it and tlie calcaneus an niinnlar ligament.

Tarsalia 4 and 5.-Represented in ligament, which is uot recog- iiizable in adult specimens.

Tursaliu 2 and 3.-Invariably united to form the splint-like

CF Ecker (17, 1’. 12S> and fig. 91).

1888.1 T H E CARPUS AND TARSUS O F THE ANURA. 177

" cnboideum " (2, 3). This componnd is interposed between the astragalus and the heads of metatarsals 2 and 3.

Naviculare (centrule), n', and hrrllux-tarsaZ.-The former articii- lates obliquely upon the epiplrysial cartilage of the astragdtis ; its outer face bears two convexities of nearly equal size, which are separated by a shallow groove. The veiitral convexity furnislies the main articulation for the pre-hallnx (see woodcut, fic. C).

The laallus-tarsal ( I ) is generally, thouch not ii~v~irialily, present ; when recognizalile as a distinct elenleiit it assiimes the form of' a small iiodule of cartilage, lying buried up in a confused inass of ligamentous tissue, in the interspace betweeu the postaxial hinder border of the riaviculare, the head of the liallux-metatarsal, and the os cuboideum (cf. 1). 147). In tlie Bull Frog (11. pipirns) this cartilage ossifies, but we have never observed that to be tlie case in I t . temporarze.

Pre-hdlux (cnZcnr).-Iiorn admits the existence of from 2 to 4 pieces in this structure ( c - p. 149), wliile lie records ( 5 , p. 233) the preseuce of three such in 12. arvalis, in coiitradistinctioti to Leydig.

111 young P r o p of 20 millim. in length we fiud four distiuct segmen'ts of' liyaline cartilage. The first or tarsal segment (ph . i . ) articulates with the iiariculare ; the second or metatarsal oiie is the largest, and bears R small retral process ; of the tliird and I'oiirth or plialaiigeal segments, the latter undergoes degradation and is repieseiitccl ouly 111 ligariieiit i n the adult. The thicc perhistcut segments ultinintely underga ossificatioii and usually remain distinct.

IV. CONCLUSIONS. Fore Foot.

1. That tlie pyramidale represents the uloare, and that there is insuficieirt evideiicc upon which to base a filial deteriiiiiiatioii of the morphological xalue of the lunatum.

2. That the naviculare is a preasial centrale, aild that its connection with the radius is always secoudary.

3. That a vestigial 5th carpal is present in the adults of the Discoglossida and Pelobatidce.

4. ' rhat that element hitherto regarded as the 5th carpal is a postaxial centrale, and that the living Aiiura are unique, as an Order, in the invariable possession of two large ceritrdia carpi.

Hind Foot. 5 . That no traces of a third proximal tarsal element are forthcoming

at any stage in deyelopmellt ; and that the morphological value of the astragalus and calcaiieus has yet to be settled.

6 . That while the tarsalia of the 4th and 5th digits are often represented in lignmeiit, skeletal vestiges of the fourth one are torth- coniing iu the Discoglossida.

Pour nctually ; but it m i l l be rolueriibercd that hc regards our coi l t ido (naricuiare) as the lire-lrallux tnrsal.

1 i 8 PROF. G . B. HOWES A N D w. RIDEWOOD ON [blar. 6,

7 . That tlie naviculare is a centrale and not the basal segment of the prc-hallux.

8. That the pre-hallux never colisists of more than four pieces, and that it conforms to the structural requiretnents of a sixth digit.

0. That tbe outer free border of the pre-hallux inxg undergo a process of fragmentation, gi! ing rise to insigiiificant nodules of cartilage iiidistingiiislinbie from those for which, in certain other Yertebrata, the value oP lost rays lias been claimed ; an11 that the grounds up011 which this cliiim is based are ~insatisfactory.

10. That there is a tciidency towards loss of independence of the hallux-tarsal in the Uiscoylossida, PeloLatide, and some Ranida , that rIerncnt in thein rcni;iining cartilaginous and sniall, or fusing w i t h one of the adjacent elements.

11. T h n t I ’ i p is alone exceptional among living forms, in that the t11ii.d digit exceeds the fourth one in length.

Pcrusal ot’ the body of this paper will show that in all the higher families of iltirira there is a geucral teudency towards couflueilce of three or more o f the carpal elenients, but consideration of tlie fact tliat snch modifications by fusion may not be even geiierically constant, shakes our faitli in them as guides to affinity. The fict wliich staiitls out most coi~spicuously is that the least modified conditions of hoth fore a i d liiiid feet are most nearly romhinctl in the Uiscoylossidre. IIoclistetter lias recently sliown (20) that in Uairt!~iiiutor tlie posterior carditial veins are retained for life, in a slightly modified form, and his observation lias becn supplemented by one ol‘ us and extended ( 2 2 ) to Alytes a i d Discoglossus. Adding tlrc,se fitcts to tho-e so well Itnown co~iccriiing the vertebral and otlier clinractcrs ot’ this faniily, there can iio longer be aiiy doubt tliat its rireiiil~ers are, by far, the least modified of all living Ariura.

Fore. IT in d . I 3 l l l C digital fL;niula of the Anura is I:-

I’ll. 2 , 2, 3, 3 . 1’11. 2, 2, 3, 4, 3. ?‘he only other Amphibia of which we have any knowledge whose

pl!alanges apliroacli this in order of arraiigemcnt are the Stegoceplialia ot the Permiau. W e nnfortunately kiiow notliing of tlieir carpus and tarsus. Uaur has lately tabulated (I, 1). 6 3 ) the digital forrrrula of all known Uroclela, axid perusal of his tables wili show how comi)lrtrlp all tlie members of that order are, in this rebpect, modified as compared with the Auura. Tlius it is seen that wliile tlie lirnb- slicleton of tlie Frogs and Toads is specialized i n the extreme for jiliysiolagical purpose?, there is retaiacd in i t a ledirig morpliologiral feature which carries us back to soirie of the oldest kiiown represen- tatiTes of the class Ampliibia ; and we must look to the Stegocelrliaha tlierntelvcs or to soine closc,ly allied forms for the aircestors of tliese familiar creatures.

1 During the 1mssagc of tliese pages tlirougli ilie i i res~, Mr. Couleiiger lias crtlleil atteiition (see below 1~11. 204-2U6) to the fact tliat iu certain l i i u i i d c p a s~~perni~uiernry p l ~ l a r i x is iiitercalatecl between tlic pciiultiluate and tcriiiiual oiies of eitcli digit i n boili lore aud hind feet. The foririuln of tliese aniiuals is thus: &3,4,4. 3 , 3 , 4 , 6 , 4 .

1888.1 THE CARPUS A N D TARSUS O F THE A N U R A . 1;9

Walterstorff has recently published (37, 38) a11 eslianstive account o f certain fossil frogs, aiid of tlie l’alzobatraeliidz i n particular. 1Ie claims to hare discovered a series of co~ir~r~ctiiig-li~il~s between tlie Arcifern ( I ’ d o b a t i r h ) mid tlic Aglossn. I Ie Iias pointed to the equality in length of the 3rd aiitl 4th digits of tlre hind limb in certain of his sprciitieiis, aiid his deductions receive additioiial support from the fact tlint in some of those tlie third digit appears to hare even exceeded the fourth in lengtli, as we h v e S ~ I O M ~ I I to be the case in the living Pka.

1‘. LIST OF L ~ U T H O R I T I E S R

1. BAUR, G.-Beitriige zur Morpliogciiie des C:ni,piis iind l‘nr>ri$ der Vertebr:iten. 1 Tlieil. lhtriicliia. Jeua (1 888).

2. BAYER, F.-Ueber dns Skelett dcr l’clo11;ititlen. Abliniitll. d. liUngl. biiliniisclien Gesellscliaft, vol. xii. (1883-5‘4), pit’. 1- 23, Gerniaii abstract.

3. U ~ R N , G.-Die secliste Zehe der Biiuren. Morpli. JdirlJ. vol. i. (1875) pp. 135-453.

4. BoRiv, G.-Zuni carpus und Tarsus der Saurier. Ibid. 1 0 1 .

ii. (1876) pp. 1-26. 5 . BORN, O.-Ueber des Sliclett dcs ‘I’arsenliiiclters v. R n n a

Jusoa uiid It. arvcilis. Stzb. der naturwiss. Sect. d r r sclile- sisclien Gesrllsch. fiir vaterliind. Iiultnr (1879) pp. 232-234.

6. BORN, G.-Naclitriige z n “ Carpus und Tarsns.” h1orI)li. Jalirh. vol. vi. (1880) pp. 49-;8.

7. BOULENGISII, G. A-Catalogue of’ Uatracliia Salientia. Uritidi Aluseuni ( 1 882).

8. XILANUT Ji lta.l.z~nunc.-nlediziiiisclie Zoologie, 1701. i. Berlin

9. B~IEYER, F. G.-Observationcs Anat. circa fabricam Rnnrt: pi1iz.

10. X I ~ O C C H I , ptl .-Ileclierches sur l’ostdologie d’un Uatracien Anoure prnveliaut d u Ur6sil (I~ierniph~acti~s). Ann. Sci. Nnt. sdr. G (1877), art. 7.

1 1 . BKUHL, C . U.-Zootoniie aller Thierklassen. Atlas, pl. 30. Wien (1876).

12. Coi.Ic, E. I).--Hketcli of the primary gronps of Bntracliia Salientia.

13. COPL, E. 1).-On the Structure and 1)istributiorl of the Arci- ferous Snurn. Journ . Acad. of Nat. Sci. lJliiladelp1iia, vol. vi. ( ISCicj) pp. 67-1 12.

14. COPE, E. D.--On the families of the Ranifom Anura. Ibid. vol. vi. (1867) pp. 189-20G.

1.5. C u Y i E R , G.-ltecherclies Bur les ossemrns fossilrs. 4”le cdit. Tome x. Paris (1836), pp. 302-309. Atlas, pl. 24.

16. D u d s , A.-lleclicrclies sur l’ost6ologie et la nijologie dcs lh t rac ie i i s >i lenrs diffirens tigee.

17. E c K I , ~ , A.--I)ie Anntoniie des k’rosclics. 1’;u.t 1 . ~ruirswirk (18(i4).

(1829).

Berol ( 1 8 1 I ) .

Nat. Hist. Review, 1 7 0 1 . v. (lHCi5) 111). 97-120.

Paris ( 1 834).

180 PROF. G . B. HOWES AND w. RIDEWOOD ON [Mar. G,

18. GEGICNBAUR, C.-Untersuchungen zur vergleichd. Aiiatomie der i\’irbeIthiere. Erst& Heft. Carpus und Tarsus. Leipzig (1864).

19. GEGENUAUR, C.-Critical rernarks on Polydactility as Atavism (Traiislation). Journ. Auat. and Phys. vol. xvi. (1882) pp. G 15-622.

m.--Beitr. z. vergleichd. Anat. uiid Eutwick- luiigsgescli. des Veiienspstems bei Arnp1iil)ia uud Fische. Morpl i . Jalirb. vol. xiii. ( l H H 7 ) pp. 117-1 72.

21. I I O ~ ~ M A N N , C.IC.--Uroiul’s lilasseii uiid Orclnuiigeii des Thier- lteiclis. &I. vi. Aniphibien. (1873-78.)

22. HOTVES, G. B.--Nole on the Azygos Veius in t,he Anurous

23. IIUXLEY, T. 11.-Article “ Ampliibia,” Ihcyclopcedia Britan- riica (9th edit.), pp. 750-771.

2-1. &CHIIER, G.-Beitriige zur Ihniitiiiss des cnrpus und Tarsus. Berickit d. iinturforscli. Gescllschaft zu Freiburg, Ed. i.

25. L I c Y n i G , F.-Uel)er d. Bau des Zrhnen bei Batrnchierii u. d. Morpli. Jahrb. vol. ii. (1 876)

26. Limnrc, $‘.--Die Anuren Batrachier der deutscli. Fauna.

27. I \ ~ A Y E R , A. J. C.-Anndecteu f. vergleichd. Anatorriic.

118. I l A y E ~ ~ , C.-Beitriipe z. eiiier ariatomisch. ?tfonogr. d. R u m Verliancll. d. Rais. Leop.-Carol. Alrad. d. Naturforscli.

29. hlicc~ti~:~, J. F.-System der vergleichentl. Anatomie. Ii.

30. Rf~vaiw, ST. Gm.-Oii the Classification of the Anurous

31. OWI~:N, It.-Anat. of Vertebrates, vul. i. (1866). 32. ltos;ImHoF, A. J . ~ ~ O J ~ : S E L voru.-Historis Naturalis Raiiarurn

33. ScHumz, A. W. F.--l)e sceleto Bufouis p1niarum. Berol

34. VAN DmN.-Aiiat. Beschreibuag eiiier monstriisen sechs-

3.3. WEINLAND, 1). P.--Ueber cleu Beutelf‘roscli (Xotodelt,hys).

36. WIEnEI<sHE1M, I<.-Lehrbuch d. vergleichd. Anatomie d.

RsTomw, \v.-ueber fossile Friische insb. d. Genus Part 1. Jahrb. cl. uaturwissensch. Vereius

3s. WAL~*ERSTOKIIF, W.-lbid. Part 2. Jahrb. (1886) pp.

. Ampliibia. 1’. %. S. (1888) pp. 122-126.

Heft 4 (1886).

pp. 165-1 96.

Uon11 (1877).

Bedeutuiig tles Fersenliiiclters.

Bonn (lH35).

p i p . Boiiii, Bcl. xii. H e f t 2 (1825), pp. 527-5.53.

Tlieil, I. Al)tli., IIalle (1824), pp. 459-460, 487-490.

13;rtrachiaiis. 1’. Z. S. (1869) pp. 280-295.

nostrtitiuiii. Niirinberg (1758).

(1S32).

fiissigeii Wasser-Froschrs (12. cscu/eitta).

hliiller’s Arcliiv ( ISM), pp. 449-477.

Wirbelth. Zweite Buflage.

Palawbtrtivchus.

Leiden (, 1838).

Jena (ISSS), pp. 204-212.

zu Jlagdeburg ( l M ) , pp. 1-81.

83-159.

lSS8.l T H E CARPUS AND TARSUS O F T H E ANURA.

VI. EXPLANATION O F THE PLATES.

181

l n order to facilitate oomparison, all the figures are drawn in the same relative position. Each represents the dorsal surface of the lelt ineinber.

Uuless otherwise s t a t d , all tlie preparations are repreaeiitetl as transpnrcnt objecls. I n drawing theiu, we have stippled ossified 1):wt.s a i d left tlie caAi1- lages clear ; the effect thus obtaiiicd is inore natural than that esprcssecl in the more orthodox custom of stippling the latter.

Ligaments are indicated in black.

PLATE VII. Pig. 1. Pijm americcma, adult, 8. Left fore foot, from above. Drawn as an

opaqiic object. x 2;. 2. I’ipu uineric~~na, very young specimen. Left fore foot. OssiGcnliou

has not yet begun. 3. X e q m s Icevis, adult 9 . Left hind foot. x 4. 4. Xeno7ms Zrevis, adult 9 .

X 30.

Left fire fool. Elements slightly disarticu-

5. 6.

7. 8.

9.

10.

lated. x 5. Discoglossus picflcs, young specimen. Left for0 foot. x 10. Discoglossus pictiks, adult 8, 48 mm. in length.

Bombkmfor ig/zczbs, adult 9, 37 mni. in length. Bombinator Z;(/izezis, adult 9, 26 mm. in length.

AZ&s obstetricam, tadpole, 34 mm. in length from mouth to vent.

Left Pore foot. All The whit,e areas represent the elcmcilts are conl]~letoly ossified.

compact bonc, the dark ones spongy bone.

x 7.

x 4. Left fore foot. X 7.

Left hind foot.

Left fore foot. x 30. -

Left hiud foot. Tlie astragnlos and calcnneus and the nietatarsals linve alone begun to ossify. x 4.

- Alytcs ohsfctricnns, adult 9 , 40 ~nin . in lengtli.

PLATE VIII . 11. Pclodyf/fcs pzi.izcfrrtus, length 30 inin. Left fore foot. Tlie linncl is

rotated u111m tho radius and ulna through more tlian a riglit. angle ; 1.0 facilitate comparison with the otlier iigurce, tlic fuzed radiiis and ulna are represented as coniplctely disarticulated and flatteiicil out,. x 7.

19. PcZudyfcs puizctatzts, length 30 mm. 13. Pdodyfes puncfafz&s, outline sketch of preaxiul port,ioii of the tarsus of

the above preparation, to show the course of the ligameiit coimecling tlie twsalia of tlie hallnx and pre-liallus.

14. Xenophrgs rnoiblicok~, young specimen, 27 m n ~ . in length. Lcft fore foot. x 5.

16. S e n o p A ~ y s momticola, enlarged drtwing of the postaxial distal portion of fig. 14. The element li lias been slightly displaced, in order to sllow more fully tlie vestigial 5th carpal. x 10.

16. Xmophr,ys m o d c o l a , adult of 40 11101. length. To show ossification of t.he vestigial 5th carpal.

17. X m o p h i y s moiiticola, length 27 win. 18. Pelobatcs fuscus, adult d, 54 mm. in length.

Left hind foot. X 8.

x 16.

x 18. Left. hind foot. x 13.

Left fore foot. The fused radius and ulna hare been disarticulatcd from the lmnd and laid flat ; during life t>he radius articulates im the concaviiy seen at the proximal end of the lunatuln (l.).

19. Hyla cmrulea, adult 9 , 75 mni. in length. 1Da. I’hyllomedusn dacaicolm-. Calcar, shorn ing absence of the prosi~ually

directed process of its second segment. 30. Hyla ccerdca, adult F , 7 5 nmi. in length. 21. f l y m arborca, late tadliolc.

x 7 . Left hind foot. x 34.

x 4. Left fore foot. x 44.

Left hind foot. Ossification has liot yct begun. x 14.

29. l’sezidophryie bibronii, d, ?3 urn. iii length. Loft forc foot. x 12,

182 M R . R. B. SHARPE O N NEW [Mar. 6,

PLATS IX. 23. IJ'i!lb cnluniitrt, young spccimcn, 18 iiiiii. in length. Left hind font.

24.

25,

26.

27. 25.

29.

30. 31. 3. 33.

x 12.

x la. CemtolilLrys ornatn, young 9 , 140 mm. i n length. Left hind foot.

LryfotlucfyIus pz fnr lnc fy l zrs , large 8 specimen, 140 iiim. in lciigtli.

I'scudis porcidom, ttdpolo. length of tail 110 UIU. Left fore foot.

I'scudis, tadpole. Left Iiind foot. x 3,. Liiwx)rlyjtasfcs tcwsiiirieizsis, youiig 8, 38 ruin. long.

Liil2720<&~?2CSfCS tasiizaniemis, young 6, 38 rum. in longth.

Left fore foot. I)raivii as an opaque ob,jcct. Rut,. size.

0saific;rtioii lias not yet. begun. x (i.

Loft fore foot.

Left. h i d x (i.

x 5. x 4%.

FORE I'OCIT.;~. i., foramen intercarpi. k., postaxinl ccutrale. Z., luiistum (mdiale). q / ~ , risviculare (preaxial centrale). p . , ulnarc (pyrtiiniclale). 210. i., carpnl of pollex. pa. ii., ~nehcarpa l of pnllcx. po. iii., P O . iv., phalangrs of lwllcx. pa. I I . , uiisegmeiited e h n e u t probably representing P O . ii., po. iii., &. p o . iv. 2, 3, 4, 5, carpals of second, tliird, fourh , arid iil'th digits. 4 k , oleinelit (cupitnfnm) forrurd by the fusioii o l tho ca.rpa1 of tho fourth digit with our post~nxial ceiitrale. 3, 4 X., element formed by the fusion of the third and fourth carpals with our postaxid contrale (ctlyitcilo-~irnintll.171 of Ecker). wk. 1 to 4, element ( c a ~ p a - crufrale) fornrcd by the fusion of the distal and central carpal elements. pk, elerriant formed by t,he fusion of' thc ulnare with the postaxial ceutmlc.

&NII woT.-u., astragalus (libiale?). c., calcmeus (Gbulnre?). n'., iiuviculnre (centrale). 1 1 1 ~ ii., second segincnt (metatarsal) of pre-hallriu. ph. iii., ph . iv., third and fourtli segiiiciits (pIia1:iii- ges) of \we-halhx. 2, 3, cdmidcmn, formed by fusion of' second and tliird tarsnliu. l', i!',.3', 4', 5', iwtakaisiils of the first, seconcl, third fourth, and fifth digits.

r. , radius. fL., ulna. 2', 3', 4', 3, metacm-l>als.

iih. i., first, scgiilcllt (tarsal) of pm-hallux.

1, 2, 3, tarsals of tbe first, second, and third digits.

2. Descriptions of some new Species of Birds from the Islsrd of Guadalcsnar in the Solomon Archipelago, discovcrcd by Mr. c. M. Woodford. By I<. ~ O J I I U L N . K , S i IARrE,

P.L.S. &c. [ReceiIed Jnuuary 18, 1858.1

The Accipitres brought by Mr. Woodford from Guadalcnnar are extremely interesting, and 110 less than t h e e new species of Asiit7. are represented in his collection.

I propose to call them :-

ASTUR HOLOMELAS, Sp. n. Adult male. Elltirely black, with a slaty-grey gloss 011 the back,

rutlip, wiiiga, and tail, and decidedly more slaty on the under surface of the body : " bill slaty black ; cere and orbital skin yellow ; iris

Download - On the Carpus and Tarsus of the Anura

Top Related