1888.] THE CARPUS AND TARSUS OF THE ANURA. 153 



goes, as will be shown later, a by no means insignificant, range of 

 modification. In no case, however, is it ever segmented into more 

 than three pieces exclusive of its carpal ; it thus never bears more 

 than two phalanges. Tlie other four digits bear each, in order of 

 succession from within outwards, 2, 2, 3, 3 phalanges, certain Ranidse 

 excepted {cf. footnote, p. 1/8). 



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 "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 Ca}-pal Elements {Carpalid). — In the more specialized 

 forms certain of these unite with each other or with adjacent 

 elements (ex. Hyla, fig, 20, Rhombophryne, fig. 32) ; 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 five carpalia 

 are represented. While we admit that such may be the case, we 

 doubt, as the sequel will show, the homology of that element which 

 our predecessors, not excluding Born (6), interpret as the carpal of 

 the 5th digit. 



Carpale 2 {Trapeze of Duges ; Trapezoides 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 Alytes tadpole (fig. 9) shows this to be its primitive position. 



It is usually of fair size, becoming well ossified. In the Bisco- 

 glossidcB it undergoes a slight reduction (tigs. 6 & 7), and this reaches 

 its maximum, amounting to insignificance in the ^ of Leptodactylus 

 (fig. 25). Comparison of this carpus with the adult Discoylossus 

 (fig. 6) and the larval Alytes (fig. 9) is sufficient 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 Brachycephalus, 

 Pipa, and lihomboj^hryne. 



Caipalia 3 and 4. — In certain forms these two elements become 

 confluent, constituting the nucleus of a large bone (3, 4 k, figs. 

 25, 26, 28), called by Ecker (17, p. 53) the " capitato-hamatum." 

 In the DiscoylossidcB, Pelobatidce, and Aglossa, however, carpalia 3 

 and 4 are distinct. Considering the general affinities of these three 

 famiUes, and that there are combined among them all the lowest 

 terms of structural 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 in 

 size of the 4th carpal and consequent displacement of the 3rd one 

 {cf. figs. 5, 7, and 18). The two, however, lie, in all, practically 

 in a line with the 3rd and 4 th metacarpals — absolutely so in the 



