42 THE SALAMANDER 



which articulates with the ulnar condyle of the humerus and bears 

 an olecranon process is entirely cartilaginous. The central portion 

 of the shaft is very slender, and is roughly triangular in transverse 

 section, with a slight bony crest running along the flexor side. The 

 distal end is but slightly expanded, and articulates only with. the 

 ulnare + intermedium. 



To summarize: the radius and ulna are entirely separate. The 

 greatest expansion of the radius is distal, and in an axial direction, 

 while the ulna expands proximally, and in a vertical plane. Both 

 elements articulate with the humerus as well as with the carpus. 



(c) The Carpus (fig. 28) of the adult Salamander consists of 

 seven elements, namely, the radiale^ a fused ulnare -\-iniermedium, a 

 centrale^ and four basalia. Much has been written concerning the 

 homologies of these basal elements and of the four digits which they 

 bear. The older authors, e.g. Gegenbaur, Wiedersheim, and Hoff- 

 mann, supposed that the hand lacked the pre-axial digit, or pollex — 

 apparently by analogy with the normal reduction sequence in the 

 mammalian carpus. When, however, the carpus of the Salamander, 

 or of any other four-fingered Urodele, is compared with the tarsus, 

 which bears five toes, a strong similarity is to be observed in the pre- 

 axial sides of the two limbs. Further, in some forms, e.g. Triton and 

 Diemyctylus, there are evident traces of a reduction occurring in 

 the post-axial region of the tarsus (cf. Steiner, 1921). Both Perrin 

 and Sieglbaur have concluded, from a comparative study of the 

 muscles of the hand and foot of Salamandra, that the pre-axial digits 

 of each are serially homologous, and that therefore it is the post- 

 axial, or fifth finger, which is missing from the hand, and this is the 

 view adopted here. It should be mentioned that both Ribbing and 

 Eisler, in spite of having investigated the muscles and nerves in 

 detail, apparently retain the older view, although they do not directly 

 discuss the question. 



Turning now to the carpalia themselves, the most pre-axial of the 

 basal carpals (c.pr.p.) was considered by Gegenbaur, Wiedersheim, 

 and the earlier authors to be the basal of the second digit (they con- 

 sidered the first digit to be missing entirely) pushed a little out of its 

 place. Strasser (1879), however, by studying the development in 

 Triton and Salamandra, showed that primitively this cartilage has 

 no connexion with the pre-axial digit, and that the next basal (his 

 Basale commune) has a double origin and is always related, in the 

 larva as well as in the adult, to each of the first two digits. He therefore 

 concludes that this element represents basalia i and 2 fused together. 

 In 1886 Kehrer suggested that the preaxial cartilage above referred 



