264 



Mr. E. S. Goodrich. 



limited below by a single bar, includes those reptiles which lead towards the 

 mammals, together with side branches. These synapsidan reptiles and the 

 Mammalia make up a monophyletic offshoot to which the name Therop- 

 sidan branch may be applied. Lastly, the majority of the reptilian orders 

 belong to the third group, culminating in the Aves, and in which two lateral 

 temporal foramina and two bars are developed. These are the Diapsidan 

 reptiles, and together with the Aves they form the great Sauropsidan 

 group. This attempt at a phylogenetic classification of the Amniota may 

 be expressed in diagrammatic form as shown in fig. 1. In these general 

 conclusions most zoologists would now concur ; it is only when we try to 

 assign certain families and orders to definite positions in the system that 

 serious differences of opinion arise. We can now pass to the evidence on 

 which these conclusions are founded. 



The Significance of the Metatarsals in Reptilian Phylogeny. 



On examining the skeleton of the hind foot of a typical Lacertilian, such 

 as the Iguana shown in fig. 2, C , it is at once seen that, whereas the first four 

 metatarsals are of the normal elongated and straight shape, the fifth meta- 

 tarsal is quite peculiar, and differs from the others in that it is shortened 

 and markedly hooked. The bent proximal end projects forwards (inwards), 

 and also extends farther proximally than the remaining metatarsals, passing 

 over the end of the fourth. This peculiar shape and disposition of the fifth 

 metatarsal has, of course, been often noticed by anatomists, but its significance 

 in classification seems not to have been fully appreciated. Much controversy 

 has taken place about the exact homology of the hook-shaped bone. Some 

 believe it to represent the modified fifth distal tarsal, others the fifth meta- 

 tarsal combined with its tarsal, and yet others that it represents the modified 

 fifth metatarsal only. This last interpretation is the one now generally 

 accepted, and is clearly shown in Sewertzoff's excellent account (31) of the 

 development of Ascalobotes (see fig. 2, D,E). The question of its homology 

 scarcely concerns us here ; the important point to establish for the purpose 

 of this paper is that this peculiarity is found in the hind foot not only of all 

 known Lacertilia, but also of all living reptiles. So far as I can ascertain, it 

 occurs in all Chelonia, Ehynchocephalia, Crocodilia, and Lacertilia, excepting, 

 of course, in those forms which have lost the hind limb. Examples are 

 shown in fig. 2. This particular specialisation of the fifth metatarsal is 

 therefore far more constant than the structure of the skull. Once 

 acquired it never seems to be lost, and it can still be clearly seen even in 

 the feet of the Chelonidse or the Pythonomorpha, which have become so 

 highly modified into swimming paddles. The hook-shaped metatarsal does 



