(89) 



THE OEIGIN OF THE MAMMALIAN CARPUS AND 



TARSUS. 



By R. Broom, M.D.,B.Sa, C.M.Z.S., Victoria College, Stellenbosch. 



(Read May 25, 1904.) 



(Plate VII.) 



Few problems in biology are of greater interest than that of the 

 origin of mammals, and within the last thirty years much discussion 

 has been given to the problem. The majority of embryologists, from 

 the consideration of certain developmental conditions, have been led 

 to the conclusion that the mammals have been descended from some 

 Batrachian ancestor or some even more primitive form which may 

 have lived in Devonian times. Most palaeontologists, on the other 

 hand, have considered that in the Theriodont and Dicynodont reptiles 

 we have forms which are so strikingly mammal-like that the ancestral 

 mammal should be looked for either in one of these groups or in some 

 closely allied order. As recent investigations point pretty conclu- 

 sively to the fact of the mammalian skull being directly descended 

 from that of the Theriodont, it becomes important to see if the 

 evidence derived from other parts of the skeleton confirms that given 

 by the skull. Already the shoulder girdle of the Anomodont has 

 been shown to be essentially similar to that of the Monotreme (1), 

 while that of the Monotreme resembles closely that of the foetal 

 Marsupial (2). In the present paper I hope to show that the 

 evidence derived from the study of the carpus and tarsus confirms 

 that of the skull and shoulder girdle. Unfortunately the carpi and 

 tarsi of the fossil reptiles are much less perfectly known than most 

 other parts of the skeleton, since unless the various elements are 

 found in undisturbed position the evidence derived from them is not 

 of a very satisfactory nature. Still it is possible even at present to 

 show from the few well-preserved specimens known tKat the- 

 mammals must have been derived from some Synapsidan reptile, 



