On the Maims and Pes of Pareiasaurus. 357 



Russian specimens, shown that the formula is probably 2, 3, 3, 4, 3. 

 In Propappus two adjacent toes have more than 3 phalanges, and 

 I regarded the formula in it as probably 2, 3, 4, 5, 3. It is now 

 pretty certain that in Pareiasaurus the formula is 2, 3, 3, 4, 3. 

 It might be thought unlikely that two animals so nearly allied 

 could have a different formula, but in the Therocephalia we almost 

 see the change taking place from the 2, 3, 4, 5, 3 formula to the 

 2, 3, 3, 3, 3. In ^Elurosaurus, though the formula at first sight 

 appears to be 2, 3, 3, 3, 3, in reality it is 2, 3, 4, 5, 3, as there 

 is a small plate-like phalanx beyond the 1st phalanx of the 3rd 

 toe and two minute phalanges in the 4th toe. In Gale-pus and 

 Galechinis, and in Dicynodon the formula is certainly 2, 3, 3, 3, 3. 

 In Pareiasaurus I cannot detect any rudimentary phalanges. On 

 the present evidence I think one may conclude that Pareiasaurus 

 with its massive body and short toes represents a stage in the 

 reduction of the digital formula, while the lighter Propappus, though 

 so closely allied to Pareiasaurus, retains the more primitive formula 

 of 2, 3, 4, 5, 3. Williston has recently shown that the American 

 Cotylosaur Limnoscelis has a foot which is very Pareiasaurian 

 in type but which has the formula 2, 3, 4, 5, 3-4, and this is 

 also the formula of Procolophon. 



