804 H. G. SEELEY ON NEW SPECIES OF 



exhibiting divided external nostrils, demonstrates that the genus 

 cannot be located in the Mononarialia. The only definition of the 

 order Theriodontia is that given by its illustrious founder in these 

 words * : — " Dentition of the carnivorous type ; incisors defined by 

 position, and divided from the molars by a large laniariform canine 

 on each side of both upper and lower jaws, the lower canine crossing 

 in front of the upper as in Mammalia." But since Procolophon 

 possesses no canine, and exhibits no modification in the dental series 

 of any importance from front to back of the jaw, it is obvious that 

 the genus does not conform to the dental definition of the Therio- 

 dontia, even if an ordinal group could be founded on such characters 

 as have been cited. Before inquiring whether the Theriodontia can 

 be defined so as to include Procolophon, I would remark on the very 

 small value which can be attached to modified form of teeth as an 

 ordinal character. It is extremely rare to find that the teeth do 

 not change their form and proportions with position in the jaw. In 

 Lizards and Crocodiles these modifications so far parallel those of the 

 supposed Theriodonts as to be worth some consideration in searching 

 for the classificatory value of the mammalian type of teeth found in the 

 South- African fossils. In Hatteria and in a large number of lizard 

 types there are teeth which are more or less well defined by posi- 

 tion and altered form as incisors, premolars, and molars ; but the 

 canine attains no development, and sometimes the character is more 

 strongly marked in the lower than in the upper jaw. Yet it would 

 be altogether unjustifiable, in the present state of science, to see more 

 than an analogy between these lizards and corresponding ordinal 

 types of dentition among Mammalia ; for it would presuppose that 

 the teeth remained comparatively unchanged, while the remainder 

 of the organism scarcely retained a semblance of lizard character- 

 istics ; whereas it is known, from the evolution of all the orders of 

 Mammalia during Tertiary ages f , that the axial skeleton alters with 

 extreme slowness, while the rapid modifications of the mammalian 

 teeth are even more striking than the changes seen in bones of the 

 limbs. But, among the Crocodiles, I recognize in the well-known 

 wavy outline of the jaws a demarcation of teeth into regions which 

 have a fair right to be named incisors, canines, premolars, and 

 molars, and constitute a dentition as Theriodont in principle, but 

 not so specialized, as is seen in the South-African fossil group. 

 In the crocodile the regions are easily recognized by the form, size, 

 and characters of the tooth-sockets, when all the teeth are drawn, 

 especially in the lower jaw. The incisors occupy a flat or slightly 

 concave region, corresponding to the premaxillary bone. Then at 

 the head of the crest is the large canine placed between the pre- 

 maxillary and maxillary bones. Next succeeds a portion of jaw 

 with concave outline occupied by small teeth, which sometimes 

 become larger from before backward : these are the premolars. 

 And, lastly, there are teeth in another concave region which have 

 the position of molars ; these may, in the young animal, all be con- 



* ' Fossil Eeptilia of South Africa,' 4to, 1876, p. 15. 

 t See Graudry, ' Mammi feres Tertiaires,' 8yo, 1878. 



