76 Transactions of the South African Philosophical Society. 



in which the posterior nares open between the hinder molar teeth. 

 Whatever evidence, therefore, is afforded by the position of the 

 posterior nares is in favour of Tritylodon being a Mammal rather 

 than a Theriodont. 



One character observed by Seeley — the presence of a distinct 

 prefrontal bone — is of much more importance than any of those 

 previously mentioned in determining the affinities of Tritylodon. In 

 1894 he could state, " The presence of a prefrontal bone is a reptilian 

 character unknown among Mammals." But, in 1896, he discovered 

 that a prefrontal bone exists also in Omithorhynchus, a discovery 

 that has since been confirmed by van Bemmelen.* So that, though 

 the presence of a prefrontal bone would remove Tritylodon from 

 the Eutheria or the Metatheria, it would not remove it from the 

 Prototheria. 



On the other hand, we have the very important mammalian 

 characters pointed out by Owen. Of these, perhaps the most 

 important is the structure of the molar teeth. The molars have 

 rows of well-developed cusps, and have distinct roots. There is 

 no known Theriodont with either the one or the other, and both 

 characters are found in the molars of known Mammals. There is 

 one important point in connection with the molars, the bearing of 

 which has not,. I think, been fully recognised. In 1898 Osbornf 

 pointed out that in " typical Multituberculates like Tritylodon" " the 

 dental series are parallel with each other as an adaptation to the 

 forward and backward motion of the jaw." This conclusion is fully 

 justified. Even if we knew nothing of such animals as Meniscoessus 

 and Cimolomys, we could be quite certain from the arrangement of 

 the upper molar cusps that the lower molars must have had the 

 cusps also arranged in rows, and further that the lower molars must 

 have worked against the upper with an antero-posterior movement 

 as in Eodents. To admit of such movement the articulation must 

 have been of the mammalian type, as no antero-posterior movement 

 would be possible in a form having the Theriodont type of articula- 

 tion ; and as the different type of articulation is the only fundamental 

 point of difference between the Theriodont and the Mammal, it 

 follows that Tritylodon must have been a Mammal. 



The condition of the anterior nares in Tritylodon is another point 

 of interest. In all known Theriodonts in which the parts are 

 satisfactorily preserved the premaxillary bones send upwards a 

 median process which meets the nasals and divides the nares. In 



* J. F. van Bemmelen, " Der Schadelbau der Monotremen. " Zool. Forschangs- 

 reisen in Austr. u.d. Malay Archipel., 1901. 



f H. F. Osborn, " The Origin of the Mammalia." Amer. Nat., May, 1898. 



