TOOTH-GENESIS IN THE CANID^. 473 



definite order, and not that several cusps appear at first and 

 subsequently some become suppressed. 



(ii.) All tbe fossil Multitubercnlata have a specialized dental 

 formula witb numerically reduced incisors and no canines : con- 

 sequently I am unable to believe that the Carnivora and Insecti- 

 vora, with their full dental formulae, have been derived from 

 these. 



(iii.) That there is a progressive increase in the number of 

 cusps in both the Thooid and Alopecoid series of Dogs, in passing 

 upwards from the more primitive forms. 



(iv.) That the teeth of the deciduous dentition are more primi- 

 tive than those of the permanent and have fewer cusps. 



(v.) Goodrich remarks that " Multituberculate forms increase 

 in number the lower we search : " this increase, of which I am 

 unable to convince myself, must be very small, and the total 

 number found at present is greatly below that of non-multi- 

 tuberculate forms, and of equal, but not greater antiquity. 



Such forms as Otocyon, which I consider to be primitive in 

 the number of its teeth, have, I believe, secondarily acquired the 

 multituberculate condition. The Monotremes may be directly 

 descended from the Multituberculata, as may also such orders 

 as the Eodentia with their specialized dental formulge, though, 

 as I have not specially worked at this point, I am not in a 

 position to express a more definite opinion. 



Having thus seen that neither the Tritubercular nor the 

 Multitubercular theory satisfactorily explains the origin of the 

 teeth of the Carnivora, the question naturally arises. Is there 

 any alternative theory that may explain it ? I venture to think 

 there is. 



In the first place, the history of the development of the cusps 

 shows that the antero-external cone is the first to develop, 

 both in the upper and lower jaws, in all four of the orders to 

 which I have referred. This uniformity cannot be without signi- 

 ficance, and I think one must regard this cone as the representa- 

 tive of the primitive reptilian cone * : in other words, I would 

 regard the Paracoue and Protoconoid as homologous cusps, and 

 to avoid confusion I would term this the primary cone. The 



* This conclusion appears to be in accordance with the views of Winge (28 • 

 From his illustrations he appears to regard the Paracone as homologous with 

 the Protoconid, and the Metacone with the Hypoconid. 



