96 



AMERICAN MESOZOIC MAMMALIA 



tiny cusp near the middle of the internal face. A curving ridge runs down the internal 

 face to this cuspule from the apex of the crown. P" is larger, has two roots, and has a 

 small posterior accessory cusp but otherwise closely resembles P^. 



P* exhibits the same general features — a main trenchant cusp with posterior ac- 

 cessory cusp, an internal cingulum, here very strong, rising to form a marked high 

 cusp just posterior to the middle of the tooth on its internal face, a small anterior cin- 

 gulum cusp, and a ridge running down from the apex of the main cusp to the internal 

 cusp. I n this tooth the latter is also followed by a small posterior accessory cusp. This 

 genus is the only one known from the Jurassic in which there is really any close com- 

 parison between the last upper premolar and the first molar. (See Fig. 36.) P^ has all 

 the elements of M^ and they are arranged in the same way. Nevertheless there is a 

 definite break between the two series. P' is much smaller than M\ its external cusps 

 are without furrows, its internal cusps relatively smaller, the anterior basin less de- 

 veloped, etc. It is, however, unique among Jurassic mammals in that it is patently 

 becoming molariform and that its parts are severally homologous with those of the 

 molars.^' 



Exterior 



_ YPM-I0t^7 



Interior 



Fig. 40. Docodon superus. Right M*'", crown view, type. I2j^ times natural size. 



The pattern is the same for all the molars save for differences in proportions. M' 

 is the largest and least oblique. Anterior to it the molars are narrower anteriorly than 

 posteriorly and posterior to it the reverse is the case. The crown of M* may be thought 

 of as divided into two parts, one external and one internal. The external part consists 

 chiefly of one large cusp, the highest and strongest on the tooth, which is sharply 

 pointed and compressed transversely but not truly trenchant. Posterior to it is an 

 accessory cusp of about half its height. Both of these are skirted externally by a sharp 

 cingulum which rises anterior to the main cusp to form a distinct basal cingulum cusp. 



^^ The belief is current that mammals were primitively homodont, within the Class as such, and 

 that the Jurassic mammals had departed but little from this condition, having a graded series with slow 

 transition from P} to the last molar. How false the supposed evidence drawn from the Jurassic forms 

 has been, should already be evident to the reader and hardly needs emphasis. Having in mind the 

 characters of the order Pantotheria as a whole, it seems justifiable to say that the incipient molarization 

 of P' in Docodon is one of its numerous specializations, and not a primitive feature retained by this 

 genus and lost by all other pantotheres. The subject, however, more properly belongs in a treatment of 

 molar evolution than here. 



