82 PROCEEDINGS OP THE ACADEMY OF [1883. 



addition, the latter genus presents a rudiment of the posterior 

 inner tubercle, as is seen in Deltatherium. An explanation similar 

 to this is admitted by Mr. Mivart to apply to the cusps of the 

 inferior molar of Centetes. It remains to ascertain whether tlie 

 cusp in this genus, Ghrysochloris, etc., represents an intermediate 

 or not. 



Secondly, as regards the Taljjidse and Soricidae, where the 

 external Vs are well marked. If we examine the external cusps 

 in the genus Didelphys, we find that the posterior one becomes 

 gradually more anterior in its position, until on the second true 

 molar it stands largely above the interspace between the roots, in- 

 stead of over the posterior root. It will also be seen that the 

 anterior intermediate tubercle is distinct, and of insignificant 

 proportions, while the posterior intermediate is large and is 

 related to the posterior external, as is the apex of a V to its 

 anterior base. In this arrangement I conceive that we have an 

 explanation of the Y's of the Talpidse and Soricidse. The first 

 true molar of Scalops is a good deal like that of Didelphys, but 

 the anterior cusp is larger and there is no anterior intermediate 

 cusp, while the posterior external is of reduced size. The poste- 

 rior V is better developed than in Didelphys, but is composed in 

 the same way, of a posterior intermediate cusp, and a posterior 

 external with a posterior heel. These are united by stronger 

 ridges in Scalops, Condylura and Blarina, than in Didelphys. 

 On the second true molar in Scalops, a V represents the anterior 

 external cusp of the first true molar. Whether this Y has a con- 

 stitution like the posterior one, i. e., is composed of external and 

 intermediate cusps joined, is diflflcult to determine ; but it is prob- 

 ably so constituted. It seems to be pretty clearly the case in 

 Blarina, where the fourth premolar and first true molar may be 

 compared, with a resulting demonstration of the correctness of 

 this view. In Condylui'a, the Vs have become more developed 

 and the external cusps reduced, so that the analysis is more 

 difficult. 



This interpretation applied to Urotrichus and Galeopithecus 

 gives them quadrituberculate molars, not trituberculate, as deter- 

 mined by Mivart. Mystoniys is tritubercular. The intermediate 

 tubercles are present, but are imperfectly connected with the ex- 

 ternal, so that Vs are not developed {vide figures of Mivart and 

 Allman). This genus offers as much confirmation of the homology 



