34 BULLETIN OF THE 



of generic divisions. All these parts, however, have recently heen found 

 to vary greatly, not only with age and sex, but in specimens of the same 

 age and sex. The form of the hinder edge of the palatines, as to 

 whether it be convex, truncate, or emarginate, has been especially 

 relied on for the distinction of both species and genera, yet the spe- 

 cimens before me show that in the same species, in skulls of equal age 

 and of the same sex, the posterior border of the palatines may be either 

 truncate or deeply emarginate. 



The situation and form of the molars also vary in a similar way, as 

 does also the depression of the palate. The general form of the skull 

 varies greatly in adults of the same sex, as shown by specimens of adult 

 males of each of the three North Pacific species now before me; so 

 much so, indeed, as to materially alter the relative proportions of the 

 different regions. The form of the frontal region, or third segment of 

 the skull, is especially liable to great variation, as indicated by the two 

 male skulls of Callorhinus ursinus figured in Plate II (Figs. 1 and 2). 

 Two skulls of the Zalophus Gillespii, received too late for illustration, 

 show much greater differences in this respect than these do. They close- 

 ly resemble in relative size and form the two adult male skulls of the 

 same species figured in the Fauna Japonica (Mamm., PI. XXII, Figs. 

 1 -4). In the figures of these skulls, as seen from above (Fig. 2 and 

 3, 1. c, Fauna Japon.), these differences are very strikingly shown. 

 Through the deep and abrupt postorbital constriction of the .skull, the 

 latero-anterior angles of the brain-case are sometimes well developed, 

 whilst in other specimens of the same species, age, and sex, through 

 the less abruptness of this constriction, they are either but slightly 

 prominent or obsolete. These differences give in one instance a quad- 

 rate form to the brain-case, and in the other a triangular form. The 

 length of the postorbital cylinder of the skull is also an exceedingly 

 variable (dement, the difference amounting in some cases to nearly 

 thirty per cent, and hence greatly changes the general form of the 

 skull. 



The great degree of asymmetry exhibited by these animals may be 

 also cited as evidence of an unusually great tendency to variation* 

 Further evidence of the same tendency is seen in the somewhat frequent 

 occurrence of supernumerary molars in the upper jaw, — instances of 

 which will be presently cited. 



* See remarks on this point beyond, under Ettmetopias Stdleri. 



