^8 Soiitlieni Cross. 



ngnin rontril»ules larj,'t'ly. On the contrary, there is no instance of a 



douMc-roottMl p.jn. . :m.l <.l" ]..ni. - it may he said that in all 



tiunes when» this t(H)th was lonnd hi situ, it also was single-rooted. 

 In n single c-aso (Xo. 4). where all the teoth had been removed from 

 the skull, and could not be certainly identified, I suspect it to have 

 lnH'n ijttubh'-riioted. 



(Xll.) Besides variation in numl)er and shape there is also 

 variation in size. The teeth of No. 807 are abnormally small; those 

 of Xo. W abnormally massive, and crowded together in the jaw. 



(XIII.) Further, the size of the incisors is conspicuously variable. 

 Its exhibited by the large incisors of Xo. 324& and the small ones 

 of Xo. SOT. 



Having thus described the variations to which the teeth of the 

 known S|)ecinien3 of Ommatophoca are subject, it is time to turn to 

 what has l)een written on the subject by Mr. Bateson and Dr. 

 Kukenthal. In fairness to these writers it should be at once stated 

 lliat Mr. Hateson's remarks were based upon an examination of only 

 two skulls — those brought home by Ross, — while, so far as 1 know, 

 l)r. Kukentlial never had an opportunity of seeing the actual 

 s|>e«iiuen9, hut based his conclusions on mere descriptions and upon 

 Mr. Hateson's arguments. It will not then appear surprising, if I 

 find myself, after the advantages of examining no less than eight 

 .skulls, unal»le to agi*ee with all that has been written on the subject. 



To dwd fii-st with Mr. Bateson. That naturalist has found in the 

 variations of skull Xo. ?>2Ah the material for a highly ingenious 

 l>ai»er, wherein he has used them in conjunction with other like 

 variations as a ram wherewith to batter the prevailing views on the 

 homologies (»f mammalian teeth. 



Am tlie result of an examination of great numbers of skulls of the 

 Trimatcs. Carnivora and Marsupialia, he finds that in many 

 oxuiuples of various genera and species " reduplication of teeth may 

 ocrur in such a way that a tooth which is usually single may be 

 n'pn-.>*ent«'d by two teeth, and that the two teeth thus formed may 

 either (I) both take place in the ordinary .scries, or (2) may stand 

 iilly and intenially re.sj»ectively." 



ine prevailing hypothesis, as Mr. Bateson points out, necessarily 

 " involves a definite conception of the mode in which variation 

 works." and, further, that " in variation the individuality of each 

 inomlHT <»r the series is resjKJcted." 



r.ut. a.H in the case also with other multiple parts, such as digits 

 and phalanges, the .litlieulty in applying this principle and" in 



