Mamiualia. 15 



strong teeth of this species is due to the I'act that it is not confined 

 to the pack-ice, and that in other regions its food-supply is derived 

 from creatures which need some hokling, but yet are not so strong 

 or active as those whicli serve to nourish Ogmorhinus. The external 

 shape of Lcptonycliotcs is that of an animal fitted for rapid motion in 

 the water, so that it is natural to suppose that fish may form no 

 small portion of its diet, 



Lastly we have OmmatopJioca, with its very feeble and variable 

 dentition, obsolete sagittal crest, yet comparatively strong lambdoid 

 crest. It seems probable that the exact number of its teeth is not of 

 importance to this animal. Even their size is sufficiently variable 

 to admit of the thought that the whole dentition is little used, and 

 it seems to me highly probable that Oinmatoplioca is in the process 

 of reducing its teeth. Its food is soft, consisting, according to Dr. 

 Racovitza, of large cephalopods. Such crushing of these as might 

 be necessary would be performed as well by the flat jaws as by 

 cusped teeth like those of the Fhoddae. Consequently the teeth 

 are in a state of disintegration. This supposition agrees very well 

 with what I have already suggested in the case of Lohodon. In the 

 last-named genus the teeth are highly developed, not for grinding 

 purposes, but for use as a sieve. In Ommatophoca, not being 

 available as a sieve, they are useless. They thus fall outside the 

 influence of natural selection, except in so far as their reduction 

 may be of use to the species. Variation and enfeeblement result, 

 processes which, if for the advantage of the animal, will no doubt be 

 carried to their fullest extent. My supposition is supported by the 

 external appearance of the animal as graphically described, I think for 

 the first time, by Dr. liacovitza. It is, he writes : " le plus phoque 

 des phoques, car chez lui toute forme de quadrupede a disparu. Son 

 corps n'est plus qu'un sac fusiforme pourvu de membrcf^ tres 

 reduits," from which I gather that Ross's Seal does not possess the 

 natural appearance which belongs to an agile carnivore such as 

 Oijiiio7^hinus. But the most remarkable confirmation of my sup- 

 position is supplied by one of the skulls obtained by Mr. Nicolai 

 Hanson. In this skull (No. 1), which, although adult, readies the 

 shortest total length of any skull of the genus known to me, there is 

 no trace of any cheek-teeth either in the upper or lower jaws. Not 

 only have the teeth disappeared, but with them have vanished, to a 

 great extent, the alveoli. There seems, then, to be some solid ground 

 for my hypothesis both in regard to Lohodon and to Ommato'phoca. 

 If I am right, there is here a striking instance of the acquirement 

 of perfectly distinct habits, and consequently of form and appearance, 



