ART. 15 ANATOMY OF THE EAEED AND EARLESS SEALS HOWELL 17 



ever, for in a large fetus the mastoid is almost as much inflated as 

 the audital bulla, and the proportions are found to be much the same 

 as in the adult. This character of inflated mastoid is common, in 

 varying degree, to all the Phocidae. As the functions, if any, of 

 mastoids of different degrees of inflation have never been discovered, 

 it is useless to speculate on the differences encountered in the pin- 

 nipeds. 



In the fetal Phoca skull there is a pair of symmetrical bones, one 

 on either side, bounded by the mastoid, parietal, supraoccipital and 

 exoccipital, and measuring 21 by 10 mm. These are found in those 

 few very young Phoca vitulina skulls that are available, but their out- 

 lines become obliterated in older animals — even in inimatures of 

 medium size. In an adult skull of Phoca groenlandica, however, and 

 a subadult of Cystophora^ these accessory bones can be perfectly 

 traced. They can not be considered as Wormian bones, for they are 

 too symmetrical and too regularly situated. It seems justifiable to 

 consider them as a phylogenetic remnant, comparable to the " rep- 

 tilian " supernumerary bones of some insectivores. (See Wortman, 

 1921.) I can not, however, find that their undoubted homologue 

 exists in the skull of any reptile which I have encountered in the 

 literature of the subject, unless they are comparable to supratem- 

 porals of such a genus as Procolophon; and I am far from con- 

 vinced that this is likely. 



In all adult otariids the temporal muscles reach the sagittal crest 

 and in old males they attain a phenomenal size, as indicated by the 

 development of the crests — in fact far larger than one imagines 

 would be of practical use to an animal with such dentition and with 

 such a diet. In a sagittal direction these fossae extend from the 

 lambdoidal crest to the supraorbital processes. In many phocids 

 (as Mirounga and some Phoca) these muscles also reach the sagittal 

 crest, but as a rule they are much weaker in this family and do not 

 encroach so far onto the frontals». It may be that no old male of 

 Phoca hisfida is available, but I have seen none in which these 

 muscles reach the sagittal line. In the skull upon which this osteo- 

 logical study is chiefly based they are far apart and weak. 



In fetal skulls the supraoccipital of Phoca has a more definite 

 rostral inclination than in Zalophus. In adults of the latter genus 

 as well as in some phocids (as Cystophora) the supraoccipital plane 

 slopes gradually, but in all phocids the occipital crest exhibits quite 

 a sharp angle in the middle portion of either half. This assumes 

 almost a right angle in the Phoca hispida^ in contrast to the more 

 even curve of this crest in Zalophus. The reason for this is not hard 

 to find, for in the otariid the different muscles of paroccipital-supra- 

 occipital insertion are more or less evenly distributed for the entire 

 86377—28 2 



