440 THE FUR SEALS OF THE PRIBILOF ISLANDS. 



36. Osimrus thaleichthys Ayres. 



Several young specimens froiri Nusliagak River were referred to this species by 

 Dr. Gilbert. Tliere is no other record of it from Alaska, and these northern examples 

 may prove distinct from the Californian species. 



37. Mesopus olidus (I'ullas). 



Petropaulski Harbor, Shana Bay, I tump Island. Also recorded from Petropaulski 

 by Stejneger. 



The posterior insertion of tlie ventral fins has been exaggerated in this genus. 

 Instead of being below or i)osterior to the middle of the dorsal, we find it approxi- 

 mately under front of dorsal, varying from slightly in advance of this point in the 

 young to slightly behind in the adults. 



Specimens from the two localities given above differ somewhat from each other in 

 length of dorsal and in scale formula. Those from Petropaulski have 05 to 07 scales, 

 and !) or 10 developed rays in the dorsal; from Iturup Island, 57 to GO scales, and 8 or 

 9 rays in the dorsal. We do not venture to separate the two lots on the basis of our 

 limited material, although the distinctive characters are constant in about 20 speci 

 mens of each. In M. oligndon Kner ( = M. olifJus) from Decastris Bay there are said 

 to be about 00 scales, but the figure of Kner shows ('>8. lloi)resentatives of this 

 species from St. Michael, Alaska, are reported as having 50 to GO scales, thus agree- 

 ing with those from Iturup. In the former, however, the paired fins appear to be 

 longer. j\I. pn-tiosiis from soutlieastern Alaska and Puget Sound differs from 71/. 

 olidus from Petropaulski no more than the latter do from iturup or St. Michael 

 specimens here called M. oluhis. It seems probable that we are dealing either with 

 one species or with three or four. 



The generic name Mesopus was regularly proi)Osed, the genus characterized, and 

 type specified on page 14, I'roceedings Academy Natural Science, Pliiladelphia, 

 1862. On the following page, in a key to the genera, there apj)e}irs in its place the 

 name Hypomrsns. In the index to the volume the name Mesopus alone appears. 

 There is nothing in the article to indicate which of the names was the final choice of 

 the author. Even were that evident, we c(»nsider it safer to conform strictly to the 

 law of priority without i)ermitting any exceptions. 



38. Leuroglossus stilbius (iilbert. 



Recorded from near Unalaska in .'?51 to 40G fathoms. 



39. Therobromus callorhini lyiicns, iipw species. Seal fisli. (Plate XLA'IL) 



Among the fishes obtained from the stomachs of fur seals by Messrs. Townsend 

 and Alexander were many examples of an undescribed isospondylous fish related to 

 the Argentinida^, although pos.sibly representing a new family. For this species the 

 name Therohrovivs eaUorhhii is i)roposed, fro.n the fact that it is so extensively eaten by 

 the fur seal. Owing to the tenderness and small size of this fish, it is so quickly acted on 

 by the gastric juice tliat nothing but bones remained of the many hundred specimens 

 that were seen and while evidently common, it can be described only from the 

 skeleton. No example of Arffentma being available it can only be said that Thero- 

 bromus differs from that genus in the shape and proportions of the component bones 

 of the jaw and gill covers, and that it finds its nearest relative in Mesopus, from 

 which it may be readily distinguished by its cranial characters as well as by the small 

 number of vertebrae, 26, 22 as against 32, 22. 



