116 SALMONID^. 



on them can bo admitted into tlic sj-stom. Therefore, although for 

 completeness' sidce we have mcntiojied them in the following list, we 

 have added only such jiarts of the original descriptions as may assist 

 in recognizing the species. 



33. Salmo purpuratus. 



Salmo purpuratus, Pall. Zoogr. Ross.- As. iii. p. 374; Cuv. Sc Val. xxi. 



p. ?>^^S. 

 clarkii, Bichards. Faun. Bor.-Amer. iii. p. 225. 



Very closely allied to S. fario, but with more vertebrae. 



B. 12. D. 13-14. A. 12-13. L. lat. 130. L. transv. 20/30. 

 Caic. pyl. 40. Vert. 63. 



Form of the head and body as in S. fario. The posterior point of 

 junction of operculum and suboperculum is nearly midway between 

 the upper end of the gill-opening and the lower anterior angle of 

 the suboperculum. Lower limb of the prajoperculum rather distinct. 

 Snout obtuse ; maxillary of moderate width ; it extends scarcely to 

 the vertical from the hind margin of the orbit in a specimen 9 inches 

 long, and scarcely beyond it in one of 16 inches in length. Den- 

 tition rather feeble ; vomerine teeth in a double series. Fins mode- 

 rately developed ; caudal fin emarginate in young specimens, sub- 

 ti'uncate in specimens 9 inches long, in which the length of the 

 middle caudal rays is contained once and three-foiu'ths in that of the 

 outer ones. There are thirteen scales in a transverse series descend- 

 ing from behind the adipose fin obliquely forwards to the lateral line. 

 Head, body, dorsal and caudal fins covered with black specks. 



Pacific coasts of Korthern Asia and America. 



a-b. Seven and nine inches long. British Columbia. Presented by 

 J. K. Lord, Esq. 



Besides these specimens, we have examined two of Pallas's typical 

 specimens, flat skins, 14 and 16 inches long. Valenciennes regarded 

 them as belonging to two different sjiecies ; however, it is possible 

 that they are merely male and female of the same fish. Pallas says 

 of this species (which is evidently non-migratory) : — 



"This species, which migrates {catanadroma) from the North 

 Pacific, usually entei's the rivers Bolschaya and Bystraya and their 

 tributaries, from the beginning of June to the end of September, not 

 in troops but singly. Some also are said to remain all the year in 

 the i-ivers and to return to the sea in the middle of May. On account 

 of its voracity, it is always found very fat, so that, from the excellence 

 of its white and firm flesh, it is to be recommended before aU the 

 other species ; for it feeds incessantly during its ascent of the rivers, 

 not only on the eggs of fish, small fish, Phryganeae, Potamogetons, 

 but also on the ' migratoiy mice ' crossing the rivers. When from 

 beneath the water it perceives a dwarf Sorbus with berries hanging 

 over, it leaps out of the water with a great jerk and seizes the berries. 

 Hence, unlike the other trout, which, during the ascent of the rivers, 





