482 



SCANDINAVIAN FISHES. 



to establish the genus Eleglnus, while Bean proposed 

 for the same reason" to restore Swainson's genus Tilesia. 

 15ut the singulai'ity in point is only a more advanced 

 development of the connexion between the air-bladder 

 and these transverse processes that occurs in other spe- 

 cies of the genus Gadiis. The two vermifoi-m processes, 

 from the front of the air-bladder, which we have just 

 descril)ed in the Common Cod, occur in exactly the same 

 form here; and a remnant of the pneumatic duct may 

 occur in some specimens, as in the case of tlie Common 

 Cod, though 1 have sometimes failed to find a trace of it. 

 These two forms are so closely connected that the 

 relationship between them must be regarded as even 

 more intimate than that between Gadus callarias and 

 G. (>(jav: and the only constant distinction between 

 tliem that has hitherto been suggested, is derived from 

 the coloration — in naraqa the dorsal side, above the 

 transverse processes, wliicli are externally perceptible to 

 the touch, is grayish green ^\■ith darker, brown spots, 

 reminding us of the common Baltic Cod, Mhile in 

 irachuja the dorsal side is plain yellow or reddish gray*; 

 l)Ut the ventral side in both forms is silvery white, 

 finely punctated with lilackish brown. Their habitats, 

 however, are distinct, as far as we can decide from our 



present acquaintance with them; and even if the above 

 character is the oidy valid one, they must, therefore, 

 be regarded at least as distinct local vai'ieties. G. naraga 

 lives in the immediate neighbourhood of the Scandina- 

 vian fauna, from the Murman coast nnd the White Sea 

 at least as tar east as the River Obi. G. gracilis was 

 first discovered by Stellek and Tilesius in Kamchatka 

 Sea, and, according to Pallas, is known by the Japa- 

 nese as fara. Nordenskiold bought it in quantities 

 of the Ciiukchees at PitlekaJ ; and we have a description 

 b^' Tuhneh, from Alaska-, as well as a drawing from 

 Noudenskiolu's winter-quarters, of the manner in which 

 the Chukchees and Esquimaux catch this species during 

 their winter-fishing in holes made in the ice^ How far 

 west irachnja goes, or how far east navaga, is a point 

 on which we have no further information. If the forms 

 are really distinct in their geographical range, they 

 give us an excellent example of systematic separation 

 in spite of the closest kinship, an example \vhich is given 

 as briefly as possible in the ajjpended tables. First we 

 give a table of averages of the most important ex- 

 ternal relations in four specimens of navaga, varying 

 between 171 and 2-lU mm. in length, and three of 

 irachnja, between 345 and 382 mm. long. 



See Joiin., Gilb., 1. c. 



According to TuuNKl: tlie fisli is griiyisli l)rci\vii iiliove, when frcsli; and in some speciuieiis siiinll, dark .spots appear on llie bides. 



In siininipr, according to Tukn'KK, it is taken in Alaska liy tisliing with long rods from jotting rocks. 



Mininuini in these specimens :^ 9'7. 



Maxiiiuini 

 Miiiiniiiin 

 MaNiniuui 

 Miniininii 

 Jlaxiiniini 



Mininuini 

 JIa\iiiiniii 

 Mininitini 

 Ma.\inuiin 



Mininiiiiii 



= 9-5. 

 = 8-7. 

 = 8-4. 

 = 15.2. 

 = 15-0. 

 = 113-2. 

 = 16-7. 

 = 11-1. 

 = 113. 

 = 3-9. 

 = 4-1. 



