A STUDY ON TROUT AND YOUNG SALMON. 259 



2) That the anal fin of the salmon is smaller than that of 

 the trout compared with the tail. 



3) That the tail of the salmon is more slender than that of 

 the trout. 



However, it is also obvious that these relations vary, and to 

 such an extent, that the amplitudes of variation in the two species 

 vaguely meet or even overlap each other. However on this 

 latter point it ought to be noted, that some of these overlappings 

 are caused by the introduction of specimens, which, in size, 

 differ only inconsiderably from the previous class or group; 

 the character in question alters in fact, during development. 



If we scan the table giving the scale-number, we find that 

 this character remains practically constant through all groups. 



Upon examining table II (on Tail-coefficient I) however, we 

 remark that this coefficient in the salmon slowly rises during 

 early growth; in other words that the anal fin of the young 

 salmon gets smaller and smaller as the fish grows, while that 

 of the trout remains nearly unaltered. 



Excepting the youngest group the curves representing this 

 character in the two species run nearly separated, with one 

 single exception, a salmon of the group 16—50 cm. This fish 

 measuring only 17,5 cm. was captured in the ocean outside 

 Ghristianssand, as will be mentioned later, and only the size 

 destinguished it from ordinary emigrating smolts. The coefficient 

 of the youngest group beeing low, the introduction of this 

 specimen will in some degree alter the real course of the curve 

 of the second group. 



The same phenomenon also occurs in the table on Tail-coeffi- 

 cient II (the slenderness of the tail), which in both species rises 

 a httle with the age, especially between 16 and 50 cm. (cfr. 

 Tab. Ill a and b). 



A classification in sizes separated by sufficiently large inter- 

 vals would possibly have yielded a more strikingly typical result. 



