riALMON AND TROUT IN ALASKA. 35 



Since the deeper water of the lakes will not fall much below 39° F., 

 the point of maximuni density, and since with the approach of cold 

 weather the diffusion of oxygen will become more general, there seems 

 to be no reason, unless there is a cessation of activity due to the cold, 

 why fingerlings should not feed more or less during the winter. 

 Some species of crustaceans are known to reach their maximum 

 numbers during the winter. This question has not been investigated ■ 

 for Alaskan lakes, nor are any data yet available uponwliich to base 

 conclusions. Frj^ from the previous season's spawn in Yes Lake are 

 ordinarily able to begin to feed about the middle of April, at which 

 time they are somewhat over 30 mm. average length. By the latter 

 part of August they have reached an average length of about 45 mm., 

 a 50 per cent increase in a period of about four months. That this 

 is crudel}^ approximate is shown by averages in the three catches 

 cited above. 



Should the figures for the Naha yearlings show the approximate 

 size of the Yes Lake migrants, these have little greater growth to 

 make between August and May than they make between April and 

 September. This would require little increase in size during the 

 winter, since the lakes seldom freeze over before January, and up to 

 that time feeding on the lake plankton can be continuous. Dr. 

 Evermann's data from Alturas Lake, as showing rapidity of growth, 

 are rendered inaccurate by the necessity of comparing small finger- 

 lings from shallow water with the larger fish from deeper water, but 

 show clearly that the ^^earlings there attain a larger size than in the 

 Alaskan lakes studied. His specimens of July averaged 48 mm.; 

 those of September 9, 83 mm. 



It is interesting to note that while the young taken in Yes Lake 

 in September did not fall below 45 mm., 46 specimens obtained in 

 Heckman Lake October 2 averaged only 36 mm. and 28 taken on 

 October 7 averaged but 40 mm. With tliis diminished size was noted 

 a corresponding slimness and lean condition (fig., p. 18). It is also to 

 be stated that the surface tows for food material in Heckman Lake, 

 as compared with the material from Patching and Yes lakes, showed a 

 diminution in number and general size of the crustacean life. These 

 facts raise the question of the possibility of overstocking waters with 

 fry. Of course, in case the surplus of young fish leave the parent 

 waters as fry, as would seem to be indicated by the observations of 

 the British Columbia parties, there could scarcely be any danger 

 from tliis source. But the ^vriter's observations in the Naha failed 

 to reveal any such movement there and it is not unreasonable to 

 suppose that some condition exists there which alters the habits of 

 the fish in that respect. If that is true, and all the sockeye young 

 spend one winter in the lake, then, with the increased millions which 

 artificial hatching is able to turn out, a serious menace is offered to 



