34 * SALMON AND TEOUT IN ALASKA. 



but the matter needs further investigation. Even these slow-moving 

 forms may be able to avoid the action of the current at the foot of 

 lakes and remain within the lake boundaries. 



FOOD SUPPLY IN RELATION TO THE HATCHING SEASON. 



The relation of food supply to the season of hatching has not yet 

 been worked out. In the latitudes in which the sockeye is found 

 it is improbable that any fry reach the feeding stage in advance of the 

 opening of tlie streams from ice, unless when the spawn is deposited 

 on the lake shores. In this latter event they would seek subsistence 

 in the lake waters. It is shown above that these fingerlings feed to 

 some extent upon surface forms when in streams or lakes at a season 

 when such food is present. Their more natural food appears to be 

 the subsurface plankton. The abundance of this material may be 

 an important factor in determining the time of departure from fTesh 

 water. 



This question has an important bearing in fish cultural work. The 

 use of spring or heated water ma}^ shorten the incubation period to 

 such an extent that fry reach the feeding stage in advance of the 

 natural production of their food. To liberate them in that event 

 must be disastrous; to retain and feed them artificially nullifies the 

 economic advantage obtained in hastening the incubation. 



The temperature at which the most thrifty fry may be produced 

 is another cpiestion that should be taken up. There are some data 

 to' show that low temperatures increase' the number of temperature 

 units required to effect the hatching. In long periods of depressed 

 temperatures errors in the thermometer or its reading will be a 

 greater factor than in short periods of observation. But if very cold 

 water does retard the hatching, then such temperatures may or may 

 not be advantageous. The success of certain methods in handling 

 an insect-feeding species in waters that never freeze can be no crite- 

 rion as to the best methods for a species of different habits living in 

 different waters. 



GROWTH IN FRESH WATER. 



Fifty-nine young sockeyes taken in Yes Lake August 24, 1905, 

 varied from about 32 to 66 mm. total length, with an average of 

 about 46 mm.; 66 seined on September 10 varied from 35 to 75, 

 with an average of 50 mm.; 59 taken September 26 and 27 varied 

 between 34 and 82 mm., with an average of 45 mm. Assuming that 

 these by the folloA\dng spring would reach the same size as the Naha 

 fish taken in 1903 when under the most nearly natural conditions, an 

 average length of 65 mm., they would have to increase from 15 to 20 

 mm. in length from September to May. 



