66 



BULLETIN OP THE BUREAU OF FISHERIES 



chosen as the most promising fields for predictive research. Not until they have 

 been subjected individually to the crucial test of extended experimentation shall we 

 be in a position to indicate the degree of dependence that can safely be placed in 

 them. At the present stage of progress they form the only basis we have for proph- 

 ecy and can be used tentatively and with caution for such value as they will be found 

 to possess. 



The 5-year fish, members of the 5 3 group, that can confidently be expected to 

 constitute the great majority of the 1927 run developed from eggs laid down during 

 the summer and fall of 1922. They hatched during the season of 1923 and remained 

 in residence in Karluk Lake until the spring of 1925, when, in company with other 

 groups of fingerlings, both younger and older, they descended the river to the sea 

 in a pronounced wave of migration. Nothing further was seen of these 3-year 

 fingerlings until the appearance in the run of 1926 of a certain number of individuals 

 that had separated themselves from their companions and had matured at the early 

 age of 4 years (the 4 3 group), while the great majority of their fellows remained 

 behind in the sea for further growth and development. 



We have available, then, as indicators for the run of 1927, (1) the number of 

 spawning fish counted through the weir in 1922; (2) the size of the downstream 

 migration, as observed (but in no way quantitatively determined) in 1925; and (3) 

 the abundance of 4 3 grilse in the run of 1926. We shall consider these in their order. 



1. The red-salmon run of 1922 to the Karluk River was one of the poorest of 

 which we have any record. For only one other year since the industry was firmly 

 established (that of 1914) has there been an equally poor commercial yield. More 

 important than the commercial take is the spawning escapement, which in 1922 

 amounted to 383,684 fish. This count, however, did not include the escapement 

 during a period from August 20 to September 4, during which the counting weir 

 was unable to operate because it was blocked by dead humpback salmon that had 

 spawned above the weir and drifted down upon it. Judging from the condition of 

 the run prior to August 21 and subsequent to September 4, a liberal estimate of the 

 fish that ascended the river during the 15 days in which the weir did not function 

 would give a sufficient number to raise the total spawning escapement for the season 

 to 400,000 fish. This is to be compared with the escapement in 1921, which totaled 

 approximately 1,500,000 fish. The returns of the 1921 spawning thus far obtained 

 indicate a probable rate of increase of three to one. If this ratio should hold with 

 some degree of approximation for the spawning of 1922, the total yield of that 

 brood would be 1,200,000 fish. Should 80 per cent of these run as 5-year fish in 

 1927, according to expectation, they would total 960,000, and the 4 and 6 year fish 

 that would accompany them from other spawnings might be expected to raise the 

 total run to a figure not exceeding 1,500,000. It will be noted that this estimate is 

 based on the assumption — which our present experience is too limited to justify — 

 that the ratio of increase in the Karluk, from spawning colony to mature progeny, 

 will, with each year, approximate three to one. The experience of 1927 will be most 

 valuable as throwing light on the extent to which this ratio fluctuates in different 

 years. 



2. No attempt was made in the spring of 1925 to estimate the number of down- 

 stream migrants, which were largely derived from the 1922 spawning. They could 



