CENTKAL AND WESTERN ALASKA INVESTIGATION. 157 



for an unexpected period, and the weather may turn exceptionally 

 warm. Then the stalest fish of the accumulated lot must be canned 

 each day, or one or more days' catch must be thrown away. The 

 regulation should be so enforced that not more than one day's surplus 

 shall be on hand at any time. This would remove all dangers from 

 the Bristol Bay pack and would at the same time be a powerful aid 

 to conservation of the fisheries. 



PORT MOLLER AND HERENDEEN BAY. 



These localities could not be visited during the past summer, but 

 one of the authors investigated the source of the runs of red salmon 

 to this region in 1918, and found that contrary to the opinions of 

 some packers the Bear River and the Sandy River produced, during 

 that year at least, all the red salmon there captured. The Bristol 

 Bay run in 1918 was very large, but no part of it skirted close in shore 

 as ifar to the southward as Port Moller. Whether it ever has done so 

 must now be judged by indirect evidence. For three years in succes- 

 sion, including 1919, yields of red salmon from this district have 

 been very poor. Still, the amount of the decline is not as great as 

 appears from the pack report of the only company which has oper- 

 ated for a term of years in the Bear River region. The Pacific Ameri- 

 can Fisheries captured about 1,125,000 red salmon off Port Moller 

 in each of the years 1915 and 1916. In 1917 their capture was 

 reduced to about a quarter of a million, and in 1918 it had recovered 

 to half a million. But 1917 is the year in which the three canneries 

 of Herendeen Bay began to compete for the Bear River and Sandy 

 River salmon, which prior to that date had been solely at the disposal 

 of the Pacific American Fisheries. If prior to 1917 the latter com- 

 pany were in fact, as seems probable, capturing a very large per- 

 centage of the available fish and if the total escape were very small 

 the advent of the new companies could do little more than subdivide 

 the year's product among the four participants. Making comparison 

 on this basis, it is seen that whereas the catch in 1915 and 1916, by 

 the Pacific American Fisheries was about 1,125,000 fish, in 1917 the 

 combined captures of the four canneries were nearly 800,000 red 

 salmon and in 1918 over 950,000 red salmon. 



While the reduction in these years is notable and, in connection 

 with the reduced pack which has followed in 1919, gives ground for 

 grave apprehensions concerning the future yields of Bear and Sandy 

 Rivers, it was not unexampled. Fully as great has been the falling 

 off in many other overfished rivers. From this point of view, there 

 is no necessity for assuming in explanation of the occurrences the 

 former participation of any portion of the Bristol Bay run. 



From another point of view, such participation becomes improbable. 

 It is alleged to have occurred in 1915 and 1916, two years in succes- 

 sion, and then in subsequent years to have failed to make itself felt at 

 the mouths of these rivers. According to this theory, the stream of 

 Bristol Bay fish moved farther from shore and became inaccessible to 

 this part of the coast on the very year in which new canneries began 

 operations there, and this diversion of the run has continued each 

 year since that time, irrespective of the size of the Bristol Bay run, 

 which was larger in 1917 and in 1918 than in the two preceding years. 



Taking all the known facts into consideration, it is believed that 

 the red-salmon captures in the Bear River region have been local 

 fish bound for Bear and for Sandy Rivers and that the reductions 



