46 
BULLETIN OF THE BUREAU OF FISHERIES. 
In several instances, however, the capture was known to be effected off the mouth, 
of Bear River, and as this stream is far more important than the Sandy River it 
seems probable that the majority of the salmon that were schooling off the Bear 
River-Sandy River beaches in 1922 were bound for Bear River. It is strikingly 
corroborative of this view that not a single individual out of the 639 tagged in the 
Port Moller district was recaptured in Bristol Bay, whereas out of the nine tagging 
experiments conducted south of the Alaska Peninsula at Unga Island and in Ikatan 
Bay all but the first two experiments (June 13, 200 specimens; June 14, 100 speci- 
mens) contained salmon afterwards taken in Bristol Bay. The inference seems 
plain and unquestionable that in 1922 a stream of migrants was traversing Isanotski 
Strait (False Pass) from the Pacific into Bering Sea, from early June to the middle 
of July at least, and that these distributed themselves to the red-salmon rivers 
along the entire northern shore of the Alaska Peninsula and throughout Bering 
Sea, from Nelson Lagoon to the Nushagak, and even to the Kuskoquim. The red 
s alm on bound in 1922 for Bristol Bay assuredly did not school close inshore until 
after they had passed the Sandy River and were perhaps approaching the mouth 
of the Ugashik. 
Table 1 . — Ikatan experiment. Tags 1 to 200, attached June 13, 1922, at P. E. Harris trap No. 7, Ikatan 
Bay. 
[Total recaptures 27=14 per cent.] 
Table 2. — Ikatan experiment. Tags 201 to 300, attached June 14, 1922, at P. E. Harris trap No. 3, Ikatan, 
Bay. 
[Total recaptures 29= 29 per cent.] 
