70 
BULLETIN OF THE BUREAU OF FISHERIES 
recapture at other places than Bristol Bay and a distinctly higher percentage of 
returns from Bristol Bay even when, as shown in the last column of Table 41, the 
percentage is figured on the number remaining after the fish taken in other places 
than Bristol Bay are taken out. This was a necessary procedure since the returns 
from Bristol Bay are seriously affected by the capture of fish in other localities, 
especially close to the point where the tagging was done. Both of these facts 
point to the same conclusion, namely, that not all of the fish bound for Bristol Bay 
and found in the region of the Shumagin Islands during the height of the fishing 
season pass close to the shores in Ikatan and Morzhovoi Bays on their migration. 
There is plenty of reason to believe that a considerable proportion of the fish found 
south of the Alaska Peninsula do traverse these waters, but it seems probable, at 
least, that not all of them do so. It is possible that the fish enter Bering Sea through 
some one of the passages farther to the westward, or they may pass through the 
center of Ikatan Bay on their way to False Pass and thus avoid the traps. If some 
of the fish tagged in the Shumagin Islands went directly through Ikatan Bay and 
False Pass, or through some one of the western channels, it would account for the 
differences noted in the returns from these experiments and those begun in Ikatan 
and Morzhovoi Bays. No such migration through the western channels has ever 
been reported, however. 
While it is felt that the figures given in Table 41 give a fairly acceptable mini- 
mum measure of the total intensity with which the schools of the Ikatan-Shumagin 
Island district are fished, and are, furthermore, sufficiently accurate for comparative 
purposes, it is not thought that the percentages taken in Bristol Bay give a reliable 
measure of the intensity of fishing at that point. All of the factors that tend to 
reduce the percentage of returns will continue to operate throughout the season, 
and will therefore be more effective in reducing the returns from Bristol Bay than 
from points nearer the place of tagging. The tagged fish actually reported from 
Bristol Bay form 7.25 per cent of the total number tagged. However, this figure is 
obviously affected by the fact that a large number of fish, undoubtedly bound for 
Bristol Bay, were taken in the Ikatan-Shumagin Island district, where they were 
tagged. If we take from the total number tagged those which were recaptured 
in other localities than Bristol Bay and calculate the percentage of the remainder 
which were taken in Bristol Bay, the result is very close to 10 per cent. It seems 
practically certain, however, that more than 10 per cent of the total run to Bristol 
Bay are taken by the fishery. The results are, of course, affected materially by the 
closing of the season in Bristol Bay on July 25, but it is not likely that this accounts 
in any great measure for the small number of returns. It is possible that this low 
percentage taken in Bristol Bay indicates that races bound for small local streams 
form a more important item in the Ikatan-Shumagin Island schools than is 
supposed. 
The tagging experiments in 1922 and 1923 have agreed in demonstrating that 
the red-salmon colonies of the Chignik and Karluk Rivers, of Olga Bay, and Cook 
Inlet do not extend their range in any numbers as far to the westward as the Shu- 
magin Islands and are therefore not at all involved in the fisheries of Morzhovoi and 
Ikatan Bays. No salmon tagged in either of these bays in 1922 or 1923 were 
captured even as far eastward as the Shumagin Islands. Only among those tagged 
