40 U. S. BUREAU OF FISHERIES 
Marine studies Investigation of the marine de of the life 
history of the Bristol Bay populations i is not only important in order 
to determine the factors influencing the salmon growth and survival, 
but is also of imporance in ascertaining the magnitude of the com- 
mercial catch, since part of the run of fish is captured 400 to 500 
miles from the home streams of the salmon. 
Tagging experiments were carried on in the Shumagin Island area 
and in the Ikatan-Morzhovoi Bay area to determine the migration 
routes and destinations of the salmon fr equenting those areas. In 
these experiments, mature salmon on their spawning migration were 
tagged by means of an aluminum strap-tag which was attached to 
the dorsal part of the caudal fin. The experiments were similar in 
nature to those carried on in 1923 by Charles H. Gilbert and Willis 
EEeRich:* 
A greater return was obtained from the 1939 Shumagin Island 
tagging than was obtained in 1923, the recovery being 34.5 percent 
as against 28.7 percent in 1923. A much hee percentage of the 
fish was caught locally, 22.6 percent in 1939, compared with 2.6 per- 
cent in 1923. A greater percentage of returns was taken to the east- 
ward, 4.3 percent in 1939, and 0.5 percent in 1923. Most of the 
returns came from Chignik, which experienced a good run during 
the past season. Lower percentages were recovered from [katan and 
Bristol Bays in 1939, 3.0, and 1.2 percent, compared to 11.1, and 9.2 
percent, respectively, i 1923. 
The 1939 study of the migration of fish around the Shumagin 
Islands supports the conclusion that as many fish traveled up Popoff 
Straits as migrated down the Straits. The results, therefore, do not 
confirm the conclusion made by Gilbert and Rich on the basis of the 
1923 study that “the majority of the salmon that escape the traps 
in the region of Kelly Rock skirt the south shore of Unga 
Island 4 Part yur 
The return from the Tkatan tagging experiments was 37.0 percent 
in 1939 as against 44.6 percent in 1923. Fewer fish were recaptured 
locally in 1939; 19.7 instead of 34.0 percent. A slightly higher per- 
centage migrated eastward; 1.0 compared to 0.3 percent in 1928. 
A smaller number was taken in Bristol Bay ; 4.0 percent in 1939 
and 6.2 percent in 1923. The lower percentage of recovery in the 
Bristol Bay area was due in part to the fishery on the south side of 
Unimak Island, which recaptured 10.8 percent of the fish tagged, 
whereas in 1923, this fishery was not in existence and consequently 
once the fish left Ikatan Bay they were not intercepted until they 
reached Bristol Bay. Of special interest is the evidence indicating 
that at least a fair proportion, if not a majority, of the Bristol Bay 
fish which pass the Shumagin Islands and Ikatan Bay proceed to Bris- 
tol Bay along the south shore of Unimak Island, and thence probably 
through Unimak Pass. Gilbert and Rich (loc. cit.) concluded that 
most, if not all, of the Bristol Bay fish found in the Shumagin Island 
and Ikatan Bay area proceeded through False Pass on the way to 
their home streams. Although those authors pointed out that fish 
may enter Bering Sea through other passes, they also stated that 
“no such migr ation thr ough western channels has ever been reported.” 
2 Second experiment in tagging salmon in the Alaskan Peninsula Fisheries Reservation, 
summer of 1923, by Charles H. Gilbert and Willis H. Rich, Bulletin, U. 8. Bureau of 
Fisheries, vol. 42, 1926, pp. 27-75. 
