MIGRATIONS OF PINK SALMON 
647 
runs in the region of Cape Fox, large numbers of the pink salmon migrating through 
Dixon Entrance began to turn northward into Clarence Strait along the east shore of 
Prince of Wales Island in the vicinity of Cape Chacon. Here again, the salmon fol- 
lowing this course of migration increased in numbers as the season progressed and 
built up a separate and definite run into Clarence Strait. From a week to 10 days 
later a third run of pink salmon began to leave Dixon Entrance and migrate north- 
ward; this time into Cordova Bay. This run, like the others, increased in abundance 
as the season progressed, thus forming three definite runs of pink salmon originating 
in Dixon Entrance. This information was secured from a study of the daily catches 
of the fishing gear operating in these waters. Studies of similar recent records by the 
Bureau indicate that three definite runs of pink salmon continue to appear in these 
waters. 
In order to provide a uniform method for classifying and reporting the recoveries 
of tagged salmon, the spawning localities of Clarence and Sumner Straits were grouped 
into definite areas. Although these areas include waters which are similar to those 
of the fishing districts, they cannot be construed as being identical. They were not 
formed with any idea of using them to replace the present districts as factors other 
than the locations of the migratory channels of the salmon enter into their forma- 
tion. They will be referred to hereafter as geograplnc areas of recovery which are 
shown in figure 1 and may be described as follows: Area 1 includes all the waters of 
Revillagigedo Channel, Portland and Behm Canals, and the waters surrounding Duke, 
Annette, and Gravina Islands. Area 2 includes all the waters of Clarence Strait 
along the east shore of Prince of Wales Island from Cape Chacon to Approach Point. 
Area 3 includes all the waters of Clarence Strait above a line from Approach Point 
to Caamano Point and below a line from Narrow Point to Ernest Point. This area 
also includes the waters of Ernest Sound, Zimovia Strait, Bradfield Canal, and Blake 
Channel. Area 4 includes the remainder of the waters of Clarence Strait above a 
line from Narrow Point to Ernest Point, and the waters of Snow Passage and Stikine 
Strait. Area 5 includes the waters of Sumner Strait, Keku Strait, Wrangell Narrows, 
and the lower extremity of Frederick Sound below latitude 56°30' N. Area 6 includes 
the waters of Cordova Bay and contiguous channels. 
Since no information was available concerning the localities in which the salmon 
comprising these migration waves spawned, the Bureau, in 1924, began three series 
of tagging experiments in the region. The first was in the vicinity of Cape Fox, the 
entrance to Revillagigedo Channel, and Portland Canal; the second in the vicinity 
of Cape Chacon, the entrance to Clarence Strait; and the third in the vicinity of 
Cape Muzon, the entrance to Cordova Bay. The locations at which these tagging 
experiments were carried on are shown by the black dots in figure 1. 
In classifying the results from the tagging experiments summarized in tables 1 
to 8, all tagged salmon recovered at the location of tagging were considered as not 
having migrated from that location and were not included either in the number of 
individuals tagged, or in the number recovered. The recovery of a tagged salmon 
cannot be considered as indicating a route of migration unless capture has been made 
at some distance from the point of tagging. Accordingly, no record is given of the 
recoveries at the point of tagging nor those where the locality of recapture is doubt- 
ful. Hence, the total percentage of tagged fish recovered in the experiments does not 
represent the entire influence of the intensity of the fishery. 
