peak proportions between 1 and 10 July. By late July 

 most of the fish have entered their respective river 

 systems, and few are left on the fishing grounds. 



Commercial fishing in Bristol Bay is done almost 

 entirely with gill nets. The nets are set from one- or 

 two-man boats and drift with the tide (drift nets) or 

 are staked or anchored along the beaches (set nets). 

 The fishery is so intense and efficient that it takes 

 almost all of the fish in a district during an open 

 fishing period. Because of this, the fishery in each dis- 

 trict is closed periodically so that spawners from all 

 time segments of the run can escape to the stream. 



The day-to-day progress and size of the run through 

 each fishing district is monitored by the Alaska 

 Department of Fish and Game, which exercises con- 

 trol over Alaska fisheries. Test-fishing is done during 

 the closed periods and in closed fishing areas both up- 

 stream and seaward of the fishery to obtain im- 

 mediate indices of the numbers of fish entering and 

 escaping the fishery. Finally, sample counts are made 

 from observation towers on each bank of all the major 

 rivers entering the bay to obtain an accurate estimate 

 of spawning escapements. 



DETERMINING THE DISTRIBUTION 



AND MIGRATORY PATTERNS OF 



BRISTOL BAY SOCKEYE SALMON 



Three different types of investigations have 

 demonstrated that sockeye salmon in the Bering Sea 

 near long. 170°W and lat. 60°N are bound for Bristol 

 Bay. These investigations involved (1) studies of scale 

 characters, morphological characteristics, and 

 parasite infestations (Margolis et al. 1966); (2) studies 

 of the direction of movement of sockeye salmon 

 caught in gill nets (Barnaby 1952) and purse seines 

 (Hartt 1962); and (3) tagging experiments (Hartt 

 1962, 1966; Kondo et al. 1965). 



In this section I examine data from these published 

 sources and from unpublished tagging studies done in 

 Bristol Bay proper to describe the distribution and 

 migration pattern of stocks in offshore and inshore 

 areas. In addition, I use changes in the age group 

 structure of fish tagged at various locations to show 

 the distribution of several major sockeye stocks in the 

 inshore area. The offshore region includes that area of 

 the Bering Sea east of long. 170°W and south of lat. 

 60° N to a line drawn between Cape Constantine and 

 the Cinder River (Figs. 1 and 2). The inshore area in- 

 cludes the remaining area to the river mouths. 



Offshore Distribution 



The distribution and migration routes of sockeye 

 salmon in the offshore area have been determined 

 from the results of exploratory fishing along north- 

 south transects by research vessels of the United 

 States and from the locations of recapture of sockeye 

 salmon tagged in the offshore area by the United 

 States and Japan. The sources of data and the years 



involved in the exploratory fishing and the tagging I 

 consider here are given in Appendix Table 1. 



Exploratory Fishing. — The exploratory fishing 

 was done in offshore waters of the Bering Sea inter- 

 mittently over a period of 28 yr. In most cases gill nets 

 of various lengths and mesh sizes were the principal 

 gear, although purse seines and longlines were also 

 used. 



Results from 8 of the years of this exploratory 

 fishing (all with gill nets) are used to show distribu- 

 tion in the offshore area. These years were selected 

 because (1) fishing was carried out during the period 

 of the spawning migration when sockeye salmon are 

 most abundant (June and July); (2) fishing was fairly 

 systematic, usually on consecutive days at stations 

 located along given transects (Fig. 2); and (3) the 

 transects fished provided the most extensive coverage 

 available. 



In some years fishing was carried out along a given 

 transect two or three times. In these instances, catch 

 data are used to examine the consistency of the 

 sockeye salmon distribution and therefore the migra- 

 tion route at various times during the spawning 

 migration. Fishing was also carried out along long. 

 170°W in 3 different years, making possible a com- 

 parison of sockeye salmon distribution between years 

 with runs of different magnitude. 



Results of Exploratory Fishing.— The gill net 

 catches of sockeye salmon along the six transects es- 

 tablished across the approaches to Bristol Bay in the 

 offshore area show that the abundance of sockeye 

 salmon increases and then decreases with increasing 

 distance in a northerly direction offshore from the 

 northwestern side of the Alaska Peninsula and the 

 northeastern Aleutian Islands (Figs. 3-10). With one 

 exception (28 June-3 July 1940— Fig. 6) this pattern 

 was consistent for all years and along all of the 

 transects. 



The area of greatest abundance of sockeye salmon 

 along the long. 170°W transect was between 161 and 

 322 km offshore in 1958, 1959, and 1961 (Figs. 3-5). 

 Catches decreased nearer the south side of the Pribilof 

 Islands. In 1959 and 1961 large catches were made 

 north of the Pribilof Islands (Figs. 4 and 5), but they 

 decreased with increasing distance to the north of 

 these islands. A comparison of the abundance of fish 

 along the long. 170°W transect at approximately the 

 same locations, but a week or two apart, shows that 

 the pattern of distribution remained essentially the 

 same (Fig. 5). This 2-wk period is sufficient for the 

 major portion of the total Bristol Bay sockeye salmon 

 run to pass through this area. These results, plus the 

 fact that the distributions of sockeye salmon in the 3 

 yr were similar, suggest that this pattern may not vary 

 greatly during the time the entire run passes through 

 this area or from year to year. 



Along the transect between Cape Mordvinof and 

 the Pribilof Islands the size of the catches increased, 



