FISHERY BULLETIN: VOL. 69. NO. 4 



usually set production goals in this period and 

 extended fishing until these goals were reached. 



As a result, one may conclude that 70 years of 

 intensive harvesting have not drastically affected 

 the timing of the Nushagak runs. All tagging 

 experiments conducted in Bristol Bay point to 

 a complete mixing of all races in the fishery and 

 exposure to the same fishing pressure in a spe- 

 cific river system (Smith, 1964; Mathisen, 1969; 

 Straty, 1969) . Thus there is very little evidence 

 of a differential rate of removal in time among 

 all the races that constitute the Nushagak sock- 

 eye salmon run. The only exception seems to 

 be the races bound for the Igushik system. Their 

 migration path follows the west side of Nushagak 

 Bay past Nichol's Spit. In earlier years when 

 the main fishing activities were concentrated 

 closer to the confluence of the Nushagak and 

 Wood Rivers than they are today, the fishing 

 pressure on the Igushik races during those years 

 was lower. 



A differential fishing pressure could arise from 

 the selectivity of the gill nets if some Nushagak 

 races consisted primarily of 3-ocean fish while 

 in others 2-ocean fish predominated. Burgner 

 (1964) has pointed out the ])reponderance of 3- 

 ocean fish to the Tikchik as one example. How- 

 ever, if a diminution of such races were of any 

 real consequence, it must manifest itself in 

 changes of the age composition through the re- 

 corded history of the Nushagak fishery. The 

 figures in Table 9 are based on the age composi- 

 tion in the commercial catches. Because of the 

 larger net sizes used up to 1926, a bias is intro- 

 duced in favor of 3-ocean fish and only the last 

 two periods are directly comparable. 



Throughout all years the majority of the fish 

 migrated to sea as age 1 smolts and returned in 

 somewhat the same proportion of 2- and 3-ocean 



fish. Over the years one can notice a shift, 

 with less 3-ocean fish in the catches of males in 

 recent years. If similar data were available 

 for the escapement, and thereby of the total runs, 

 one would in all probability see more of a con- 

 trast in the shift from 3- to 2-ocean female fish, 

 especially in the years when mesh sizes were 

 larger than 5\U inches. A mesh size experiment 

 conducted in 1928 by the Bureau of Fisheries 

 illustrates this point. The log ratio of catches 

 made with nets of .51 o-inch and 6-inch mesh sizes 

 are plotted by centimeter groups in Figure 6 

 and form an expected straight line. The aber- 

 rant points toward the upper size range are due 

 to a much larger sampling error because of the 

 very few fish present at these sizes. 



The essential element of an escapement is not 

 the total number of fish present but the poten- 

 tial egg deposition they represent. During the 

 first period of the Nushagak fishery, when net 

 sizes ranged from 6Vi to 5% inches, escapements 

 of the same numerical magnitude as in later 

 years must have represented a substantially 

 higher potential egg deposition since a much 

 higher proportion of 3-ocean females was in- 

 cluded in those escapements than in years with 

 .51/2-inch mesh size. On the average, 3-ocean 

 females produce 650 eggs more per female than 

 2-ocean fish. The mean fecundity of these two 

 groups are 3,639 and 4,290 eggs, respectively 

 (Mathisen. 1962). 



This net selection has another, more intangible 

 aspect. Not only is fecundity greater in the 

 larger 3-ocean females, but egg size is also a 

 function of the size of the females (Mathisen, 

 1962). Thus there may be a higher survival of 

 the progeny in this case than from eggs pro- 

 duced by 2-ocean females in the same environ- 

 ment. This concern was expressed in 1927 by 



Table 9. — Summary of age in the commercial catches of sockeye salmon in Nushagak. 



760 



