fishing intensity and number of generations, 

 would be modified. The consequences of this 

 modification would be a reduction in CPUE on 

 the fishing grounds, but not a reduction in 

 abundance or population size. There are many 

 other conceivable attributes besides migration 

 which could be genetically controlled and se- 

 lected by fishing. 



The discussion of the effect of fishing on the 

 genetic structure of the population raises the 

 question of whether fishing selection might to 

 some extent replace natural selection. The 

 answer to this question would be affirmative in 

 the sense that, according to the theory of fish- 

 ing, fishing mortalities increase the mortality 

 rate and in some situations fishing deaths 

 might even exceed the natural deaths in the 

 unfished population. In fact, fishing might 

 even be more selective (in the sense that the 

 act of fishing Imposes a greater probability of 

 capture of individuals with certain genetically 

 controlled attributes) than natural deaths. We 

 can envision that all fishing deaths could result 

 from a monolithic type of selection. In an un- 

 fished population, however, the deaths would 

 tend to be from a variety of causes that might 

 even cancel one another. A monolithic type of 

 fishing selection then might to a large extent 

 replace or modify natural selection. The added 

 selection from fishing might have dire conse- 

 quences upon the population. These dire con- 

 sequences would result from an elevation in 

 selection such that its cost (see Haldane, 1957) 

 would be in excess of that which could be toler- 

 ated by the population in the face of other en- 

 vironmental demands for genetic plasticity. 

 Thus, such a phenomenon could account for 

 some of the catastrophic declines in fish popu- 

 lation size in which (even under very high fish- 

 ing pressure) the decrements in the population 

 owing to fishing deaths are considerably less 

 than the total deaths in the population. It must 

 be emphasized that the key to the whole selec- 

 tion argument involves the nonrandom removal 

 of certain genes from the population. If we ex- 

 amine the life history of any fish, then we see 

 that the greatest opportunity for the selective 

 removal of genes from the population is during 

 the first few months of life when upwards of 

 99.5 percent of any cohort usually dies. Thus 

 any fishing or natural genetic selection that 

 operates during the adult stages of life might 

 be relatively unimportant when compared with 

 the natural selection that might operate during 

 a fish's larval existence. Further, many fish- 



ing operations, in contrast to that for the alba- 

 core in the North Pacific, operate on the popu- 

 lation after some spawning has taken place. In 

 these instances, the effects of fishing genetic 

 selection would be materially reduced. 



SUMMARY REMARKS 



A decline in the apparent abundance of alba- 

 core in the North Pacific longline grounds is 

 evident. Effort statistics do not appear suffi- 

 ciently precise to quantitatively relate to this 

 decline. The decline does not appear to be re- 

 lated to changes in average length for the data 

 which we have available nor do the positive and 

 negative deviations from the long-term trends 

 in any year appear to be related to fluctuations 

 in size for that year. Thus, the evidence that 

 these deviations might owe to fluctuations in 

 year-class strength is not strong. Based on 

 the migratory route postulated in this paper, 

 the fishing intensity in the pole-and-line fishery 

 might have a stronger effect on the longline 

 fishery than fishing effort in the longline fish- 

 ery per se. If the longline fishery is strongly 

 size-selective then an increased fishery-induced 

 mortality rate would reduce the average size of 

 the fish in the population and the consequences 

 of this reduction would be a reduction in long- 

 line CPUE without a reduction in average size. 

 The complicated interactions of the various 

 fisheries make the assumptions used for com- 

 puting the usual steady-state rates tacit over- 

 simplifications. The spatial statistics also do 

 not show any obvious relation to the changes in 

 CPUE. The spatial statistics do indicate, 

 however, interesting features of albacore biol- 

 ogy. These features involve the predictable 

 spatial position of the albacore, which suggests 

 need for a detailed study of the biological and 

 physical oceanography of the area that the al- 

 bacore move through, and the relatively slow 

 rate of apparent motion, which suggests, upon 

 comparison with the relatively rapid velocities 

 that tuna are capable of achieving, that the path 

 of the albacore on the longline grounds is es- 

 sentially nonlinear. This nonlinear path may 

 indeed, be highly convoluted, a circumstance 

 which could account for a higher rate of hook- 

 fish contacts. 



The key to understanding the dynamics of the 

 reaction of the various North Pacific fisheries 

 for albacore upon fishing depends to a large 

 extent in understanding the amounts of fishing 

 intensity applied to each age group. In order to 



35 



