equatorial area. This hypothetical route of the 

 albacore is shown schematically in figure 19. 



DISCUSSION AND EVALUATION OF 

 PRESENT UNDERSTANDING 



Decline in Apparent Abundance 



The apparent abundance of albacore on the 

 North Pacific longline grounds exhibited a 

 marked decline during the study period. The 

 decline generates the question of the extent to 

 which changes in actual abundance, changes in 

 fishing intensity, or changes induced by the 

 environment influenced the reduction in appar- 

 ent abundance of albacore during the study 

 period. Unfortunately, the appropriate data, 

 customarily used for studies of this nature, 

 were not available for the present investiga- 

 tion. As an example, estimates of nominal 

 fishing effort would enable studies of the corre- 

 lation between apparent abundance and fishing 

 intensity. Suda (1963b: p. 1262) presents in- 

 dices of fishing effort for albacore for each 

 North Pacific longline fishing season (1951-59). 

 His estimates are based on estimating the 

 "number of standardized trips in North Pacific 

 and other areas," and the ratio of number of 

 trips in North Pacific to the number of stan- 

 dardized trips. Using these values Suda com- 

 puted an index based on the ratio of the number 

 of trips in the North Pacific to the number of 

 trips standardized to a 1951 base of about 

 14,000 North Pacific trips. For an alternative 

 estimate of effort on the albacore longline 

 grounds, we have used Rothschild's (1966: pp. 

 96-99) estimates of the percent reported long- 

 line effort expended in each 20° quadrangle of 

 the Pacific Ocean in 1953-61 and Otsu and 

 Sumida's (1966) estimates of total Pacific Ocean 

 longline effort for each year. We have applied 

 Rothschild's percentages for quadrangles 9872 

 and 9873 (lat. 20°-40° N., long. 140° E.-180°) to 

 Otsu and Sumida's estimates of total effort to 

 arrive at an estimate in millions of hooks ex- 

 pended in the area lat. 20° to 40° N., long. 140° 

 E. to 180°. These estimates in millions of 

 hooks per year are: 1953, 7.2; 1954, 29.1; 1955, 

 27.2; 1956, 25.7; 1957, 13.3; 1958. 23.8; and 

 1959, 32.6. 



The procedures used by Suda and us to es- 

 timate effort can be discussed from several 

 points of view. Suda's is based on fishing effort 

 applied to other species such as yellowfin or 

 bigeye tuna. His computations are based on a 

 trip as a unit of effort; the "trip" is a less 



fundamental unit of nominal effort for longline 

 fishing than the "hook." His estimates of ef- 

 fort include fishing trips during the summer 

 months when longline fishing for albacore in 

 the North Pacific is essentially not prosecuted. 

 Our estimates are based on a more restricted 

 area than those of Suda but probably include 

 some effort for bigeye and bluefin tunas. We 

 have used the hook as a unit of nominal effort. 

 Our estimates also include effort that is ex- 

 pended outside the peak fishing months, but 

 since our area is restricted to the albacore 

 grounds it is much less likely to include non- 

 albacore effort even though data from the sum- 

 mer months are included in our computations. 

 Unfortunately, our estimates of effort do not ap- 

 pear to be correlated with those given by Suda. 

 Suda's estimates show a tendency for effort to 

 decline during the 1951-59 period. Our esti- 

 mates for the 1953-59 period do not exhibit a 

 clear decline. Our CPUE index shows a nega- 

 tive correlation with our estimates of effort but 

 not with Suda's estimates of effort. Since both 

 estimates are derived through tortuous com- 

 putations, it is difficult to determine, without 

 additional detailed study, which is the better 

 estimate. 



The tentative nature of the estimates of effort 

 creates difficulties in appraising the variables 

 that influence the per-head rate of death. The 

 conclusion that the per-head rate of death in- 

 creases during the 1949-61 period essentially 

 concurs with the studies of Suda that were 

 summarized by Kamimura (1966). Rather than 

 interpret the decline in apparent abundance as 

 linear Suda indicated that the decline was quite 

 rapid during the early 1950's, but by the mid- 

 1950's the decline halted and CPUE became 

 stabilized at a relatively low level. A rapid 

 decline followed by a stabilization of CPUE 

 implies that the regression of CPUE on time 

 should be curvilinear. We have already indi- 

 cated that there is some evidence for curvi- 

 linearity. In addition it is quite obvious that a 

 linear decline in apparent abundance could not 

 last for an indefinite period of time without the 

 stock of fish becoming extinct. In one sense 

 the question of linearity or curvilinearity is 

 somewhat academic for our data since both 

 methods should yield declines in apparent abun- 

 dance that are of the same order of magnitude. 

 This implies, of course, that estimates of the 

 coefficient of total mortality based on either 

 the linear or curvilinear interpretation of the 

 decline in CPUE should be roughly equal. But 



28 



