FISHERY BULLETIN: VOL. 73, NO. 4 



albacore in each of the latitudinal subdivisions. In 

 the areas north of lat. 20°S, both male and female 

 length-frequency distributions had a single well- 

 defined mode. In the areas south of lat. 20°S the 

 modes were less well defined. Koto and Hisada 

 (1967) found a similar pattern in the length- 

 frequency distribution of albacore in the South 

 Pacific in 1961. 



Otsu and Sumida (1968) computed the mean 

 length of albacore in the fishery during the period 

 from 1962 to 1965. They divided the fishery into 5° 

 bands of latitude and noted that the fish were 

 smallest near the equator, largest at lat. 20° to 

 25°S and tended to be smaller again south of lat. 

 25°S. The results agree in general with those of 

 Otsu and Sumida; however, as seen above, the 

 length-frequency distributions indicate a more 

 complex situation than do the mean sizes. North of 

 lat. 20°S the mean sizes may be good indicators of 

 the general size of albacore because the length- 

 frequency distributions showed that the fish were 

 composed of single uniform size groups. South of 

 lat. 20°S, the catches were composed of several size 

 groups of fish and the mean does not indicate the 

 presence of different size groups of fish. For 

 example, although Otsu and Sumida stated that 

 albacore south of lat. 25°S tended to be smaller, my 

 data show that large fish were also present in these 

 latitudes. 



Although it is not readily apparent in the 

 length-frequency distributions, there appears to 

 be a declining trend in the mean size of albacore in 

 the fishery. Otsu and Sumida (1968) noted that 

 albacore taken in 1964 and 1965 were shorter on 

 the average than those caught in 1963. My data 

 show that the declining trend in the mean length 

 of albacore has continued (Table 3). 



Table 3.-Mean lengths of albacore, sexes combined, 1963-71. 



The mechanisms that produced such a pattern in 

 the distribution of sizes in the fishery are probably 

 complex. The unique character of the length- 

 frequency distributions among the four lati- 

 tudinal bands must have resulted from some 



nonrandom distributional process. The larger 

 numbers of smaller albacore in the more southern 

 waters suggest that albacore are initially recruit- 

 ed into the fishery in the area south of lat. 20°S. 

 The high CPUE experienced in the second and 

 third quarters south of lat. 20°S may be an in- 

 dication of this. Also, it has been shown that 

 juvenile albacore which originate from spawning 

 that takes place north of lat. 20°S migrate south as 

 they grow larger (Yoshida 1971). 



SUMMARY 



A comparison of various indices of apparent 

 abundance of albacore indicated that catch per day 

 and catch per 100 hooks were better indicators of 

 apparent abundance than catch per trip. The mean 

 catch per day and catch per 100 hooks of albacore 

 have generally declined over the years, which 

 suggests that the apparent abundance of albacore 

 has declined in the American Samoa longline 

 fishery. To compensate for the reduced CPUE, the 

 longliners fished more days per trip and traveled 

 farther from the home base seeking areas of good 

 catch rates. 



The fishery apparently has had an effect on the 

 albacore stock. Although the annual landings have 

 continued to increase with increased fishing ef- 

 fort, the CPUE has declined. That the apparent 

 overall effect was not greater was due to the fact 

 that the fishing grounds have expanded, especially 

 into areas south of lat. 20°S where good catch rates 

 were obtained. The mean annual CPUE plotted 

 against fishing effort for selected, discrete areas 

 north and south of lat. 20°S indicated that the 

 fishery has not as yet had as great an effect in the 

 south as it has to the north. There are at least two 

 possible reasons for the better condition of the 

 fishery to the south. First, the area south of lat. 

 20°S has not been exploited as long as the area to 

 the north. Second, it was shown that the albacore 

 are first recruited into the fishery in the latitudes 

 south of lat. 20°S, which may account in part for 

 the higher apparent abundance. 



There was some indication that there were 

 temporal changes in apparent abundance of al- 

 bacore south of lat. 20°S. Because of poor weather 

 conditions or because the fishermen have learned 

 through experience that catch rates are better 

 during certain seasons, or a combination of these 

 and other reasons, fishing effort expended south 



764 



