were due to nonrecording or to absence of fish. The 

 tests indicated that the catch rates, and therefore the 

 recording of skipjack tuna caught, were not significantly 

 different (Table 2) between large and small vessels. 

 Hence, it seems unlikely that smaller vessels fishing 

 only in the western Pacific had biased the catches geo- 

 graphically. 



Finally, because longliners are fishing for species of 

 tunas other than skipjack, the selection of fishing sites 

 and competition for hooks could affect the catches of 

 skipjack tuna adversely. Any such effect, however, would 

 lose its importance for this study, since in most areas 

 where other tunas are abundant, skipjack are also pre- 

 sent, and competition with other tuna species would 

 occur throughout the range of skipjack tuna distribution. 



In summary, although the longline catch data for 

 skipjack tuna are affected by certain characteristics of 

 the gear and of the fishery, they appear adequate for 

 my purpose. 



Table 2. — Probabilities (two-tailed) of differences in recording skipjack 

 tuna caught on longline between small (20-50) and large (50-200) vessels 

 by quarters (vessel sizes in gross metric tons). The numerals in 

 parentheses indicate vessels having recorded significantly more skipjack 

 tuna: (1) small vessels, (2) large vessels. 



DISTRIBUTION OF EFFORT 



Rothschild (1966) published a summary of effort of 

 the Japanese longline fishery for 1953-63, in which each 

 year's distribution of effort in each 20° area was given 

 as a percentage of that year's total effort. His report 

 shows a gradual eastward expansion of the fishery, and 

 although fishing had approached the coastal waters of 

 North, Central, and South America by 1962, it was not 

 untU the following year that adequate coverage was 

 attained in the eastern Pacific. In my study the quarter- 

 ly effort in numbers of hooks for each 5° area was 

 averaged for 1964-67 and contour lines were drawn at 

 100,000-hook intervals (Fig. 3). 



The figure shows that heavily fished areas are lo- 

 cated near Japan in all four quarters, off southeastern 

 Australia in the second and third quarters, and in the 

 eastern Pacific in the first, third, and fourth quarters. 

 Near Japan, the concentration of effort is not only the 

 highest, but remains high throughout the year because 

 the area is accessible to large vessels as well as small 

 ones that make repeated trips during the year. The 

 concentration of high effort shifts from south of Japan in 

 the first and second quarters to east of Japan in the 

 third and fourth quarters, reflecting the seasonal move- 

 ment of the albacore, Thunnus alalunga, (Van Campen 

 1960) and the bigeye tuna, T. obesus, which are the 

 main species of tunas fished by the longliners in this 

 region. Both the route and time of the shift generally 

 agree with that of skipjack tuna caught by surface gear 



within the area (Kimura 1949). The heavy effort off 

 Australia in the second, third, and perhaps the fourth 

 quarters is expended mainly for southern bluefin tuna, 

 T. maccoyii, and albacore (appendix figures 2 and 5; 

 Fisheries Agency of Japan, Research Division, 1967a, 

 1967b, 1968, 1969); whUe in the eastern Pacific, fishing 

 is concentrated off the northern coast of South America 

 in the first and fourth quarters for bigeye tuna and 

 striped marlin, Tetrapturus audax, and off Baja Cali- 

 fornia in the third and fourth quarters for striped marlin 

 (Kume and Joseph 1969). 



The effort is more evenly distributed, though at 

 lower levels, over the equatorial region between lat. 

 20°N and 20°S. Moderate fishing prevails in an almost 

 continuous band along the equator from the eastern to 

 the western Pacific for the greater part of the year. 

 This band is interrupted over a relatively short distance 

 only during the last quarter. 



RELATIVE ABUNDANCE OF 

 SKIPJACK TUNA 



Distribution of Catch Per Unit Effort 



To show the relative abundance of skipjack tuna 

 taken by longlines, the catch per unit effort (CPUE), 

 expressed here as number of skipjack tuna caught per 

 1,000 hooks per quarter in all 5° areas fished, was 

 calculated and plotted for the 4 yr. Because of skewed 

 distribution of the catches, the geometric mean was 

 calculated for each quarter's catches and contour lines 

 were drawn to emphasize high-CPUE areas (Figs. 4-7). 



The distribution of CPUE (Figs. 4-7) indicates wide- 

 spread occurrence of skipjack tuna in the Pacific Ocean. 

 The north-south distribution is expanded latitudinally in 

 the west due largely to the greater poleward dispersion 

 there of warm water inhabited by the tunas. The 

 distribution is generally narrow in the eastern Pacific 

 where cooler waters flowing toward the equator tend to 

 restrict the poleward dispersion of the skipjack tuna. 

 The predominance of zero-catch areas (indicated by -'s) 

 at both the northern and southern fringes during most 

 quarters assured that fishing had occurred over the 

 whole range of adult skipjack tuna distribution. 



The contoured areas of above-average CPUE appear 

 either singly or in clusters in various parts of the 

 Pacific. To facilitate the discussion the Pacific Ocean was 

 partitioned into western (A), central (B), and eastern (C) 

 regions (Figs. 4-7). Boundary lines were drawn as close 

 as possible along lines of apparent breaks in the distri- 

 bution of above-average CPUE areas. 



In region A, north of the equator, high catch rates 

 occurred sporadically; in 1964, for example, they occur- 

 red in all four quarters, but in 1965-67 in only two 

 quarters. The above-average CPUE cells there were 

 small and seemed to be associated with seasons and 

 areas: in the second and third quarters off Japan, and 

 in the first, second, and fourth quarters east of the 

 Philippine Islands. South of the equator high-CPUE cells 

 generally occurred off northeastern Australia in the 

 fourth and first quarters and off southeastern Australia 

 in the second and third quarters. Cells of high CPUE, 

 however, seemed less consistent from year to year off 

 Australia than in the adjacent region B. In 1966, for 

 example, high CPUE occurred only in the first quarter. 



