dominant forage food items (fish and crusta- 

 ceans) in the stomachs of skipjack tunas was 

 correlated with the size of tunas. Spearman's 

 rank correlation coefficients of 0.771 for fish 

 and -0.771 for crustaceans were not significant. 

 I did not make a similar analysis of the sizes 

 of yellowfin tunas and the frequency of occur- 

 rence of the three main food categories because 

 the number of size classes (four) was too small. 



Spearman's rank correlation analysis was 

 also applied to test whether the percentage 

 volumes of the forage food categories of skip- 

 jack and yellowfin tunas were correlated with 

 the size of fish (table 3). All fish suitable for 

 such a test (487 skipjack and 338 yellowfin 

 tunas) were grouped into 10-mm. length in- 

 tervals. Correlations were performed for the 

 two dominant food categories (fish and crusta- 

 ceans or mollusks). The data from single 

 cruises were treated separately, because each 

 cruise is representative of the same general 

 area and the same time. I observed a significant 

 correlation (Spearman's rank correlation co- 

 efl^'icient of 0.512, P<0.05) only between yellow- 

 fin tuna and fish on one cruise, indicating that 

 as the size of tuna increased, the percentage 

 consumption of volume of forage fish increased 

 also. 



The size ranges of skipjack (410-639 mm. 

 fork length) and yellowfin tunas (390-629 mm. 

 fork length) were for the most part probably 

 not great enough to relate possible differences 

 in feeding habits to size. Even so, the fact that 

 one significant correlation was noted for yellow- 

 fin tuna may indicate that larger tunas consume 



Table 3. — Spearman's coefficients of rank correlation 

 (r,) between the lengths of skipjack and yellowfin 

 tunas and percent volumes of the dominant food 

 categories for several cruises 



more fishes than do the smaller ones. The 

 greater dependence of larger skipjack and yel- 

 lowfin tunas on fish and of smaller skipjack 



tuna on crustaceans was also observed by other 

 investigators (Nakamura, 1965; Yuen, 1959; 

 and Alverson, 1963). 



King and Ikehara (1956), who analyzed the 

 relation between the volume of stomach con- 

 tents and body weights for bigeye and yellow- 

 fin tunas from the central Pacific, found that 

 the mean volume per stomach for both species 

 increased with the size of the fish and the 

 average content of stomachs per unit of body 

 weight decreased with increase in size. Because 

 about 98 percent of the specimens I studied 

 fell into a narrow range of sizes (2-4 kg.), I 

 was unable to make a similar analysis, but I 

 did find — from examinations of scatter dia- 

 grams — that the relation between the volume 

 of the stomach contents and the weight of the 

 fish appeared to agree with King and Ikehara's 

 results. 



VARIATION IN VOLUME OF 

 STOMACH CONTENTS 



As indicated in the studies by Magnuson 

 (1969), the maximum capacity of the stomach 

 of a skipjack tuna is about 7 percent of the 

 weight of the fish, but during 1 day a fish can 

 consume the equivalent of 15 percent of its body 

 weight. On the basis of Magnuson's study, I 

 assume that the maximum capacity of yellowfin 

 tuna stomachs is also about 7 percent of the 

 body weight. 



The volume of stomach contents was less 

 than 20 ml. in 75 percent of the skipjack tunas 

 and in 85 percent of the yellowfin tunas. Stom- 

 ach volumes above 20 ml. were higher in areas 

 1 and 2 than in area 3 — an observation that 

 might be related to the fact that the diet of 

 skipjack and yellowfin tunas in areas 1 and 2 

 consisted primarily of fish, which have higher 

 average volumes than crustaceans. In their 

 study of food of yellowfin tuna in the central 

 Pacific, Reintjes and King (1953) found that 

 food volume was less than 25 cc. in 58 percent 

 of the stomachs of yellowfin tunas. Assuming 

 that 1.0 ml. of stomach contents is equivalent 

 to 1.0 g., I made a rough comparison between 

 the estimated weight of the stomach con- 

 tents and the body weight of tunas. ]\Iy 

 calculations showed that in 99 percent of the 

 observations the stomach contents were well 



SKIPJACK AND YELLOWFIN TUNA FOOD IN ATLANTIC OCEAN 



453 



