ANKHNHKANPT: R)()I) HAHl'I'S ( )K SKIIMACK Tl'NA 



IRI and M\'RM were chub mackerel, Scomber 

 japonicus. and Thyrsitops lepidopoides*^ (Fig. 3). 

 Crustaceans other than A\ fiirriilis occurred frequent- 

 ly in the stomachs (F = 22.6%), but as a relatively 

 low percentage of the total volume {V = 2.0%). 

 Cephalopods were usually insignificant in the diet 

 (see below). Pteropods, siphonophores, beetles, 

 rocks, and unidentified materials were the consti- 

 tuents of the miscellaneous category (App. Table 1). 



Seasonal Variations 



The data were divided into four quarters: January- 

 March 1982 (I), April-June 1982 (II), July-September 

 1982 (III), October-December 1981 and 1982 (IV). 

 The results (App. Tables 2-6) are illustrated in 

 Figures 4 and 5 both with the IRI and the MVRM of 



•^Identified by Y. Matsuura, Universidade de Sao Paulo, Institiito 

 Oceanografico, Sao Paulo, Brazil, October 1984. 



dominant food items in each quarter. Note that items 

 important in one quarter may be negligible or absent 

 in another. When evaluated in terms of MVRM, the 

 prey ranks sometimes did not coincide with those 

 determined by the IRI (Figs. 4, 5). Based on the IRI, 

 E. similvi was the dominant food in the first quarter, 

 followed by other fish and M. muelleri (Fig. 4). 

 According to the MVRM, other fishes, other crusta- 

 ceans, and E. similis were ranked first, second, and 

 third in importance, respectively. The importance of 

 E. similis in this quarter based on the IRI was exag- 

 gerated by their high frequencies of occurrence. 

 Scomber' japonicus and frigate tuna, Auxis thazard, 

 were secondary in importance to M. muelleri as the 

 main fish species consumed. 



The rankings of the food categories in the second 

 quarter were the same for both the IRI and the 

 MVRM (Fig. 4). Euphausia similis and M. muelleri 

 were the dominant food items, followed by Caranx 

 ruber. 



Figure .5. -Relative importance based on mean volumetric ratio of selected food items for skipjack tuna grouped by length for each quarter 

 (I-IV) and all months. * indicates a significant difference (P < 0.05) in the mean percent volume of that food item by length when tested with 

 the Kruskal-Wallis nonparametric one-way analysis of variance. 



383 



