FISHERY BULLETIN: VOL. 78, NO. 3 



large prey: Pleuromamma xiphias, P. ah- 

 dominalis, and large aetideids, were eaten fre- 

 quently. Several small to medium copepods, the 

 most frequent of which was Corycaeus, had inter- 

 mediate ASV's (0.41-0.70 m^). Very few prey had 

 low ASV's; half of these were ostracods and 

 euphausiids. 



Diaphus fragilis (Table 10) 



Few D. fragilis were available, and number of 

 prey items was low. The data are most similar to 

 those for D. schmidti. Microzooplankton ac- 

 counted for about one-half the diet. The corrected 

 ASV for Oncaea spp., the dominant microzoo- 

 plankton, and those of most macrozooplankton 

 were low. Only two prey types — each taken only 

 once — had ASV's over 0.40 m^. 



Diaphus trachops (Table 10) 



Data for D. trachops are few, but indicate that its 

 diet is quite different from those of the other 



Table ll. — Stomach contents of Melamphaes danae and Breg- 

 maceros japonicus . Format as in Table 4. 



Diaphus spp. in that few microzooplankton were 

 eaten. Most prey items were medium to large 

 forms and had high ASV's. 



Melamphaes danae (Table 11) 



Microzooplankton made up minor fractions of 

 the diet of M. danae; most were either small os- 

 tracods or gastropods. Among the other prey only 

 chaetognaths and euphausiid larvae were consis- 

 tently important. At 70 and 90 m, 22 and 26% of 

 the prey had high ASV's; most other types had low 

 values (Table 2). At 110 m, the great majority of 

 prey types and items had high ASV's. For most 

 prey types, the ASV's at different depths were 

 either consistently high (euphausiid larvae, 

 A^eocaZa^Ms, amphipods) or low (ostracods), but the 

 value for chaetognaths at 110 m was much higher 

 than shallower, 



Bregmaceros japonicus (Table 11) 



Bregmaceros japonicus ate few microzooplank- 

 ton; small macrozooplankton were also taken in- 

 frequently and usually with low ASV Most prey 

 were medium to large and, except for chaeto- 

 gnaths and Euphausia spp., ASV's were moderate 

 to high. 



DISCUSSION 



The fishes considered here clearly showed pref- 

 erence in a broad sense, i.e., some abundant zoo- 

 plankton were rarely or never taken, and the 

 ASV's of types regularly eaten were variable. 

 Though there were exceptions, these fishes gener- 

 ally grazed on relatively large, visible crustaceans. 

 Other taxa were rarely taken and then usually 

 with low ASV's. Most other taxa in the plankton 

 were either translucent forms, e.g., chaetognaths 

 and tunicates, or quite small, e.g., gastropod 

 veligers. Among the crustacean microzooplank- 

 ton, the densely pigmented and relatively 

 opaque Oncaea spp. were the only types that were 

 taken regularly and had high ASV's. Some appar- 

 ently less visible forms such as Clausocalanus and 

 small Scolecithricella spp. were abundant in the 

 plankton samples (in spite of mesh escapement), 

 but rarely eaten, and the undoubtedly more 

 numerous smaller types which mostly passed 

 through the plankton net mesh were absent from 

 almost all the fishes' diets. Among the crustacean 

 macrozooplankton, several translucent or weakly 



632 



