FISHERY BULLETIN: VOL. 78, NO. 3 



high, indicating a real preference for Oncaea. Ex- 

 cept for three prey types not taken by the plankton 

 tows, the ASV's for other microzooplankton are 

 low even without any adjustment for undersam- 

 pling. 



Except for the large fish from 110 m, macrozoo- 

 plankton were taken very infrequently and mostly 

 had low ASV's. The large fish from 110 m had eaten 

 Pleuromamma and candaciids frequently, and this 

 was the only group from which euphausiids were 

 recorded. The data indicate some preference for 

 candaciids. ASV's for these copepods were high for 

 the large fish from 110 m and sometimes fairly 

 high in other groups. 



Diogenichthys atlanticus (Table 8) 



About three-fourths of the items eaten by D. 

 atlanticus were microzooplankton — mostly On- 

 caea spp. The ASV for the grossly undersampled 

 small Oncaea is meaningless, but if ASV's for the 

 other microzooplankton are reduced by a factor of 

 4, there is reasonable indication of preference for 

 the large Oncaea spp. and Acrocalanus spp. Most 

 of the macrozooplankton were small to medium 

 copepods, and ASV's of most types were low. 



Benthosema suhorhitale (Table 9) 



Benthosema suhorhitale usually does not occur 

 as deep as 110 m, but the number collected at that 

 depth was considerably larger than that expected 

 from catches in transit. Thus the data are probably 

 not seriously affected by fish caught at shallower 

 depths. The sample from 90 m, which was taken at 

 a different time of the year, had too few B. suhor- 

 hitale to merit analysis. 



Microzooplankton were important fractions of 

 the diet of B. suhorhitale; they made up over half 

 the items from the small fish and slightly less for 

 the larger ones. Almost all were Oncaea spp. — 

 mostly the larger forms. Macrozooplankton were 

 mostly medium to large copepods, but also in- 

 cluded euphausiids and large amphipods. Such 

 prey, especially P. xiphias and candaciids, were 

 eaten more frequently by the large fish from both 

 depths. ASV's for most macrozooplankton prey 

 types were 0.40 m^ or less. If the ASV's for the 

 large Oncaea spp. are reduced by a factor of 4, they 

 are commensurate with those of the macrozoo- 

 plankton. 



630 



Table 9. — Stomach contents of Benthosema suhorhitale. For- 

 mat as in Table 4. 



Diaphus schmidti (Table 10) 



The numbers of prey per fish and diversity of 

 prey were relatively high for D. schmidti; several 

 small copepods and noncrustacean prey that were 

 either rare or absent in the diets of other species 

 were taken relatively frequently. 



Microzooplankton made up SO-SO^f of the items 

 (Table 2); half to two-thirds of these were Oncaea. 

 If ASV's for Oncaea are roughly corrected, they 

 are still quite high. ASV's for other types of micro- 

 zooplankton were variable. 



Although the composition of macrozooplankton 

 prey was generally similar for all groups, there 

 were some differences between sizes or depths. 

 Pleuromamma and Euphausia spp. were eaten 

 more frequently at 70 m than at 90 m. Overall, 

 Corycaeus spp. were the most frequently eaten 

 prey, but at both depths, frequency and ASV's were 

 higher for the small fish. About half the prey of the 

 small fish from 70 m had high ASV's. These were 

 mostly Corycaeus spp., but also included several 

 medium to large prey types. Among the large fish 

 from 70 m, a few types of large prey had high 

 ASV's, but most prey from both these and both 

 groups from 90 m had low ASV's. The generally 

 higher ASV's associated with the small fish from 

 70 m appear to have resulted mostly from higher 



