DIEL FEEDING PATTERNS OF 16 SPECIES OF 

 MESOPELAGIC FISHES FROM HAWAIIAN WATERS 



Thomas A. Clarke^ 



ABSTRACT 



Diel patterns of stomach fullness, as percent of dry weight, were determined for 16 species of 

 mesopelagic fishes. Nine species of myctophids and one melamphaid, all vertical migrators, appeared 

 to feed solely or principally at night in the upper layers. These species encountered higher tempera- 

 tures and prey concentrations at night. Four species of stomiatoid fishes appeared to feed during the 

 day regardless of the extent of their migration or the absence thereof Prey concentrations encountered 

 by the stomiatoids during the daytime appeared to be higher than or similar to those encountered at 

 night. One myctophid and one gonostomatid showed no diel pattern; diel changes in the environmental 

 factors considered were relatively small in spite of the fact that both species undertook limited vertical 

 migrations. 



Crude estimates of instantaneous evacuation rate and daily ration were made from data for four 

 species. These indicated that evacuation rate was increased at night in the upper layers and that daily 

 rations of species which migrated into the upper layers were similar to values for shallow-living 

 zooplanktonivores, while rations of deeper living species were lower. Thus while the adaptive value of 

 upward migration in the species which feed at night is obviously related to feeding activity, the upward 

 ascent by the daytime feeders may allow processing of larger daily rations than if they remained at low 

 temperature all day. 



The extensive diel vertical migrations of certain 

 mesopelagic fishes have been well documented in 

 a vanety of oceanographic situations. While a 

 number of theories have been proposed to explain 

 the adaptive value of the behavior — in both fishes 

 and migrating invertebrates as well — data to 

 support any of them are few. One of the most 

 frequently proposed hypotheses (e.g., Marshall 

 1960) is that the organisms ascend at night to feed 

 in the upper layers where food is presumably at 

 higher concentrations and descend during the day 

 to avoid predation while the upper layers are well 

 lighted. Several studies of mesopelagic fishes (to 

 be cited below) have considered the relationship 

 between feeding chronology and vertical distribu- 

 tion in an effort to support at least one-half of the 

 hypothesis, but the results have for the most part 

 been rather equivocal. Apparent diel trends in 

 stomach fullness or details thereof are often ques- 

 tionable owing to low numbers of specimens 

 examined, insensitive methodology, or incomplete 

 diel coverage. Furthermore, all such studies on 

 mesopelagic fishes, with the exception of Merrett 

 and Roe (1974), have been conducted in high 



'University of Hawaii, Department of Oceanography and 

 Hawaii Institute of Marine Biology, P.O. Box 1346, Kaneohe, HI 

 96744. 



Manuscript accepted March 1978. 



FISHERY BULLETIN: VOL. 76, NO. 3, 1978. 



latitude or neritic situations and have dealt with 

 only one or, at most, three species. 



This study considered the feeding chronology of 

 16 species from 5 families of mesopelagic fishes 

 from the north central Pacific Ocean. Vertical dis- 

 tribution and certain other apsects of the ecology 

 of these fishes are covered in Clarke (1973, 1974) 

 and Clarke and Wagner (1976); results from re- 

 lated investigations in the same study area are 

 summarized in Maynard et al. ( 1975). Comparison 

 of diel patterns of stomach fullness and diel 

 changes in temperature and prey concentration 

 allow consideration of adaptive value of the verti- 

 cal migrations undertaken by most of these 

 species. In four species, rough calculations of daily 

 ration are possible using equations similar to 

 those presented by Eggers (1977). 



MATERIALS AND METHODS 



Field Sampling 



Specimens for this study were all collected with 

 a 3-m Isaacs-Kidd midv/ater trawl ca. 20 km west 

 of the island of Oahu, Hawaii (ca. lat. 2r20-30'N; 

 long. 158"20-30'W) in waters 2,000-3,000 m deep. 

 In order to reduce the concentration of zoo- 

 plankton in the cod end of the net and thus 



495 



