FISHERY BULLETIN: VOL. 86, NO. 4 



for fish prey and 0.35% BWD for squid prey. Since, 

 fish and squid each constituted about one half of the 

 stomach contents by weight in this size group, the 

 actual daily ration lies within the range of the two 

 values, 0.64% and 0.35%. 



DISCUSSION 



This study demonstrates size-dependent prey pref- 

 erence by Greenland halibut; halibut <20 cm fed 

 primarily on euphausiids whereas those >20 cm 

 were largely fish and squid eaters. It is not surpris- 

 ing to find that walleye pollock was the dominant 

 prey (Table 1) of Greenland halibut because the esti- 

 mated biomass of walleye pollock in the Bering Sea 

 area is about 10,000,000 metric tons (Bakkala and 

 Wespestad 1983). Livingston et al. (1985^, 1986) 

 noted that walleye pollock is a major food source 

 not only for marine birds, marine mammals, and 

 man, but also serves as a major food source for domi- 

 nant components of the eastern Bering Sea ground- 

 fish complex. Other studies have also shown the 

 importance of walleye pollock as food of Greenland 

 halibut (Moiseev 1953: Mito 1974; Smith et al. 1978) 

 in the eastern Bering Sea. 



In this study, large Greenland halibut (>70 cm) 

 ate fish almost exclusively. Bowering and Lilly 

 (1985) found that 65-69 cm was the length at which 

 Greenland halibut in the northwestern Atlantic 

 began to switch from smaller pelagic fish {Mallotus 

 villosus) to larger groundfish {Gadus m.orhua, Sebas- 

 tes sp., Anarhichadidae, Pleuronectidae, Zoarcidae) 

 as food. Mikawa (1963) noted increased piscivory 

 with size in Greenland halibut sampled in several 

 areas of the North Pacific. Mito (1974) also showed 

 the same trend; he found that 65-90 cm Greenland 

 halibut ate 20-40 cm long walleye pollock. These 

 observations suggest that large Greenland halibut 

 (^70 cm) feed on larger sized groundfish which may 

 be lower in the water column whereas the smaller 

 Greenland halibut (<70 cm) feed on smaller sized 

 pelagic fish in the upper water column. 



Shuntov (1970) and Mikawa (1963) noted seasonal 

 depth migrations for Greenland halibut and inter- 

 preted the summer movement into shallower waters 

 as a feeding migration related to migrations of 

 walleye pollock. Based on the size distribution of 



^Livingston, P. A., M. S. Yang, and D. Wencker. 1985. The 

 importance of juvenile pollock in the diet of key fish species in the 

 eastern Bering Sea. Unpubl. manuscr., 19 p. Presented as the 

 workshop on comparative biology, assessment, and management 

 of gadoids from the North Pacific and Atlantic Oceans, 24-28 June 

 1985. Northwest and Alaska Fisheries Center, National Marine 

 Fisheries Service, NOAA, Seattle, WA 98115. 



walleye pollock in midwater trawl catches near the 

 Pribilof Islands and westward over the Aleutian 

 Basin, Livingston and Dwyer (1986) found that small 

 (age 0) pollock occurred in near-slope and shelf 

 areas, medium (age 1) pollock in shelf areas, while 

 larger (>1 year old) pollock occurred in all areas dur- 

 ing summer. Therefore, it can be concluded that in 

 the slope area, where juvenile walleye pollock (age 

 and 1) of the appropriate size for smaller Green- 

 land halibut were not available, the smaller sized 

 (30-69 cm) Greenland halibut ate the available prey, 

 cephalopods and deep-water fishes, while the larger 

 ones (^70 cm) consumed mostly larger walleye 

 pollock (>30 cm) and other fish regardless of depth 

 or season. 



No clear diel feeding trends were found. The lack 

 of trends may be related to the large variations of 

 the time of sunrise and sunset in different seasons 

 in the Bering Sea. Other studies have varied find- 

 ings. Mito (1974) reported that Greenland halibut 

 fed primarily from sunset to midnight based on 

 hmited sample sizes (six specimens in some time 

 periods). Shuntov (1970) showed that this species 

 fed continuously in the Okhotsk Sea, although feed- 

 ing was somewhat higher during the night. By com- 

 paring day and night catch rates, Chumakov (1969) 

 concluded that Greenland halibut (in the Iceland 

 area) made daily vertical migrations (staying close 

 to the bottom during the day and moving up in the 

 water column at night). However, he did not cor- 

 relate this behavior with diel feeding. Thus, the 

 literature and this study show no definite diel feed- 

 ing trend in Greenland halibut. 



Daily Ration 



Daily ration calculations were based on the evacu- 

 ation rate of one prey item (pollock or squid) using 

 Elliott and Persson's (1978) model. Other authors 

 (Durbin et al. 1983; Dwyer 1984) have calculated 

 total daily ration by adding up the separate daily 

 rations of the different prey items. Persson (1984) 

 demonstrated that the evacuation of a specific food 

 item can be dependent on the ingestion of other food 

 items. Therefore, it may be erroneous to apply the 

 food consumption model to estimate the consump- 

 tion of individual prey types separately. Persson 

 (1984) suggested that the only practical solution to 

 calculate the daily rations of different prey items 

 is to calculate the mean weight of each food item 

 remaining in the digestive tract over 24 hours and 

 multiply the fraction it constitutes of the total mean 

 content with the total daily ration. This is necessary 



688 



