180° 178° 176° 174° 172° 170° 168° 166° 164 162° 160° 158° 156° 



Figure l. — Location sites where walleye pollock stomach samples were collected in the eastern Bering Sea. 



Table l. — Summary of walleye pollock stomach samples col- 

 lected in the eastern Bering Sea. 



tifiable matter was classed as "digested material" 

 and also weighed. Percentage of food weight for 

 each major food category, by fish-length group, 

 was calculated as was the weight for each major 

 food category per fish for each length group. 

 Empty stomachs were not included in the 

 analysis. 



Detailed length data from foreign fishing ves- 

 sels were available only from the Japanese fishing 

 vessel Tenyo Maru. These data were analyzed by 

 10-cm fork length classes. Fish lengths from the 

 Japanese fishing vessel Chikuhu Maru and the 

 Soviet fishing vessel Tiraspol were recorded only 

 as greater or less than 35 cm (the approximate 

 length at sexual maturity). This is also the size at 

 which pollock become markedly cannibalistic 



(Takahashi and Yamaguchi 1972). Data from all 

 three observer cruises were combined using these 

 two major size categories to obtain sufficient sam- 

 ple sizes for comparison with the data collected in 

 1974. 



Data collected in 1974 (RV Oregon) were 

 examined by 5-cm length classes. The larger 

 number of stomach samples collected during this 

 cruise allowed a finer analysis of size-related 

 changes in feeding habits. The methods used for 

 processing samples from this cruise were approx- 

 imately the same as for samples from the foreign 

 vessels. 



Results 



An examination of stomach content weight by 

 fish-length group provided evidence of related 

 shifts in principal food categories in the diet of 

 pollock (Figures 2, 3). In both spring 1977 and 

 summer 1974, the percentage of copepods as food 

 biomass tended to decrease with increasing size of 

 pollock. The percentage offish in pollock stomachs 

 tended to increase with the size of pollock. 

 Euphausiids were important food components in 

 most length classes in both sampling periods. Am- 

 phipods, however, were only abundant in 



305 



