variation? woxild then expose fish 

 feeding at different times to 

 different food supplies. 



The first objection to the assumption 

 is the one most esisily overcome. Variation 

 between the stomach contents of the same 

 species taken in the same area was small. 

 This lack of variation probably makes the 

 sampling size adequate. 



The second criticism is not so easily 

 overcome. All evidence at present points 

 to the fact that the bulk of the salmon in 

 oceanic watei^ are found in the surface 

 layers (Bamaby, 1952; Fukuhara, 1953; Tano- 

 naka, 1955). If the differences in feeding 

 behavior were due to the fish feeding in 

 areas other than their point of capture, it 

 is probable that the differences in feeding 

 were due to the salmon's horizontal rather 

 than vertical distribution. If this criti- 

 cism to the hypothesis is valid, then a 

 wide variation in feeding behavior in the 

 different areaa would have been expected; 

 however, the stomach content of fish from 

 various areas of a single region showed 

 little such variation. Different species 

 exposed to the same food supply consistently 

 fed differently. From area to area, for 

 example, the sockeye was consistently more 

 selective than the chum salmon. 



of the different food organisms. In the 

 Inshore waters amphipods and, to a lesser 

 degree, crustawean larvae formed the major 

 portion of the food supply. As the waters 

 became more oceanic, copepods and euphau- 

 slids assumed increasing importance. 

 Amphipods seemed ubiquitous, appearing 

 with great frequency in all regions. Cope- 

 pods and euphausilds, however, assumed 

 roles of importance only in the more 

 oceanic waters. This does not mean that 

 euphausilds and copepods may not have been 

 in the Inshore waters. It meems only that 

 they were not utilized by the salmon in 

 the inshore waters at the time of the year 

 when the samples were taken. 



Qualitatively, the items consumed by 

 the three species were similar. The dif- 

 ferences established were mainly quantita- 

 tive. Size of food items also was involved 

 in feeding behavior (table 7)' Chum salmon 

 tended to feed more on the larger forms 

 (squid, fish, and euphausilds). Sockeye 

 fed more on the smaller foims (amphipods 

 and copepods). Pink salmon were Intenne- 

 dlate in degree of selectivity of food 

 items. Table 7 also indicates that the 

 larger food organisms became more important 

 in the diet of all three species as the 

 area became more oceanic. 



The third objection cannot be resolved 

 at the present stage of our knowledge. 

 There was seme indication, in fact, that 

 diet differences were due to time differ- 

 ences In feeding. The contents of chum 

 salmon stomachs were, on the average, f\ir- 

 ther digested than the contents of either 

 the pink or the sockeye. This may be due 

 to differences in digestive rates, but it 

 may be due also to a difference in the 

 feeding time. Further information concern- 

 ing the digestive rates of the Bacific 

 salmon is necessary to determine the validi- 

 ty of this objection. 



Even if the objections are valid, the 

 fact remains unaltered that different spe- 

 cies of salmon feed differently. The data 

 seem to indicate that the sockeye Is the 

 most selective and the chum the most omni- 

 vorous feeder of the three species. The 

 feeding behavior of the pink salmon seems 

 closer to the chum than to the sockeye. 

 Feeding behavior of fish in different 

 regions was also different. This, of course, 

 is a reflection of the varying availability 



Table 7. — Percentage contribution by volume 

 of fish, squid, and euphausilds 

 to the diet of the sockeye, pink, 

 and chum salmons 



Species 



Region A Region B Region C 



It Is emphasized that the data pre- 

 sented were only for a portion of one year. 

 It may be expected that seasonal variation 

 pleiyed an important part in determining 

 the availability of food and the resulting 

 diets of the salmon. In Region A, for 

 example, crustacean larvae were an impor- 

 tant part of the fishes' diet for the time 

 of the year under consideration. These 

 larvae, however, are meroplankton and it 

 may be expected that their availability 

 throughout the year would fluctuate 



10 



