percentage of the stomachs; however, 

 volumetrically they played a secondary 

 role In the diets. Crustacean larvae are 

 not found in any of the sockeye or pink 

 salmon stomachs, although they did appear 

 in 10.8 percent of the chum salmon. Fish 

 were of lesser Importance to the sockeye 

 than to the pink and chum salmons. 



Identification of food items 

 in the stomach contents 



Complete species identification was 

 not possible at this time. Some of the 

 more important food items, however, have 

 been identified. Of the copepods, the 

 greatest number were of the genus Calanus . 

 C. eristatus was a very common species, 

 perhaps even the predominant species. A 

 few specimens of another copepod, Metridia 

 lucens , were also identified. None of the 

 Euphasiids have been identified. It ap- 

 peared, however, that only a single species, 

 probably Thysanoessa sp. , was of any impor- 

 tance to the diet of the salmon. Five spe- 

 cies of amphipods were identified from the 

 stomachs. These are the gaimnarid, Scopelo - 

 cheirus coecus ; and the hyperiids, Hyperia 

 gaudichaudii, Parathemisto japoalca , P. 

 libellula , and Primno n. sp. Of these spe- 

 cies only P. Japonica was of any importance 

 and it probably accounted for over 90 per- 

 cent of the total amphipods. The pteropod 

 found was almost always Limwcina sp. , al- 

 though there were a few specimens of Euclio 

 sp . The fish found were mostly lantern 

 fishes, family Myctophidae, and juvenile 

 rockfishes, Sebastodes sp. None of the 

 squid were identified. The crustacean lar- 

 vae, primarily megalops and zooea stages of 

 decapods, also were unidentified. 



Patchiness of organisms and 

 selectivity of feeding 



Some of the stomachs contained a single 

 species but more typically, particularly for 

 the sockeye, a single species accounted for 

 more than 90 percent of the volume of the 

 diet. An example of a stomach with an al- 

 most pure content of copepods is shown in 

 figure k. The contents of a number of other 

 stomachs, rather than being a homogeneous 

 mixture, were definitely stratified. An 

 example of this for the sockeye salmon is 

 shown in figure 5- The contents of the pos- 

 terior region of this stomach were entirely 

 emphlpods, Parathemisto Japonica , while the 

 anterior portion was filled with euphausiids. 



Similarly stratified stomachs were found in 

 chum and pink salmon. 



The above resxilts can be best explain- 

 ed by assuming a non-homogeneous distribu- 

 tion of food organisms and some degree of 

 feeding selectivity by the fish. The strat- 

 ified stomach shown in figure 5> for example, 

 was taken from a fish capture in Area 12. 

 The dominant item in the diet of 17 fish 

 from this area weis the amphipod, P. japonica ; 

 the stomach of one fish contained only 

 euphausiids, while the stcanach of another 

 fish (shown in fig. 5) contained one -half 

 amphipods and one -half euphausiids. If the 

 amphipods ajad euphausiids were well mixed 

 in the area where the fish were feeding, it 

 would be necessary to assume a high degree 

 of feeding selectivity on the part of the 

 fish. If, however, those food organisms 

 had a patchy distribution, and the amphipod 

 patches were larger or more numerous than 

 the euphausiid patches, it is not necessary 

 to assume that the fish fed with extreme 

 selectivity. A fish found to have a stomach 

 containing only amphipods would have con- 

 fined its feeding to the amphipod patches; 

 a fish found to contain only euphausiids 

 had fed only in a euphausiid swaim; while 

 fish in which the stomach contents was 

 stratified had fed first in one sweuih and 

 then in another. 



Discussion of Results 



The basic assimiptlon in analyzing the 

 information reported in this study is that 

 fish taken in the sEuae area at the same time, 

 and in the same gear, had the same food 

 supply available. Differences in stomach 

 content would then reflect differences in 

 feeding behavior. This assumption may be 

 challenged on several points: 



1. The size of the sample may not 

 have been adequate for the three 

 species. 



2. The fish may not have been caught 

 in the area in which they had fed; 

 therefore, the various species 

 may have been exposed to differ- 

 ent food supplies prior to their 

 capture . 



3- The time of feeding of the three 

 species in the same area may have 

 been different. Diurnal plankton 



