FISHERY BULLETIN: VOL. 70, NO. 2 



:3 







</> 



I 



ft 



ft 



ro- 



se 



5.0- 



4.0 



200 400 600 800 1000 



Mean number of organlsnns per fry foregut 



Figure 3. — Relationship between the mean number of 

 organisms per foregut in fry and yearlings. 



asymptotic upper limit with respect to food 

 abundance in the Lake. The relationship might 

 have been nonlinear for both fry and yearlings 

 if lower and higher zooplankton densities had 

 occurred. 



The relationship between the number of or- 

 ganisms per foregut and fish size within each age 

 group was not examined because size range of 

 fry and yearlings in any sampling area was 

 small. However, the yearlings contained an 

 average of 40 Yr more organisms than the fry 

 in the late summer of 1967, probably because of 

 their larger size and feeding capacity. The feed- 

 ing intensities of both age groups are compared 

 in Figure 3. The correlation coefficient, r = 

 0.67, between the mean number of organisms per 

 fry foregut and the natural logarithm of the 

 mean number of organisms per yearling foregut, 

 significant at P = 0.05, indicates an exponential 

 relationship between the feeding intensity of the 



two age groups. This was expected as fry tend 

 to approach an upper limit of food intake (Fig- 

 ure 2). 



COMPOSITION OF THE DIET AND 

 THE ZOOPLANKTON SAMPLES 



The food of juvenile sockeye salmon in the 

 limnetic area of Iliamna Lake consisted primar- 

 ily of zooplankton. Insects averaged less than 

 1% of the total number of organisms in the fore- 

 gut except for yearlings from area I during the 

 early summer of 1967, when they constituted 

 42.6 % . Cyclops and Bosmina were usually dom- 

 inant in the zooplankton hauls and in the fish 

 foreguts, and averaged 75% in the zooplankton 

 hauls and 90% of zooplankton in the fry and 

 yearling foreguts (Table 3). Yearlings con- 

 tained a higher percentage of Cyclops and a low- 

 er percentage of Bosmina than fry in late sum- 

 mer 1967 when both age groups were sampled. 

 The percentage of calanoid copepods averaged 

 higher in the zooplankton hauls (13%) than in 

 the fish foreguts (3%) and was consistently 

 higher for all areas and sampling period. Daph- 

 nia and Holopedmm occurred in most samples 

 but only constituted 11% of the organisms in the 

 zooplankton hauls and 7% in the fish foreguts. 



To compare the percentage composition of or- 

 ganisms in the zooplankton hauls with that of 

 the foregut contents, a "two-way crossed" anal- 

 ysis of variance (fish-zooplankton by area) was 

 performed for each sampling period and for each 

 organism {Cyclops, Bosmhia, and calanoid cope- 

 pods). Daphnia and Holopedium were not 

 tested. Mean squares and degrees of freedom 

 are shown in Table 4. The percentage compo- 

 sition of the foregut contents and the zooplank- 

 ton hauls (Table 3) were significantly different 

 except for Cyclops in late summer 1966 and early 

 summer 1967. However, differences in the zoo- 

 plankton hauls and the foregut contents were 

 not consistent in all areas for Cyclops and Bos- 

 mina as indicated by the significant interaction 

 between the fish-zooplankton samples and areas. 

 Interaction was not significant for the calanoid 

 copepods. 



A modification of Tukey's test (Snedecor, 

 1956, p. 251) showed a significant difference 



358 



