FISHERY BULLETIN: VOL. 71, NO. 2 



PS = 100 

 c 



0.5 S [a-b] =S min(G,6); 



where PS = percent similarity of community 

 samples at different stations, and a and b 

 are, for a given species (or group), the per- 

 centages of samples A and B which that species 

 (or group) represents. This method, as adapted 

 for food organisms, permits comparison of 

 diets within stations being studied. It quan- 

 titatively measures the relative similarity in 

 terms of species numerical composition, in this 

 case, occurrence in the stomach contents of 

 juvenile chinook. 



Percent similarities for all possible monthly 

 combinations of samples were compiled and 

 entered in a diamond matrix (Figure 5). Values 

 of combined samples for all upper and lower 

 stations ranged from a low 58.5 in May, 68.8 

 in April, 70.4 in June, to 75.5 in July (dark 

 areas). 



NO SAMPLES AT STATION C IN AMY 

 '^'nO samples at STATION B IN JULY 



Figure 5.- — Diamond matrix comparison of percent 

 similarities (fS^.) between and within upper (A, B, C) 

 and lower (D, E, F) stations based on major food 

 organisms consumed, April-July 1969. 



Monthly computed figures within and be- 

 tween individual stations ranged above and 

 below these values. No consistent pattern was 

 evident. An upper station (A, B, or C) some- 

 times showed a high similarity value when 

 compared with a lower station (D, E, or F) 

 and sometimes a low value. For example, 

 values for Station A versus F were high 

 in April (77.7), June (74.8), and July (78.3) 

 but low in May (49.2). The calculations were 



consistent with the conclusion that, despite 

 the general similarity in diet (PS^ above 50% 

 in most cases), proportional variations between 

 stations occurred randomly in response to site 

 habitat features and the feeding activity of 

 individual fish. 



Variability in Feeding Intensity 



Mean numbers of insects in the stomach 

 contents revealed an increasing trend from 

 March to July (Figure 6). This increase cor- 

 responded with the span when fish size and 

 water temperature were simultaneously in- 

 creasing. During March and April, when fish 



20 



MARCH: MEAN 39.4mni, TEMP. 3-6C 

 _J I I D I I L 



20 - 



20 



MAY: MEAN 44.1mm, TEMP. 9-13C 



U 



Ll 



I III 



JUNE: MEAN 49.1mm, TEMP. 13-16C 



R 



J U 1 U L 



1 



100 



JULY: MEAN 58.7mm, TEMP. 16-20C pi 



Jl 



Jl 



I 



D E 



STATION 



. Figure 6. — Number of insects in stomach contents of 

 juvenile chinook salmon at six primary stations, in 

 relation to mean fish lengths and water temperature. 



were small and temperatures were low (3°- 

 9°C), mean numbers of insects contained were 

 minimum, about 10 per fish. Mean numbers 

 increased to 38 in May and decreased to 28 

 in June. A peak of 47 was reached in July 

 when the fish were large and temperatures 

 ranged from 16° to 20°C. The May samples 

 were influenced by high utilization of Hypo- 

 gastruridae at Station A on 20 May, and the 

 July samples by adult midges at Station F 

 on 29 July, the last time fish were available. 

 If these samples were excluded, mean numbers 

 of insects contained in May and July did not 



394 



