the fish was plotted for each 4-h interval. Dry 

 weight estimates for each fish were based on the 

 entire carcass. These estimates were derived ex- 

 perimentally using a dry weight to wet weight 

 conversion (21.1%) based on the combined results 

 of 15 individuals of each species. 



Daily meal and daily ration of subyearling coho 

 salmon and steelhead were determined by sum- 

 ming interval values for the 24-h period. The 

 amount of food consumed at each 4-h interval, was 

 estimated from the Elliott and Persson (1978) 

 formula: 



Ct = 



(St -Soe-^')Rt 



1 



-Rt 



whereC; = amount of food consumed in t hours 



St = mean stomach contents at the end of 



the interval 

 So = mean stomach contents at the begin- 

 ning of the interval 

 R = exponential rate of gastric evacua- 

 tion in t hours. 



Estimates of R were derived using available 

 information for subyearling rainbow trout (Win- 

 dell et al. 1976). 



Variability in the diet of coho salmon and 

 steelhead over the 24-h period, and diet similarity 

 between coho salmon and steelhead between sam- 

 pling periods spaced 4 h apart were examined 

 using the overlap formula of Morisita (1959) 

 modified by Horn (1966): 



c.= 



2 IX,- Y, 

 / =1 



~ 2 ^ 2 



1 Xi" + 1 Y, 



i = l 



( =1 



where C\ = overlap coefficient 

 s = food categories 

 Xi = proportion of the total diet of fish 



species X contributed by food 

 category i 

 Y, = proportion of the total diet of fish 

 species Y contributed by food 

 category i. 



Ck can range from 0, when samples contain no food 

 in common, to 1, when there is identical represen- 

 tation of food between samples. 



Results 



A total of 130 coho salmon and 142 steelhead 

 were examined for diet composition and feeding 

 periodicity (Table 1). Subyearling coho salmon 

 which emerge approximately 6 wk earlier than 

 subyearling steelhead were substantially larger 

 at the time of collection. Terrestrial invertebrates 

 were the major components (51.3-62.3%) in the 

 diurnal diet of coho salmon fry, whereas aquatic 

 invertebrates were the major prey utilized at 

 night (89.5-96. 99f ) (Table 2). Predation on terres- 

 trial invertebrates was greatest between 1600 and 

 2000 h and greatest on benthic invertebrates 

 between 2000 and 2400 h. Of the aquatic taxa 

 immature chironomids were the most important 

 item ( 17.1% ) during the 24-h period although some 

 diel variation in importance is suggested. Other 

 less important aquatic items include adult chi- 

 ronomids (7.7%), caenid (7.0%), and leptophlebiid 

 (6.8%) mayfly nymphs along with hydropsychid 

 caddisfly larvae (6.0%). Homopterans (6.5%), dip- 

 terans (5.4%), and coleopterans (4.8%) were the 

 major terrestrial invertebrates consumed. Over- 

 all, aquatic invertebrates composed 74.7% and 

 terrestrial invertebrates 25.3% of the diel diet of 

 coho salmon (Table 2). 



Aquatic invertebrates were the principal food of 

 subyearling steelhead throughout the 24-h period 

 (Table 2). Although steelhead did not utilize 

 terrestrial invertebrates to the extent that coho 

 salmon did during diurnal feeding, temporal pat- 

 terns in the consumption of terrestrial prey were 



Table L — Number examined, mean total length (millimeters ± 957^ confidence limits), mean dry weight 

 offish (grami and stomach contents (milligrams), andC\ of subyearling coho salmon and steelhead. Data 

 are from sampling intervals spaced 4 h apart, 13-14 July 1979, in Orwell Brook, NY. 



372 



