to 



Z 

 UJ 



17.5h 



o> 



15.0 - 



Ox 



io 



O ^ 10.0 



CO X 

 u. o 

 O UJ 



.5 - 



n 



oc E 

 Q ^ 



z 

 < 



UJ 



7.5 • 



5.0 



2.5 



SUBYEARLING 

 STEELHEAD 



' SUBYEARLING 

 COHO SALMON 



0800 



1200 



1600 2000 



HOUR 



2400 



0400 



Figure l. — Relative food contents (with 95% confidence limits) 

 of subyearling coho salmon and steelhead at 4-h intervals, 13-14 

 July 1979, in Orwell Brook, NY. 



food consumed among individual fish occurred 

 during peak feeding periods in both species. Coho 

 salmon appear to be primarily nocturnal feeders 

 since their stomachs during periods of low light 

 intensity (2000-0800 h) were much fuller 

 (x = 6.87 mg stomach contents /g fish) than during 

 daylight periods (0800-2000 h) {x - 3.23 mg 

 stomach contents /g fish). Conversely, steelhead 

 are predominately diurnal feeders. Their stom- 

 achs during the day {x = 9.03 mg stomach 

 contents /g fish) were considerably fuller than at 

 night ix = 5.60 mg stomach contents /g fish). 



We estimated the daily meal (amount of food 

 consumed per day) and the daily ration (amount of 

 food consumed per day expressed as a percentage 

 of the fish's body weight) for subyearling coho 

 salmon and steelhead in Orwell Brook. The daily 

 meal of coho salmon was estimated as 10.6 mg and 

 the daily ration was 1.7% of their dry body weight. 

 For steelhead the daily meal was estimated as 7.8 

 mg and the daily ration as 2.8%. 



To investigate the hypothesis that the diet of 

 coho salmon fry was more variable than that of 

 steelhead fry over a 24-h period the overlap 

 formula was used to compare diet similarity for 

 each species between each 4-h sampling interval. 

 When employing this formula, Cx^0.60 is as- 

 sumed significant (Zaret and Rand 1971). Using 

 this criterion, all possible combinations for steel- 

 head are significant (Table 3). Only 4 of the 15 

 comparisons for coho salmon are significant while 



Table 3. — Diet similarity values (C\) for subyearling coho 

 salmon and steelhead between sampling periods spaced 4 h 

 apart, 0800 h 13 July to 0400 h 14 July 1979. 



their mean Ck, 0.43, is much lower than that of 

 steelhead, 0.77 (Table 4). 



The diet of coho salmon and steelhead during 

 similar 4-h intervals differed. The greatest over- 

 lap occurred between 2400 and 0400 h and the 

 lowest from 1200 to 1600 h (Table 1). Overall, the 

 mean diel overlap in diet between coho salmon 

 and steelhead was significant {Ck = 0.67). 



Discussion 



The results of this study support the hypothesis 

 that the diet composition of subyearling coho 

 salmon is more variable over a 24-h period than 

 that of sympatric subyearling steelhead. The C\'s 

 for each species between different 4-h intervals 

 (Table 4) indicate that the diet composition of coho 

 salmon was variable, whereas that of steelhead 

 was fairly uniform over the 24-h study period. In 

 fact, the data show that at no time did the diet of 

 steelhead fry (mainly aquatic invertebrates) differ 

 substantially from diets at other 4-h intervals 

 (e.g., all C\'s were significant). The low similarity 

 in the diel diet of coho salmon, as postulated, was 

 due to the predominance of terrestrial prey in 

 their daytime diet and predominance of aquatic 

 prey in their nocturnal diet as C\s between day- 

 time (0800-2000 h) and nocturnal (2000-0800 h) 

 periods for coho salmon were generally the lowest 

 ix = 0.32). Unfortunately, samples of the inverte- 

 brate drift were not taken at 4-h intervals con- 

 current with these fish collections. We suspect, 

 however, that the variability in the diet of coho 

 salmon was due to diel fluctuations in the com- 

 position of the drift, which are well documented in 

 streams (Hynes 1970). In many streams during 

 the summer, terrestrial invertebrates dominate 

 the diurnal drift (Hinckley and Kennedy 1972; 

 Johnson and Ringler 1980) while aquatic inverte- 

 brates dominate the nocturnal drift (Hynes 1970; 

 Hinckley and Kennedy 1972). Since the diet of 

 coho salmon fry in Orwell Brook has been shown 

 to be significantly associated with the composition 



374 



