FISHERY BULLETIN: VOL. 74, NO. 4 



length changes occurred during preservation. The 

 number of individuals for each species in each size 

 interval was recorded for every sample. 



The remaining 90% of a bottom sample-that 

 not sorted under magnification-was dried at 70°C 

 for 4 days and then ashed at 600° C. Ten percent of 

 its organic weight was arbitrarily assumed to have 

 been lost during preservation. Subtracting the 

 estimated biomass of small (<4 mm) invertebrates 

 within this subsample from the total loss of igni- 

 tion yielded the ash-free dry weight of filamentous 

 algae, some diatoms, detritus, and organisms too 

 small to be seen during the sorting process. Con- 

 version to energy units (kcal) was accomplished by 

 multiplying the plant-detritus biomass by 4.05, the 

 mean of five samples combusted in an oxygen 

 bomb calorimeter. 



Computations of invertebrate biomasses were 

 based on live specimens collected from a nearby 

 stream, grouped according to size and species, and 

 weighed after drying 4 days at 70° C. Their aver- 

 age dry weights were converted to calories by 

 values obtained from Cummins and Wuycheck 

 (1971) or determined directly by calorimetry. 

 When no representatives of a certain size were 

 available, a value for that interval was estimated 

 by interpolation. Very similar forms were as- 

 sumed to have identical values. For bottom sam- 

 ples, the biomass (kcal/m-) of each size class of 

 each taxon was taken as the product of the number 

 of individuals in that class, the estimated caloric 

 value for individuals of that size, and the appro- 

 priate area conversion factor. The product of the 

 number of individuals and the caloric value was 

 divided by the total amount of water passing 

 through the nets to give biomass estimates per 

 unit volume (cal/m^) for the drift samples. Sum- 

 ming the values of all size intervals gave the total 

 caloric content for each taxon. 



Fish 



Fertilized chinook salmon eggs were obtained 

 from the Marion Forks Salmon Hatchery of the 

 Oregon Department of Fish and Wildlife. Eggs for 

 the 1972 experiment, taken 3 October 1971, were 

 from a single pair mating. Eggs used in the 1973 

 experiment, taken 1 October 1972, were obtained 

 by crossing three females with four males. This 

 was done in order to increase genetic heteroge- 

 neity among fish in the 1973 experiment. Follow- 

 ing fertilization, the eggs were transported im- 

 mediately to holding facilities where they were 

 incubated at 12°C. 



In 1971, eggs were introduced into the streams 

 when they reached the eyed stage. They were 

 hatched in floating baskets and the fry were 

 released shortly before yolk absorption was com- 

 pleted. Owing to accelerated development in 

 warmer water, fish in the heated stream were 

 released sooner than those in the control, although 

 the initial number of individuals placed in the two 

 streams was identical (425). A 10-wk recoloniza- 

 tion period following repairs delayed introduction 

 of salmon until mid-March 1973, when 200 fry were 

 released simultaneously into each stream. 



When the fish had reached approximately 0.4 g 

 wet weight, they were all removed from the 

 streams for measurement of individual length and 

 weight every 3 wk until an experiment was ter- 

 minated. From 5 to 20 fish were randomly drawn 

 from the populations for stomach analyses. A 

 blunted 22-gauge needle on a 5-ml syringe was 

 inserted through the esophagus of an anesthetized 

 fish into the anterior limb of the stomach. Several 

 milliliters of water were gently injected into the 

 stomach, forcing the contents out through the 

 mouth into a collecting beaker. The combined 

 whole food organisms and identified fragments of 

 each taxon were weighed to the nearest 0.1 mg, 

 and each taxon was assigned a percentage of the 

 diet based on its fraction of the total wet weight of 

 the sample. 



Direct effects of the model stream temperature 

 regimes on chinook salmon growth rates at 

 different levels of food availability were studied in 

 concurrent experiments. Fish of the same paren- 

 tage and size as those in the model streams were 

 placed in insulated streamside troughs, where 

 they were fed live Tubifex at rations ranging from 

 near maintenance to near repletion. The troughs 

 received water directly from^ the model streams, 

 and temperature differences between the troughs 

 and streams were never greater than 0.3°C. Ten- 

 day growth experiments were carried out once 

 each season during 1973. Each experiment was 

 preceded by a 10-day period of acclimation to 

 temperature and ration size. Numbers of in- 

 dividuals tested at each ration level ranged from 

 10 to 20 depending upon fish size. 



Average relative growth rates (Warren 1971) of 

 the salmon were calculated as: 



ARG = 



W, - n\ 



0.5 (W, + \V.,)'t 

 where ARG represented growth, Wi and W2 



766 



