Juvenile coho salmon, Oncorhynchus kisutch (Walbaum), were held at 20° C. in 12-gallon bottles, usually 10 to a bottle, in continuously renewed fresh water having various dissolved oxygen concentrations. Reduced oxygen concentrations were maintained by bubbling nitrogen through the inflowing water. In most of the experiments, the fish were fed beach hoppers (marine amphipods) to repletion twice daily so that food was available for 10 hours each day. Growth and food consumption rates of age-class 0 juveniles observed in the year 1956 declined slightly with reduction of oxygen concentration from a mean of about 8.3 to 6 and 5 mg/l, and declined more sharply with further reduction of oxygen concentration. Weight gains in grams per gram of food consumed (food conversion ratios) were slightly depressed at concentrations near 4 mg/l, and were markedly reduced at lower concentrations. At concentrations averaging 2.1 to 2.3 mg/l, many fish died and surviving fish consumed very little food and lost weight, although some food was assimilated. Interesting seasonal variations were noted of the dissolved oxygen requirements of coho salmon, their food and oxygen consumption rates, and their gross food conversion efficiencies. The conversion efficiencies apparently increased, whereas food consumption rates declined in the fall of 1956. The seasonal variations are not ascribable entirely to size differences. Very poor growth and low food consumption rates at reduced oxygen concentrations near 4 and 5 mg/l and unexpected mortalities at lower concentrations near 3 mg/l observed in the summer of 1955 probably were due to synergistic action of a toxicant leached from rubber (latex) tubing supplying water to the test vessels. The stated results of 1956 tests were not likewise invalidated.