control group do not compare to previous results at Leavenworth in which 

 losses ranged from 18 to 30 percent for a similar 12-week period (Bur- 

 rows, 1951) « The differences in mortality may be attributable to 

 differences in water temperature patterns throughout the period of the 

 experiments. The average water temperatures per biweekly period for 

 1950 at Leavenworth and for 191?1 at Entiat are as follows : 



1950: hS.k, U5.1, W1.8, hh'9, UU.5, U7.8, mean U5.U 



I95IJ U3.0, U3.1, h3.9, hh.9, U5.5, U6.8, mean UU.5 



the 1950 temperature pattern does not differ significantly from those of 

 19[|.8 and 19U9 at Leavenworth. In every instance the temperature either 

 dropped slightly or remained constant during the first 10 weeks of the 

 experimental period. The twelfth biweekly period occurring the last of 

 June was a transitional one in which the water temperature rose rapidly. 

 The 19^1 pattern at Entiat differs sharply from those of previous trials 

 in that the temperature was on a slight but consistent upward trend. It 

 is believed that the difference in temperature pattern and the resultant 

 increase in food intake is responsible for the marked reduction in mor- 

 tality during the 19^1 trials. Experiments, using controlled temperature 

 equipment, are planned to test this hypothesis. 



The results of the cold-water phase of the first experiment indi- 

 cated that either the beef liver - beef lung or hog liver - beef lung 

 combinations v/ere diets comparable in nutritional adequacy to the beef 

 liver and hog liver combination — the most satisfactory starting diet 

 developed to date. A hog liver plus salmon viscera and a hog liver - hog 

 spleen - salmon viscera combination proved comparable to the standard 

 meat-viscera control diet in both their grovrth potentials and survival 

 rates. These combinations offer excellent possibilities as cold-water 

 diets. Tuna liver in combination with salmon viscera was inferior to the 

 hog liver and salmon viscera combinations but proved an adequate substi- 

 tute for beelf liver in the meat-viscera mixture. Preserved salmon eggs 

 rfere inferior to frozen eggs but comparable to viscera in growth potential 

 when substituted for these products in variations of the meat-viscera 

 mixture. Neither the substitution of predigested meal for vacuum-dried 

 salmon viscera meal nor the supplementation of the meal with lysine and 

 methionine alleviated the mortality incurred from feeding salmon meals 

 during periods of cold water. The fortification of the meal with 

 vitamins of B-complex reduced the mortality but the results were incon- 

 clusive. 



First Experiment, 2l4.-Week Period 



Certain of the diets of the first experiment were continued an 

 additional 12 weeks making a total of 21; weeks for the experiment. Diets 

 which were of no further interest either because they were primarily 

 cold-water diets v/hich would be too costly to feed when the food intake 

 was high or had been previously tested in warm water were discontinued. 



