The meat and viscera mixture has proved superior to any single 

 or combination of raw products tested on blueback salmon. This diet, 

 which consists of 22.2 per cent, each, of beef liver, hog liver, and 

 hog spleen, and 33. Ij. per cent of salmon viscera, produces excellent 

 gains with a minimum of mortality during periods of low water tem- 

 peratures—less than 50 degrees o At water temperatures above 50 

 degrees any of the salmon waste meals may be incorporated into the 

 diet at the 10 per cent level to produce a significant growth res- 

 ponse without an increase in mortality „ Of the salmon waste meals 

 tested, low temperature (100 and lU5 degree) tunnel dried, salmon 

 viscera meals exhibited the greatest growth potentials. 



That hog liver in combination with salmon viscera may meet the 

 nutritional requirements of blueback salmon, at least during periods 

 of warm water temperatures, is indicated from these diet experiments. 

 In the 19U5 feeding trials, a combination of 30 per cent bog liver, 

 20 per cent hog spleen, u0 per cent salmon viscera, and 10 per cent 

 flame-dried, salmon offal meal produced gains comparable to the 

 standard meat-viscera-me^l combination (Table 3> Diets 8 and 9) « 

 At the conclusion of the 1^-week experimental period no deficiency 

 symptoms were discernible. During the 2u-week trials in 19U8 an 

 alteration of this diet, which included u0 per cent, each, of hog 

 liver and salmon viscera plus 20 per cent of lu£ degree, tunnel- 

 dried, salmon viscera meal, was acutely anemic at the conclusion 

 of the experiment but the total gain of the fish was comparable to 

 those fed a meat-vise era-meal mixture (Table 5, Diets 2 and 6). 

 The 20 per cent level of meal in Diet 6 was used to improve the 

 feeding consistency and it may be that this high level of meal was 

 responsible for the production of the anemic condition, further 

 experimentation using the hog liver-salmon viscera combination as a 

 substitute for beef liver appears to be justified. 



The recommended Cortland diets, using either 15 per cent beef 

 liver and 35 per cent hog spleen or 50 per cent hog spleen in com- 

 bination with 00 per cent of the Cortland No. 6 meal mixture, pro- 

 duced gains comparable to a meat-viscera-meal mixture when fed foir 

 a 12-week period at an average water temperature of 5>1| degrees 

 (Table h. Diets 2, 3, and U). At the low water temperatures experienced 

 during the first 12 weeks of the I9I4.8 feeding trials, the Portland 

 diets were unsatisfactory for blueback salmon. Although the addition 

 of 15 per cent beef liver resulted in a reduction in mortality and 

 an increased growth rate when compared with the 50 per cent spleen 

 ration, both diets showed significantly higher mortalities and lower 

 gains than did the meat-viscera-meal mixture (Table 5, Diets 2, U, 

 and 5). The deleterious effects of the Cortland No» 6 meal mixture 

 carried through into the warm water period. Although the growth 

 rate of the fish accelerated, it was still significantly below that 

 of the comparable meat-viscera-meal ration at the conclusion of the 

 experiment. From these results it is indicated that the Cortland No. 6 



31 



