water temperature of Ul4.5° F. The response of blueback salmon to 

 different' diets varies vath marked chances in water tem.peratures. Cer- 

 tain of these 'phenomena have been described by Burrov/s et al. (195^1) 

 and confirmed by Robinson et al. (l?5la). By definition average water 

 temperatures below U^° F. are considered cold water and average water 

 temperatures above $0° F. are regarded as warm water. Because of the 

 variation in response to diets at different temperatures the design of __ 

 the experiments vias such as to test diet components at either cold 

 water, warm water, or a combination of both phases of temperature. 



A summary of the cold-water phase of the first experiment is given 

 in Table 2. The diets of this part may be divided into three main 

 groups. In the first group (Diets 1 through 7 and Diet 12) evaluations 

 were made of individual components and limited combinations, usually 

 two, of the components. In the second group (Diets 8 through 11) 

 variations in the standard meat-viscera mixture were tested. In the 

 third group (Diets 13 through 18) an attempt was made to determine the 

 cause for the high mortalities encountered when feeding salmon meals 

 during periods of cold water. 



Eight diets composed the first group. Beef liver (Diet 1) was 

 included as the standard control. Beef lung was fed at the 100 percent 

 level (Diet 2) and in combination with equal parts of beef liver 

 (Diet 1|) and hog liver (Diet S) . Equal parts of hog liver and beef 

 liver in combination (Diet 3) were included for comparative purposes. 

 Salm.on viscera -in combination with hog liver (Diet 6), hog liver and 

 hog spleen (Diet 7), and tuna liver (Diet 12) completed the diets in 

 this group, 



B.eef lung was the only component in these trials v/hich had not 

 been tested previously by this laboratory. McCay _et al. (1931) found 

 beef lung comparable to beef liver and kidney when fed at the 25 per- 

 cent level in combination v/ith dry feeds. He concluded that beef lung 

 contained the anti-anemic factor. Titcomb et al. (1929) could demon- 

 strate no difference in growth rates between trout fed diets consisting 

 of 100 percent each of either beef spleen or hog liver or beef lung. 

 Wales and Moore (1939) came to the same conclusions on diets consisting 

 of either beef liver, beef heart, or beef lung. Lord (1935) found that 

 lungs and spleen gave inferior growth in comparison to beef, hog, or 

 sheep liver. 



Previous experiments at this laboratory (Burrows _et al, 1951) 

 indicat^ed that a combination of 50 percent each of beef liver and hog 

 liver was superior to either beef or hog liver fed separately or in 

 combination #ith 50 percent of hog spleen. These results were con- 

 firmed in the present trials in which the beef liver - hog liver combi- 

 nation (Diet 3) produced a significantly greater gain in weight than 

 did beef liver (Diet 1). 



Beef lung at the 100 percent level (Diet 2) produced gains in 

 weight superior to those of beef liver (Diet 1) and comparable to the 



k 



