Analyses of the vitamin A content of the standard meat-viscera-meal 

 control (Diet 23), whale liver (Diet 20), and the tuna liver - salmon 

 viscera meal diet (Diet 31) indicate the possibility of a hypervitamino- 

 sis A being responsible for the high mortality encountered in the whale 

 liver and tuna liver diets. Results of the analyses made by the Seattle 

 Technological Laboratory of the Fish and Wildlife Service are as follows; 



Spec trophotome trie units 

 Diet No. Oil Content per gram of oil per 100 grs . diet 



20 (whale liver) 3-37^ 9U,200 317, U5U 



31 (tuna liver) 2,9h% 19,965 -58,697 



23 (control) 6.15/^ 1;,)400 27,060 



It will be noted that the vitamin A content of Diets 20 and 31 was much 

 higher than that encountered in Diet 23 the standard control. A high 

 mortality, sufficient to force a discontinuation of the experiments, 

 occurred in Diet 20 after U weeks of feeding and Diet 31 after 8 weeks 

 of feeding. Hyper vitamin os is A has not been reported previously in 

 salmon but has been produced in both man and rats . The work of Rodahl 

 as reported by Marier (1951) i£ particularly pertinent in that he was 

 able to demonstrate hypervitarainosis A by feeding whale-liver oil to 

 rats. All the evidence strongly supports the hypothesis that a hyper- 

 vitaminosis A was present in the fish which vrere fed diets containing 

 high levels of either whale or tuna liver. 



The results of the second experiment indicated that whale liver had 

 few possibilities as a diet component for blueback salmon because of its 

 high vitamin A content. Both herring and arrovf-toothed halibut when 

 substituted for hog spleen in the meat-viscera-meal combination produced 

 a significant increase in the growth rate of the fish but the diet con- 

 sistency was poor. Of the tvfo, herring appeared to have a higher growth 

 potential than the halibut. Low-temperature-dried whale meal was an 

 adequate substitute for vacuum-dried salmon viscera meal. Beef lung 

 substituted for hog spleen in the meat-viscera-meal mixture produced a 

 combination equal to the standard control. The addition of hog liver, 

 beef lung, or a combination of hog liver and beef lung to salmon viscera 

 and salmon viscera meal produced combination diets equal to or better 

 than the standard meat-viscera-meal control. Tuna liver in a similar 

 combination was unsatisfactory due to the development of a hypervitaraino- 

 sis A. 



Third Experiment 



Although no controlled-feeding trials had been conducted previously 

 on Chinook salmon at the Salmon-Cultural Laboratory, in regular produc- 

 tion feeding a difference in nutritional requirements appeared to exist 

 between Chinook and blueback salmon. Chinook, when fed the standard 

 meat-viscera mixture, developed an acute anemia in at least a portion 

 of each lot fed. This condition did not develop in. b-luebadc fed an 



17 



