viscera meal, A similar ration, Diet 32, from which the APF supplement was 

 omitted, served as the control. 



The mean weight produced by Diet 31 which contained the APF con- 

 centrate (without aureomycin) was significantly less, than the CGmjarable diet, 

 Diet 30, which contained no APF concentrate. In contrast, tfie mean weight pro- 

 duced by Diet 33 which contained the APF concentrate (with aureomycin) was not 

 statistically different than the weight produced by the comparable diet, ^^iet 

 32, which contained no APF concentrate. 



The blood counts and hemoglobin determinations showed that Diets 

 31 and 32 had a possible anemic tendency. 



The conclusions to be drawn from the weights are definite. The 

 APF concentrate ( without aureomycin) reduced the growth rate, significantly^ 

 in Diet 31 and the APF concentrate (with aureomycin) had no effect on the 

 growth rate in Diet 33. The conclusions from the hemoglobin determinations 

 and the blood counts can not be regarded as nearly so definite. It can be 

 said that the addition of the APF concentrate (without aureomycin) to Diet 31 

 produced a tendency toward anemia, and that the APF concentrate (with aureomycin) 

 possibly averted a tendency toward anemia in Diet 33. The reason for the anemic 

 tendency in Diet 32 is not clear since previous trials indicated that Diet 32 

 produced no anemia. Some of the possible reasons for this tendency may have 

 been a difference in meals, a difference in the beef liver, hog spleen, and 

 canned salmon, or a difference in fish size. 



From these conclusions it is indicated that more experimental work 

 is needed before APF concentrates may be used to fortify diets for blueback 

 salmon. 



POSSIBLE PRODUCTION DIT-JTS DERIVED FROM THE SECOND EXPERIMENT 



For immediate practical use, these experiments indicated that 

 mackerel offal meal, hake, whole cod, haliBut sawdust, and tuna liver could 

 be substituted for various components of the standard meat-viscera-raeal mix- 

 ture during periods of warm water. 



Air-lift dried mackerel offal meal gave less growth than the vacuum- 

 dried salmon viscera meal but should give no less growth than the commercially 

 available flame-dried salmon offal meal. As was previously men- 

 tioned, the comparison of mackerel offal meal and flame-dried salmon offal was 

 made on the basis of previous tests. The diet testing mackerel offal meal 

 consisted of 20 percent each of beef liver, hog liver, and hog spleen, 30 

 percent salmon viscera, and 10 percent of air-lift-dried mackerel offal meal. 



Whole hake, whole cod, and halibut sawdust all proved to be satis- 

 factory substitutes for hog spleen in the meat-viscera-meal combination, but 

 of these fish products only halibut sawdust produced as much : growth as hog 

 spleen. It should be mentioned that hake in previous experiments proved satis- 

 factory in cold water feeding. The diets testing these products consisted of 

 either 20 percent hake, 20 percent cod, or 20 percent halibut sawdust added to 

 a mixture of 20 percent beef liver, 20 percent hog liver, 30 percent salmon 

 viscera, and 10 percent vacuum-dried salmon viscera meal. 



19 



