from Lot 35 was added to viscera less gonads for alkali digestion, re- 

 sulted in an oil containing 8,060,000 units per pound of oil. In this 

 instance the head oil contributed at least 1.9 percent of the vitamin A 

 in the final oil. On the basis of the previous experiments, some oil 

 would be expected from the viscera itself, but in this experiment there 

 was no additional recovery of oil, probably due to saponification and/or 

 mechanical losses in processing. 



A similar series of expa-iments wei*e conducted using pink salmon 

 waste. The oil recovered from heads and viscera less gonads contained an 

 average of 1,207,500 units per pound of oil. The oil from the digestion 

 of the heads alone contained 116,700 U. S. P. units of vitamin A per pound 

 of oil. When this head oil was added to the viscera less gonads the oil 

 (Lot 12) recovered contained 985»180 units of vitamin A per pound, an 

 apparait net loss compared to Lots 7 and 8. Here again the heads contributed 

 only 1.2 percent of the total vitamin A obtained. 



In Lots 5 and 6 pink salmon heads were combined with pink salmon livers 

 for digestion. The results were at considerable variance. For Lot 5 the 

 heads contributed 116,700 units per pound of oil, or i»2 percent of the total 

 278,000 units per pound of oil recovered. The heads in Lot 6 supplied 13 

 percent of the total vitamin A recovered, assuming their vitamin A con- 

 tent was the same as that found for pink salmon heads in Lot 1. 



The oils shown as Lots 13,23,28 were composites of all the oils pre- 

 pared from the waste of pink, chum, and red salmon respectively. All oils 

 from king and coho salmon waste were pooled to make up Lot 28. The E value 

 ratios indicate that Lot 13, pink salmon oil, contained a disproportionately 

 high amount of head oils although the vitamin A content is approximately 

 representative of the combined lots. Tl>e chief purpose in saving these ccm- 

 posites was to get an evaluation of the salmon oils in animal and poultry 

 feeding tests. Thus any gross discrepancies in the vitamin A concentration 

 as measured by physico-chemical methods compared to the actual biological 

 vitamin A concentration may be brought out. 



Summary 



The alkali digestion process was found to be adaptable for the pre- 

 paration of vitamin A bearing oils from total salmon cannery waste. 



Several variations were made in the type of raw material selected from 

 the total cannery waste to observe the effect cf the presence or absence 

 of specific parts of the waste on the digestion process and on the concen- 

 tration of vitamin A in the oil produced therefrom. If the processor 

 wishes to recover the vitamin A in an oil with the highest possible con- 

 centration, the best portion of the cannery waste to utilize is the viscera. 

 Some increase in the facilitation of the digestion may be made by the re- 

 moval of the gonads from the viscera. 



43 



