Discussion of Results 



The experimaits conducted were not sufficiently numerous oc any one 

 of the variations mentioned to warrant conclusive statements regarding 

 the potential supply of vitamin A oils that could be obtained from Alaska 

 salmon cannery waste. In Tables 1 through 5 there are presented the 

 pertinent vitamin A data on the oils prepared and tested from pink, chum, 

 echo, king, and red salmon waste respectively. Our aim in this preliminary 

 project was primarily the evaluation of the alkali digestion process. In 

 view of the satisfactory manner in which the alkali digestions progressed, 

 it may be stated that the process can be utilized to recover the oils from 

 the waste. 



Salmon heads were digested separately to give a reference point for 

 subsequent combinations of parts of the waste. There was no difficulty in 

 the digestion of the heads whether they were processed whole or chopped. 

 The time required for completion of processing was scmewhat longer in the 

 former instance. Qnulsions were more frequently encountered in the prep- 

 aration of the head oils than in the viscera oils, but in all such cases 

 re-centrifuging of the emulsion resulted in a satisfactory oil. Probably 

 further refinements in minor details of the procedure would minimize or 

 eliminate this difficulty with emulsion formation. Since the vitamin A 

 content of the head oils was uniformly very low, ranging from 175 U. S. P. 

 units per gram of oil for chum salmon to 540 vinits for coho salmon, no 

 further experiments seem to be indicated if vitamin A oils are the pri- 

 mary interest. The E value ratios at 300/328 millimicrons ranged from 

 0.891 to 1.577. In the vitamin A industry the cu&tomary maximum acceptable 

 ratio at 300/328 millimicrons is 0.72. Higher ratios usually mean that 

 there are substances present iirfiich increase the apparent vitamin A content 

 so that the true biological vitamin A value of the head oils may be even 

 less than the small potencies given. 



Observations made during the digestions using the total visceral 

 portion of the waste differed somewhat from those made when other portions 

 were processed. When the viscera was added to the agitated mixture of water 

 and sodium hydroxide (3.0 percent) a very viscous mass resulted. The material 

 could be picked up on a paddle and stretched into a fine sheet resembling 

 cellophane. This condition continued for the first 30 to 45 minutes of 

 the digestion, then as stirring and heating continued, the liquor became 

 gradually more fluid, and after a total period of 1^ to 2 hours the di- 

 gestion went to completion. These tests were made using both cold water 

 and hot (190°) water at the beginning of the process. No differences 

 were apparent with respect to the viscous stage or to the digestion as 

 a whole. 



The oil yield from the viscera ranged frcan 1.2 percent for chum salmon 

 to 4.2 percent from red salmon. Not too much significance should be 

 attached to oil yields in small scale tests because the mechanical losses 

 could be high and variable from one lot to another, but these results do 

 give some conception of the approrlmate oil yield that could be expected 

 from the material by alkali digesticai. The vitamin A content, in U.S. P. 

 units per gram of oil, varied frcni 2,844 units for pink salmon to 66,820 



41 



