The above material was used for the follovdng preparation. The 

 frozen material was thawed, spread on hardware cloth screens, and air- 

 dried at 100°F, until it contained approximately 10 percent moisture. 

 The dried meal was ground, sampled for analysis, packed in 5-gallon tins, 

 and placed in the cold storage at 0°F. until used for the feeding tests. 



Frozen, ground Alaska pink salmon offal was thawed and dumped into 

 a large wooden box and allowed to spoil. At inteinrals the mass was stir- 

 red thoroughly to keep the action uniform throughout. Since the experi- 

 ment was ccxxiucted over a two-week period during the winter season, the 

 decomposition by enzymatic and bacterial action was not as rapid or com- 

 plete as had been desired. After two weeks the spoiled material was 

 cocked and dried as described previously; drying was done at 100°F. 



Fish solubles concentrate was prepared from the liquors drained from 

 the retort during the cooking process for the Alaska pink salmon viscera. 

 The liquors, consisting of the condensed steam, dissolved protein and 

 water-soluble vitamins CB vitamin complex) and any oil or suspended pro- 

 tein solids, were placed in shallow stainless steel pans and acidified 

 to pH 1.5 to 2 with concentrated hydi\3chloric acid. These pars were plac- 

 ed on shelves in a Stokes vacuum-drier oven. The liquors were concentrat- 

 ed by the simultaneous application of live steam to the Jacketed even and 

 shelves and the vacuum evacuation of the oven. The maximum ten^jerature 

 attained by the concentrate was 135°F. when the gauge reading on the vac- 

 uum line was 29. Concentration was continued until the material contained 

 approximately 50 percent solids. The hot concentrate was placed in 1/2- 

 pound size flat cans, hermetically sealed and stored until ready for use. 

 Prior to the feeding tests, the various lots of concentrate were warmed 

 in the unopened cans, the cans were then opened, and a homogeneous mass 

 made of the lots. Samples were taken for analysis. The balsmce of the 

 concentrate was placed in small vials of 10 milliliters capacity to facili- 

 tate addition of small amounts to the diets as required without undue ex- 

 posure of the unused portion to the action of air. The vials were stored 

 at room temperature during the feeding test period. 



Methods of Vitamin Assay 



The general methods used for the riboflavin and niacin assays are 

 those set up by the Association of Vitamin Chemists (1947) and Roberts 

 and Snell (19^)^). The samples of raw material were composite samples 

 taken v^en the meal was prepared. They were macerated in a Waring blend- 

 or and kept frozen. The ground meal samples were stored at 35°F. The 

 hog liver and spleen and the beef liver are samples from material being 

 used in the diets at the Leavenworth Hatchery. Approximately 10 pounds 

 of each were ground in a meat grinder while still frozen, thoroughly mix- 

 ed, sealed in cans and stored at 0°F. 



81 



