a yacuum of 26 inches was reached. The material in the chamber was 

 agitated mechanically during the drying process and the steam pressure 

 controlled so that the temperature of the material did not exceed 150^. 

 The dr3ring time depended upon particle sixe, ccaisistency, and amount of 

 Bjaterial. Certain materials tended to stick to the blades of the agitator, 

 thus prolonging the drying time. One hundred pounds of raw material could 

 be processed in 6 to 12 hours. The dried scrap was grovind in a Wiley mill 

 using a 3/l6-inch hole size screen. 



2. Steam-cooked, air-tunnel dried meals: The raw material was 

 spread on trays made of galvanized welded wire fabric of one-half inch 

 mesh. The loaded trays were stacked in a basket with a conical aliminum 

 drain plate between each tray. The basket was placed in the vertical 

 pressure cooker, and the lish were processed at 15 pounds pressure for ten 

 minutes. The use of spaced trays made it possible to subject the material 

 to a uniform heat treatment, and the drain plates for each tray deflected 

 the freed liquor and condensed steam from lower trays, thus Biinlmizing the 

 leaching of water-soluble materials from the cooking fish. The cooked 

 material was spread on screens made from one-half inch mesh galvanized wire 

 fabric, using approximately 1-1/2 poimds of material per square foot of 

 surface. The screens were placed in a tunnel-type hot air drier. The 

 cooked material was dried at either 100° or 150^. for l-±/2 to 2 hours. 

 The dried scrap was ground in a Wiley mill using a 3/16 -inch hole size 

 screen. 



3. Acetone-extracted meals: The raw material was mixed with enough 

 acetone to precipitate the protein, approximately 4 volumes of acetone to 

 1 volume of raw material. After allowing the mixture to stand with occa- 

 sional stirring, the acetone layer, which contained the water, oil, and 

 possibly other acetcme-soluble materials, was ronoved by decantation. For 

 a second extraction, enough acetone was added to make a definite layer of 

 liquid on top of the solid phase. The acetone remaining in the meal after 

 decantation was allowed to evaporate at rocm temperature. 



4. Meal from spoiled pink salmon offal: Frozen, groimd Alaska pink 

 salmon offal was placed in a large wooden box and allowed to stand in an 

 unheated ro<»i. The experiment was conducted during the winter and the 

 tonperatures in the roc» were almost the same as outside, fluctuating 

 around iiO°F. The mass was stirred thoroughly at intervals in order to keep 

 the spoiling action uniform. After two weeks the spoiled offal was cooked 

 and dried in an air tunnel at 100°F. as described in the preparation of 

 steam cooked air-dried meals. 



5. Lyophilized meals: The raw material was finely ground and a small 

 amount was placed in a round bottom flask. The material was frozen in a 

 thin layer on the inside surface of the flask using dry lee and alcohol. 

 The sample was dried by exposing it to a high vacuum trtiich roioved the 

 moistiure. 



17 



