6. Flame-dried salmon offal meals: These samples were prepared 

 coBjaercially frcm *4iole cannerj waste. The meal was dried by use of a 

 gas, direct flaime drier. 



7. Blue crab ( Callinectes sapidiis ) meal: The meal was prepared 

 ccmmerciallj in a rotary flame drier fr«m the *4iole scrap of blue ciab. 



8. Dungeness crab ( Cancer magister) meal: The commercial meal was 

 prepared by dry-rendering Dungeness crab scrap at atmospheric pressure. 

 The laboratory meal was prepared at the Seattle Technological laboratory 

 by vacuum drying the scrap in a Stokes steam-jacketed drier at a tempera- 

 ture of 100° to 145°F. 



9. ifeickerel ( PneumatoiAorus dlego ) offal meal: The mackerel meal 

 was prepared in a c<aimercial plant by a wet-rendering method. The press 

 cake was dried in an air-lift drier. 



Discussion of uesults 



Results of the feeding tests at Leavenworth aire discussed in detail 

 in several reports (Burrows suad Karrick 1947; Burrows, Robinson, and Palmer 

 1951; Robinson, Palmer, and Burrows 1951; Robinson, Payne, Palmer, and 

 Burrows 1951), and are svunmarixed in Part IV of this report. Results of the 

 vitamin smd proximate analyses are given in tables II-l, 2, 3, and 4. 



Although quality of protein affects the nutritive value of the diets, 

 the results reported here are for total protein only and give no indication 

 of any alteration that may have occurred to the protein during processing. 

 Limited time and personnel precluded a study of protein quality. 



Natural materials vary in composition. Therefore, it must be empha- 

 sized that the analyses reported are on samples which are representative 

 of ccnuparatively small amoxints of material and are not representative of 

 the raw material as a vdiole. For example, beef liver undoubtedly varies 

 to a considerable extent frcm one portion to another of a lot as well as 

 frcn one lot to another. The analyses of the slaughterhouse waste used in 

 the diets were core samples taken frca a single block of beef liver furn- 

 ished by the hatchery and thus were representative only of that particular 

 block of material. 



One hundred percent beef liver is used as the standard diet for hatchery 

 feeding tests. For this reason the composition of both mixed diets and the 

 raw materials being studied as potential substitutes are compared with beef 

 liver. 



The composition of the samples of hog and beef liver examined was 

 essentially the same. Hog spleen contained more fat, less protein, and had a 

 much lower vitamin content than tne liver products. The salmon viscera 

 samples had a lower vitamin content than did tne beef liver. 



There were no significant differences in the composition of chinook 

 salmon ( Oncorhynchus tshawytscha) viscera frrai the Columbia River and pink 



18 



