136 U. S. BUREAU OF FISHEEIES 



the bureau so that the resultin<r product was comparatively light in 

 color and above the average grade. Both haddock and menhaden 

 meals gave quite satisfactory results with yearling trout. 



The cod-liver meal was also difficult to use without waste, as it was 

 very fine and light. Even when mixed with meat before feeding it 

 was easily separated on contact with the water; and, as with shrimp 

 and haddock meals, the trout left it alone after its precipitation on the 

 bottom of the trough. 



SHRIMP MEAL 



Shrimp meal is a waste product of the shrimp canneries and large 

 quantities are produced annually, althougli at the present time only 

 a small percentage of the available supply is being utilized. As 

 ordinarily sold it consists chiefly of the dried fragments of the head, 

 thorax, and legs of shrimp. The entire bodies of shrinq) which are 

 loo small for canning are also included, and not infrecjuently these 

 may form a considerable percentage of the total. It will thus be seen 

 that the product varies considerably in nutritive value but always con- 

 tains a high percentage of protein and chitin. Of course, the chitin 

 is indigestible, but may serve a valuable purpose as " roughage." In 

 this respect shrimp meal is quite similar to the natural food of trout. 

 Like most marine products, shrimp meal contains a relatively high 

 percentage of essential minerals and from this standpoint alone 

 should make a valuable adjunct to the fresh meats which are deficient 

 in this respect. 



In addition to the coarse meal referred to above, which is frequently 

 called shrimp bran, there is also available a finely ground product 

 which can be fed to small fingerlings. The ordinary shrimp bran is 

 too coarse to feed to fish under 4 or 5 inches long. 



When fed to yearlings in combination with fresh meats shrimp 

 meal has given results comparable to those obtained with fish meals, | 

 (Fig. 11.) While the growth is not as rapid as with "clam heads'" 

 it is quite satisfactory, nevertheless, and owing to tlie relatively low) 

 price of shrimp meal it nuiy prove to be more econfunical than the fishi 

 meals. When fed in combination with " clam heads " the growth was 

 nearly equal to that obtained when the latter was the only dry con- 

 stituent in the ration. The mortality in lots fed a ration containing 

 shrimp meal was low, and the fish were always vigorous with excelleni 

 coloration. 



Our experiments indicate that nothing is to be gained by the 

 addition of shrimp meal to mixtures of meat and fish meals, but 

 a beneficial effect was evident when it was added to a mixture of 

 sheep liver and soybean meal at a 10 per cent level. 



Shrimp meal is a less satisfactory food foi- fingerlings than fox 

 larger fish, a mixture of equal parts beef liver and shrimp meal 

 producing a relatively slow growth. (Fig. 5.) In our experiments 

 for 1928 (fig. 9) only menhaden meal was inferior to shrimp meal 

 in this respect, and in the following year the mixture of beef livei 

 and shrinq3 meal made the poorest showing of any of the beef-liyei 

 combinations. In view of these results the inclusion of any consid- 

 erable quantities of shrimp meal in the diet of young fingerlings is 

 not recommended. 



