126 U. S. BUREAU OF FISHERIES 



Owing to the difficulties in obtaining yearling fish of sufficient 

 uniformity for experimental work the number of fish in our lots has 

 varied from 75 to 225 in different years. In no case, however, has 

 the number been sufficient to result in overcrowding. Each season 

 the same number of fish was used in each lot except in a few compart- 

 ments which were somewhat larger than the standard size. In such 

 instances larger numbers of fish -were used so that the number of fish 

 per unit area of water surface was the same in all compartments. 



FRESH MEAT AND FISH PRODUCTS 

 BEEF LIVER 



At this late date in the history of fish culture, despite marked dif- 

 ferences of opinion on other foods, no one challenges the real value of 

 raw beef liver as a trout food. Time-tried and time-proved, its use 

 as a single element in the diet of hatchery-raised fish has doubtless 

 met with greater success than has any other meat. 



Because of its steadily advancing price, investigators have been at 

 work for some time past to discover some substitute for beef liver. 

 Heart, liver, lungs, and spleen of cattle, pigs, and sheep, as well as 

 other meat products, have all received attention; but beef liver has 

 withstood the competition, especially as a growth producer. Not 

 that these other meats failed to give good results, but beef liver gave 

 better, particularly with the smaller fingerlings. 



Embody (1918) in search of cheaper trout foods, showed that cer- 

 tain combinations of dried animal meals when cooked with flour could 

 be used successfully, providing 45 per cent of raw beef liver was 

 fed also. On the other hand, very carefully conducted experiments 

 by Morgulis (1918) led him to conclude that while beef heart was 

 very satisfactory, beef liver was not a good food for trout. This 

 conclusion was based on the ease with which beef liver gave up its 

 components to the water, a great deal of nourishment being thus 

 wasted. Fish culturists early recognized this fact when feeding 

 finely ground liver, and the bureau's reports contain frequent sug- 

 gestions as to how to feed liver less wastefully. One of the most 

 common practices is to mix the ground liver with some flour or 

 middlings which absorb the meat juices and act as a binder. 



One of the first experiments undertaken at the Pittsford experi- 

 mental station was the testing of the comi)arative values of beef 

 liver, beef heart, and sheep liver for brook and rainbow trout finger- 

 lings. In comi)aring results from these three meat diets with the 

 brook trout (fig. 1) it was found that until the ninth week of feeding 

 there was little difference in the growth of the respective lots; but 

 as the weeks went by those on beef liver began to forge ahead, 

 finishing at the close of the experiment definitely in the lead. The 

 mortality among these fish was higher than in the beef-heart lot, 

 although it was nmch lower than in the sheep-liver lot. There 

 appeared to be little doubt that the increased growth after the ninth 

 week was because it was by then unnecessary to grind the beef liver 

 so fine, and there was less waste. 



Rainbow fingerlings on these diets (fig. 2) reacted in practically 

 the same manner. In comparing the growth curves of the two 



I 



