1931 ANNUAL REPORT, 1930 47 



As beef liver is generally accepted as one of the most satisfactory foods for 

 trout at the age of those experimented upon, it was taken as a standard or check. 



The fish were fed at intervals of three hours; by beginning at 7.30 a.m., five 

 daily feedings could be accomplished conveniently. Increase in weight was 

 taken to indicate the rate of growth. 



"Before the fish were taken over for these experiments they had been fed 

 on beef heart to which a little beef liver was added, the proportion as estimated 

 by the hatchery manager being nine parts of heart to one part of liver." The 

 effect of the change of food, with the exception of beef heart, beef liver and beef 

 heart, beef liver and dry skim milk, caused a slower rate of growth as compared 

 with that later on. The change to beef liver alone appeared to cause a rather 

 unexpectedly large initial lag. 



"Horse meat did not prove to be a satisfactory food when compared with 

 beef liver, beef heart, or even beef melt." The addition of clam meal seemed to 

 make it less desirable, but when dry skim milk as well as clam meal was added to 

 horse meat, a very good growth was obtained. As it is difficult to feed just 

 the right amount of food to fish at the age experimented upon, it is possible 

 that they did not have to rely very much on the horse meat ingredient. 



Beef melt gave good growth after the initial lag was overcome. The trout 

 took to this food readily and ate considerable quantities, and it appeared to keep 

 their bowels in good condition. 



From the results obtained with beef melt, it merits consideration as a 

 constituent of fish diet. Although the proportion in the diet giving good results 

 was high, the actual cost at 2.5 cents per pound is low. 



"The cooked tripe was quite unsatisfactory." It appeared to be very 

 distasteful, and the fish lost considerable weight at first and the mortality was 

 high. 



Addition of clam meal to the diet gave good results and produced nearly 

 as good growth as that of beef liver and dry skim milk. Unfortunately, the 

 supply of meal in bulk at the present time cannot be depended upon. 



The experiments performed with trout showed that the order of the diets 

 according to their relative values was as follows: (1) beef liver and dry skim 

 milk; (2) beef liver and clam meal; (3) beef liver; (4) beef liver and beef heart; 

 (5) horse meat and dry skim milk and clam meal; (6) beef heart; and (7) horse 

 meat. Considerable irregularity in the growth curves were shown with the 

 diets, horse meat plus clam meal, beef melt, and tripe. 



A study of the weights of the fish seemed to show a tendency of groups of 

 fish having an initial advantage in weight though small, not only to retain this 

 advantage but to produce an accelerated rate of growth as compared with those 

 in duplicate troughs with slightly lower weight. Of the nine pairs of troughs, 

 five pairs showed this tendency. 



"This tendency would appear to show the necessity of careful selection of 

 fish for feeding experiments in order to get them as nearly equal in size as possible. 

 Incidentally, this may also show a racial difference in fish, a view which careful 

 observations of growing fish undoubtedly support. There is little doubt that 

 selective breeding, even that of mass selection, would materially raise the quality 

 of the fish." 



A record of the amount of food fed was also kept, but as the experiments 

 were conducted for so short a time and that during the period of the life of the fish 

 when it is difficult to estimate the optimum amount of food, the calculated 

 efficiency of the diets is no doubt lower than it should be. 



