1104 U. S. BUREAU OF FISHERIES 



mental diets. The results were very consistent and in general agree- 

 ment with those obtained in previous years. The advisability of 

 using dry meals in the ration as a measure of economy was fully 

 demonstrated and it is confidently believed that by the utilization of 

 these products the food bill can be reduced at least one-third without 

 adverse effects on the growth or mortality. 



While the results of these experiments clearly indicate that trout 

 can not be kept successfully on dry foods alone for any length of 

 time, they also show that the amount of fresh meat required is less 

 than was formerly thought to be the case. Excellent growth accom- 

 panied by low mortality was obtained with rations containing as 

 high as 60 per cent of dry meals and it is probable that the amount 

 of meat can be still further reduced without appreciably lowering 

 the efficiency of the ration. 



Unfortunately, owing to local conditions, it has been found im- 

 practicable to continue the feeding experiments during the winter at 

 the Pittsford station. Since it is evident that the practical value of 

 these experiments is dependent to a large degree upon the length of 

 time they can be continued without interruption, it is highly desirable 

 that facilities be provided elsewhere for carrying on feeding experi- 

 ments throughout the year. Furthermore, the investigations at 

 Pittsford have shown that the nutritional requirements of other 

 species of trout are in some respects quite different from those of 

 brook trout. Consequently it will be necessary to conduct similar 

 experiments with rainbow and brown trout before the most desirable 

 rations for these fish can be worked out. 



BREEDING EXPERIMENTS 



As noted in previous reports, experiments in selective breeding 

 have been carried on at the Pittsford station during the past three 

 years. Owing to the necessity of rearing a large brood stock in 

 preparation for the breeding work it was possible to mate only a 

 few pairs of fish previous to 1928. In the fall of that year 34 pairs 

 were mated and the offspring kept in separate compartments in 

 hatchery troughs during the following summer. These fish were 

 all kept under as nearly identical conditions as possible so that the 

 young of different parents would be directly comparable. Although 

 all the parents were specially selected fish it was found, as was to be 

 expected, that the young in different lots showed marked differences 

 in rapidity of growth. The variations in growth among the fish in 

 each lot were, however, much less than is usually the case among 

 fingerlings of mixed parentage. In some lots the average growth 

 was over three times that of other lots, and there is no question but 

 that this difference in the growth rate must be ascribed very largely, 

 if not entirely, to difference in parentage. Only those lots of finger- 

 lings showing exceptional vigor and rapidity of growth are to be 

 retained for further breeding work. 



During the present season an even larger number of individual 

 pairs have been mated and the eggs and young will be kept separate 

 throughout the spring and summer. For the first time it has been 

 possible to mate selected young of known ancestry and the results 

 of these matings are awaited with a great deal of interest. 



