PROPAGATION AND DISTRIBUTION OF FOOD FISHES, 1921. 29 
HATCHING EGGS IN GRAVEL, 
At the Afognak (Alaska) station an experiment in hatching eggs 
in gravel after the so-called Robertson method was undertaken dur- 
ing the season of 1920. Eggs to the number of 5,500,000 were placed 
in two old hatching troughs, mixed thoroughly with gravel, and 
bedded in the bottom of the creek. The troughs were fitted with 
a cover and water was introduced through a 1-inch pipe leading 
from a small dam, so that there would be a gradual flow through each 
trough. The intake of this pipe was carefully protected by a screen. 
On opening the troughs in the spring it was found that very few 
of the eggs had hatched. Though the winter was a mild one it is 
probable that the eggs had been frozen, as there had been a very 
hight fall of snow, leaving the creek unprotected, and frost may have 
penetrated to a greater depth than in seasons of lower temperature. 
Eggs placed in gravel boxes patterned after the Robertson plan and 
held in the hatchery developed nicely, and the majority of the fry 
remained in the gravel until the yolk sac was absorbed. 
SUGGESTIONS FOR POSSIBLE IMPROVEMENT AND ENLARGEMENT OF WORK. 
Further extension of the feeding operations at all stations where 
chinook, sockeye, and silver salmons are handled is desirable. This 
would necessitate an increased allotment for the Pacific coast work, 
since at the present time there is not sufficient capacity or funds to 
feed all stock resulting from the eggs collected. In fact, it has 
been necessary to close certain substations and curtail expansion in 
all fields. The extension of operations in other fields is essential if 
the work is to be kept at its highest efficiency. The take of eggs at 
many of our stations has been reduced through extensive commercial 
fishing operations during the period of the war, when heavy demands 
were made for this class of food, and it will take many years to re- 
store the streams to their former productiveness. 
The small statutory salaries provided for the lower grades of the 
bureau’s service and the consequent inability to maintain a trained 
and efficient personnel have been important factors in the declining 
output. Because of these conditions it has been necessary to employ 
and train new men at most points every season, and as competent 
men could not be secured or retained at the statutory wage provided 
station allotments have been drawn upon in many instances for the 
employment of temporary help at the prevailing high wages, thus 
greatly reducing the funds available for actual propagation work. 
The returns from the marking of young salmon on the Columbia 
River and at other points have not been wholly satisfactory. One 
reason for the small returns from these experiments may be the fact 
that it has been the custom to hold the fish intended for marking 
for several months beyond the natural spring migration, and this 
may have had a serious effect on the life of the fish involved. The 
thought has occurred that better results might be secured if the fish 
used in the marking experiments are held over one year, or until the 
migration of the following spring. The results of the marking ex- 
periments also suggest the desirability of a more thorough study of 
the early life of the salmon, with the view of correcting any possible 
weaknesses in present methods of propagation and distribution. 
This is a phase of the work that is respectfully referred to the 
division of scientific inquiry. 
