52 U. S. BUREAU OF FISHERIES 
content and by lowering the temperature, which is normally about 
54° F. Eggs were also incubated in aquaria supplied with circu- 
lating soft water. 
The results were inconclusive and difficult to analyze, but appar- 
ently none of the variations in the normal water supply were par- 
ticularly beneficial. Both brook and rainbow eggs in soft water high 
in oxygen gave the poorest results. Efforts to incubate brook trout 
eggs in brook water at lower temperatures failed on account of silt 
deposited on the eggs. On the whole, both eggs and fry in normal 
hatchery water did as well or better than the other experimental lots. 
The failure to find any definite fault in the water supply indicates 
the desirability of investigations along other lines, and it is proposed 
to study the effect of environmental conditions and diet of brood 
stock on the developing eggs. That this may prove a fruitful field of 
research is indicated by the fact that egos from hatchery trout that 
had lived several months in a large pond where they were dependent 
on natural food gave a much better hatch than the average. 
California trout investigations —Good progress was made during 
the year on the inland research projects of this unit. As noted pre- 
viously, the work has proceeded along two major lines of effort; the 
Hot Creek brood stock experiments, and coastal stream steelhead 
studies. Personnel remained the same with Dr. P. R. Needham in 
charge of the fteld program, assisted by A. C. Taft. Two assistants, 
Leo Shapovalov and Leo Erkkila are assigned to the work by the Cal- 
ifornia Division of Fish and Game as part of the joint cooperative 
program actively initiated in 1932. 
Work at the Hot Creek station continued as planned and the first 
eggs were taken from the three select lots of fall-spawning rainbow on 
October 26, 1936. A total of over 800,000 rainbow eges were taken, 
of which about 500,000 were eyed and hatched at Hot Creek. A con- 
siderable loss of both female and male adult brook rainbows was 
caused by furunculosis. 
Experiments in rearing trout on natural food in ponds were con- 
dintiedl for the fourth consecutive year at Hot Creek. This year only 
loch leven trout were planted in the four ponds available. The 
ponds were stocked in June and operated for approximately 3 
months. An average survival of 72 percent was obtained, the best 
since these exper iments were begun in 1932. In previous years’ opera- 
tion of the rearing ponds during the summer, the survival was only 
about 40 percent. The reason for the higher survival rate obtained 
in the summer of 1936 may lie in the fact that few predator birds fed 
in the ponds this season and but few of the large leeches, He/obdella 
stagnalis, abundant in the ponds in previous years, were present. The 
growth rate of the fish per ‘month was about the same as that obtained 
im previous years, or approximately one-half inch per month. 
Detailed records of anglers’ catches were again taken during the 
open fishing season on Upper Angora and Convict Lakes. <A series of 
gill-net sets were made in Upper Angora Lake after the fishing sea- 
son had closed, to determine from samples of both marked and un- 
marked fish taken what age groups, species, and sizes of fish had sur- 
vived. The complete data from catch records obtained over three 
fishing seasons are now being analyzed and prepared for publication. 
Detailed plans were drawn up for the development of a major 
stream- and lake-management project on the Convict Creek drainage 
