26 U. S. BUREAU OF FISHERIES. 
to ascend to their natural spawning grounds. It is probable that 
the numbers of fish thus released for natural spawning exceeded 
those involved in the egg collections. This substation is located 
within 10 miles of the “top” of the Rocky Mountains, in the State 
of Idaho, and to reach this mountain stream for the accomplishment 
of their supreme mission in life the salmon travel approximately 
1,500 miles without food after entering the Columbia River at 
Astoria, Oreg. The conditions which were observed here in regard 
to the splendid spring run of chinook salmon were not confined to 
the Snake River. Similar conditions were observed in all tributaries 
of the Columbia River frequented by the so-called spring-run fish. 
Very excellent results in the rearing of young salmon have been 
obtained from the use of a comparatively new article of fish food. 
This is the immature spawn of the salmon taken by the commercial 
fishermen. The spawn is purchased at a comparatively low cost 
and held in cold storage until needed. In commenting on its value 
in his annual report for 1920 the superintendent of the station states: 
The salmon fry fed on salmon spawn and beef spleen produced remarkably 
fine fingerlings. It is by far the most satisfactory food that has ever been 
used in this field. No trouble was experienced with the fish at any point 
where its use was adopted. They were uniformly sturdy and readily accepted 
it throughout the year. 
While the expense involved in feeding fish on an extensive scale 
is considerable, in view of the quite general belief that the future 
supply of salmon in the Columbia and other rivers of the State is 
dependent very largely on the numbers of fingerling fish released 
from the hatcheries, the work is considered of first importance. 
Because of the long period during which so little is known of the 
life of the salmon—from the time of the seaward migration as a 
fingerling or yearling until its return for the reproduction of its 
kind—it is difficult to effect any positive check on the actual results 
of hatchery work. Nevertheless, the evidence in favor of the work 
at certain points, notably in the Columbia, Clackamas, and Rogue 
Rivers, is of a convincing nature, and it appears to fully warrant 
the expense necessary to continue the feeding and intelligent plant- 
ing of fingerlings. Upward of 68,000 pounds of fish food was used 
at the stations in the Oregon field during the fiscal year 1921. The 
details are given in the table of fish food on page 75. 
BAIRD (CALIF.) STATION AND SUBSTATIONS, 
[W. K. Hancock, Superintendent. ] 
In this field are to be recorded a number of untoward circum- 
stances resulting in a serious curtailment of the fish-cultural work. 
From the fall run of fish the aggregate egg collections for the entire 
field amounted to 7,910,000, as compared with 11,785,000 the preced- 
ing year, which included 1,349,700 eggs of the spring run chinook 
salmon taken in the McCloud River at Baird. 
Racks were again installed in the McCloud River in time to inter- 
cept the spring run of chinooks, but no fish appeared. This is di- 
rectly attributable to the dam erected in the Sacramento River at 
Redding for irrigation purposes. This dam had the effect of divert- 
ing into the irrigation ditches most of the water in the river during 
periods of low-water stages, and though an opening was left in the 
