648 U. S. BUREAU OF FISHERIES 
of sockeyes. It purposely was situated on a tributary in which native blue- 
backs formerly spawned, and an effort was made to simulate further natural 
conditions. The marking experiment conducted at that station resulted in 
the recovery of 2,500 adults in the commercial fishery, but none returned to 
the hatchery stream. Why they did not return to the place of liberation or 
where they did go is not known. There may be some significance to the fact 
that this particular race of sockeyes, in their native river in Alaska, has a 
migration route of only a few miles from the ocean to the spawning grounds, 
whereas the station at which they were liberated on the Columbia is approxi- 
mately. 500 miles from the ocean. The instinct to return to the place of libera- 
tion may have been present, but the fish may have been physically unable to 
make such a long migration. The small amount of fat stored by them may 
have been insufficient to supply the energy neeessary for the long migration 
without food. The importance of this feature of the homing instinct lies in 
the fact that a hatchery can not become permanent in its operations unless it 
can produce its own breeding stock. Eggs can not be introduced indefinitely 
from some other locality. 
The chinook marking experiments are too numerous and too varied in nature 
to be discussed in detail at this time. Only the most general conclusions will 
be given. From the standpoint of the number of returns, the chinook experi- 
ments have been far less successful than those with the sockeye salmon. The 
reported recoveries have varied from 1 out of each 50,000 fingerlings liberated to 
1 out of each 390. The reported recoveries are not as representative of the 
actual number of returns as in the case of the sockeye experiments, but this 
discrepancy will not account for the great difference in the returns from the 
two species. The reasons for the difference in the returns is unknown. One 
of the chinook experiments was identical in every detail with one of the sock- 
eye experiments. The fish were even reared together in the same pond, and 
the same mark was used on both species, but the return of the chinooks was 
only one-sixth as great as that of the sockeyes. It is hoped that future experi- 
ments, which have been designed to involve a greater variety of hatchery prac- 
tices, may give better returns and point the way to more successful hatchery 
operations. 
As with the sockeyes, the question of the quality of the flesh has been an 
important one. The chinooks of the Columbia vary in quality from a maximum 
for which canners are willing to pay to 13 cents per pound to a minimum for 
which they can pay only 2 cents. An effort has been made to determine if this 
character is hereditary or if it can be changed by changing the environment of 
the fish through artificial propagation. The results are not conclusive as yet, 
but the indications are that, as with sockeyes, the quality of the flesh is deter- 
mined by heredity. 
The homing instinct has been found to be the same with chinooks as with 
sockeyes. They instinctively return to the tributary in which they spend the 
early part of their life, whether this is their native tributary or one into which 
they were introduced. Their migration may be altered, however, by unfavor- 
able conditions. ; 
Another feature of the spawning migration that has been found to be heredi- 
tary is the time of year when the fish leave the ocean. Chinooks enter the 
Columbia during every month of the year, but fish having the same ancestry 
are to be found only during a definite and regular part of each season. 
The knowledge of the nature of these two features of the spawning migration 
has thrown doubt upon the advisability of indiscriminately transferring the 
eggs from one tributary to another. If under natural conditions all or nearly 
all of the salmon return to spawn in their home tributary, it would follow that 
each tributary supports a separate colony or race of salmon. As a result, there 
would be a constant interbreeding between fish of the same colony and an 
absence of cross breeding between different colonies. This would lead to the 
development of racial differences through which the fish would become adapted 
to the conditions that prevail in their home tributary. We have found that 
character (of starting the spawning migration at a definite period) to be 
among these racial differences. If, then, the progeny of any race are trans- 
ferred to a tributary in which different conditions prevail, the resulting adult 
fish may not be able to adjust themselves to the new conditions, and as a 
result they may not be able to deposit their eggs. Several marking experiments 
have indicated that this is possible. These experiments were with a race of 
echinooks that normally spawns at a considerable distance from the ocean and 
that starts its upriver migration three or four months before spawning time. 
