38 U. S. BUREAU OF FISHERIES 
served as a check on the relative value of the different marks used. 
Recoveries of fish in which only one ventral had been removed 
amounted to 663, while recoveries of the fish in which both ventrals 
had been removed totaled only 569. Apparently the fish marked by 
removal of the adipose and both ventrals had a lower survival value 
than did the other group. While complete returns from this experi- 
ment will not be at hand until 1940 these results indicate that caution 
should be exercised in comparing marking experiments in which 
different combinations of fins have been removed. 
A counting weir was again operated in the Karluk River to deter- 
mine the number of fish going up the river to the spawning grounds. 
Counting was continued until September 3, at which time 875,678 
red salmon and 1,639,589 pink salmon had passed upstream. Owing 
to the large number of dead spawned-out pink salmon drifting down- 
stream the weir had to be removed on September 4. While the 
escapement figures are consequently not quite complete, they are of 
considerable value. Data on the escapements and the returns from 
given escapements have been collected for a number of years and it 
is important that no breaks should occur in the series of data. 
More than 11,500 scale samples were collected during the season 
for the purpose of calculating the age group composition of the 
escapement and the commercial catch. The number of fish returning 
from each brood year is determined in this way and credited to the 
proper year. This study is of basic importance because it indicates 
what variations may be expected in the returns from known escape- 
ments and what is the optimum sized escapement for this and other 
watersheds of a similar magnitude and nature. 
Ovaries from fish ranging in size from 20 to 24 inches were pre- 
served and counts are being made to determine the number of eggs 
produced by each fish. The results of the present investigation will 
be compared with similar work done in 1926 to determine whether or 
not any change is taking place in the fecundity of the species. 
The seaward migration of red salmon fingerlings began on May 
96 and continued intermittently until June 16. For the purpose of 
determining the age composition and the length frequency of the 
various age groups making up the migration, samples were collected 
on all days the migrants were in the river. The migration appeared 
to be of average magnitude and the fingerlings were of the usual 
strong, sturdy stock so characteristic of this watershed. 
Predatory species —As the char Salvelinus malma, or Dolly Var- 
den trout, is very abundant in the Karluk watershed and is con- 
sidered to be one of the most important salmon predators, a study of 
this species is being carried on in conjunction with the studies of 
the salmon population. In 1937 a series of marking and tagging 
experiments were initiated to aid in determining the size, homo- 
geneity, growth rate, migration habits, and mortality of the popu- 
lation of chars inhabiting the Karluk River and Karluk Lake. 
During the past year three more experiments were started, in which 
external tags were used in lieu of the internal tags used in the 
1937 experiments. To recover the internal tags and discover the 
distinguishing number it was necessary to kill and dissect the fish. 
The external tag has a marked advantage in that a recaptured tagged 
fish can be identified by the number on the tag, measured, and then 
released. Recaptures of chars from these experiments are conse- 
