SOUTHEASTERN ALASKA SALMON STATISTICS 
463 
distant places. The only explanation seems to be that an error was made in desig- 
nating the locality. 
The easternmost locality on the outside coast of the mainland is Dixon Harbor. 
It was fished irregularly from 1908 to 1924; all species of salmon were caught in some 
years but in no case did the total catch in any season exceed 6,000 fish. The runs, 
small as they were, became commercially valueless in a few years, and the harbor 
was abandoned, or if catches were made after 1924 they were reported as coming 
from other localities. 
Three small streams on the western slope of Yakobi Island have been fair pro- 
ducers of red salmon, and have been fished regularly since the establishment of a 
cannery on Dundas Bay. They are tributary to Hoktaheen Cove, Surge Bay, and 
Takanis Bay. Hoktaheen Cove is the least important as the catch there fell from 
the relatively high levels maintained prior to 1910 to very low levels in recent years 
while in 3 years no salmon were reported from the cove. The catch has now 
declined almost to the vanishing point, a sure indication of depletion. Takanis 
Bay appears to have been fished each season from 1905 to 1927, except 1921. It 
produced mostly red salmon, though a few thousand cohos, chums, and pinks also 
were taken there. The catches have fluctuated noticeably but without indications 
of periodicity, except as to pinks which were more abundant in the odd years from 
1913 to 1923, inclusive. Thereafter, pinks were taken in larger numbers in the even 
years. In respect of red salmon, the fluctuations in catch are peculiar, being high 
and low in alternate years from 1907 to 1915, with the peaks of production occurring 
in the even years. The downward movement continued however in 1916, and in 
the next few years, or until 1921, the better catches were made in the odd years, a 
complete reversal of the earlier record. The catch of 1922 was 1,334 reds, the 
smallest recorded up to that time, but it was followed in 1923 and 1924 by progres- 
sively larger catches. This marked the beginning of another period in which the 
even years took the lead in production. Nothing is known of the age of the red 
salmon of Takanis Bay. No scales have been studied and the record of the catch 
gives no indication of the sort of periodicity that would indicate their probable 
age. The trend of the catch was slightly downward to 1920 but since then it has 
apparently recovered, notwithstanding the extremely small catch in 1927. The 
significance of this apparent recovery is not, however, at all clear and it is not im- 
probable that within the next few years the direction of the trend will again be 
downward. 
Surge Bay is the most productive red-salmon locality on Yakobi Island. 
Omitting 1921, when it was not fished, the catch fell below 10,000 only once — in 
1913. There were three peaks in the production of red salmon, the first and highest 
occurring in 1907 and 1908, the second in 1915, and the third in 1924. Though no 
catch was reported in 1921, it is probable that the run was small and that the actual 
abundance was not above the average of the years immediately preceding. Since 
no fishing was done in 1921 it may be assumed that the escapement of fish to the 
spawning grounds was larger than usual — if, indeed, the run was approximately the 
same as in the preceding years. It is possible that the increased production from 
1924 to 1926 was the result of this. Unfortunately the routine observations on the 
spawning grounds which are now a feature of the work of the Bureau of Fisheries 
were not being made at this early date or we might now be in possession of some 
very valuable information as to the results to be expected from such an increased 
escapement as presumably took place in 1920. This bay produces all species of 
