PRINCE WILLIAM SOUND, COPPER AND BERING RIVER SALMON STATISTICS 213 
with surprising results. Out of a total production of 868,594 salmon in the Montague 
district in 1924, Macleod Harbor produced 862,773 — a catch which has not since 
been equalled. There was a material falling off in catch of pinks and chums in 1925, 
cohos decreased slightly, while kings and reds increased. In 1926, large gains were 
made in the catches of chums, pinks, and reds, but kings declined 8 per cent and 
cohos about 50 per cent. The catch of chums, pinks, and reds fell off in 1927, whereas 
that of cohos and kings increased. In the five years for which data are available, 
the average yield of cohos was approximately 30,000; chums, 15,000; pinks, 416,000; 
kings, 1,562; and reds, 11,000. 
The outstanding feature of the Macleod Harbor fishery is the consistently good 
catch of king salmon, which is larger than that in any other locality of the sound in the 
same years. The district is too new and data are too few for comprehensive analysis, 
but it is safe to say that the bulk of the catch came from passing runs rather than 
from runs to streams of Macleod Harbor. Catches were made exclusively by traps 
set near the entrance of the harbor, which in itself is an indication that the local runs, 
if any, are unimportant so that seining is not profitable. 
Port Chalmers . — Chum and pink salmon were reported from this locality in six 
years, 1917 and 1918 and from 1924 to 1927. Catches of cohos and reds were insig- 
nificant, but those of chums and pinks have been of more importance. Contrary to 
the common rule, the largest catches of pinks were made in the odd years of 1925 and 
1927, while all other localities in the Montague Strait district show larger catches in 
the even years. This place seems to have been fished chiefly by companies located 
at Cordova using seines. The total absence of kings and the small number of reds 
in the catches afford some reason for assuming that the salmon came from local runs 
to the streams of Port Chalmers. Evidence of the interception of migrating salmon 
at this point, as noticed at the more southerly localities of the Montague shore, is 
wholly lacking here. 
Stockdale Harbor . — Small catches of pink salmon were made at this bay in three 
years, but the data are entirely too few for analysis. Although the catch in 1926 
was larger than that of the other two years, none was made in 1927, and it would seem 
that the run is commercially unimportant. The unallocated catch in this district 
includes salmon reported as taken at Marsha Bay in 1917, at Montague Point in 
1925, at Montague Island in 1917, 1919, 1926, and 1927, and at Rocky Point, Sandy 
Point, and Green Island in 1927. 
The Montague Strait district, considered in its entirety, constitutes an area of 
relatively recent exploitation since, prior to 1920, it was not known to offer any possi- 
bility for profitable fishery development. Small catches had been made in the bays 
of the east coast of Knight Island and along the northwest coast of Montague Island, 
but not until traps were driven in 1920 at points on the southwest coast of Montague 
was it discovered that a large part of the Prince William Sound run entered through 
Montague Strait and could be reached by traps driven from the shore. In the last 
four years, 1924 to 1927, catches have reached rather large proportions and occasioned 
interest in the possible effect they may have upon the runs of salmon to the inland 
waters of the sound. In reviewing the data for this district, it was pointed out that 
there was little or no evidence to support the notion that salmon moving along the 
west coast of Montague were bound to local streams. That dea seems untenable in 
view of the physical peculiarities of the island which is long and narrow and traversed 
lengthwise by a high and rugged range of mountains. The most plausible theory, 
and one supported by the evidence of tagging experiments, is that the salmon passing 
