498 
BULLETIN OF THE BUREAU OF FISHERIES 
ities where the chums run later than the pinks. Of equal importance in this connec- 
tion is the fact that the catches from 1921 to 1927 were made with fewer seines and 
traps than were operated in the preceding years of intensive fishing. Some places in 
the district show signs of depletion of certain species, as for example, Basket Bay in 
reds, and Chaik, Kelp, and Sitkoh Bay, and Tenakee Inlet in all species; but aside 
from these localities there are no definite signs of weakness in the runs of any species. 
The salmon fisheries of the northern part of the strait, therefore, may be regarded as 
having held their strength against the exploitation to which they were subjected. 
The southern part of the Chatham Strait district includes 12 localities, equally 
divided between the Kuiu Island and Baranof Island shores, which have produced 
several thousand salmon of all species through many years, while the strait itself pro- 
duced yet other large numbers of salmon. The history of its fisheries is similar in 
some respects to that of the northern part in that it dates from 1892 and shows the 
exploitation of the runs of red and coho salmon at Gut Bay, Bay of Pillars, and Te- 
benkof Bay, in the same manner as the red salmon streams of the northern section 
were fished. Not until 10 years later was any serious effort made to utilize other 
species, but beginning in 1902, pink salmon were taken and in a few years they be- 
came the most important fishery product of the district. The six localities on the 
Baranof shore are small bays which support insignificant runs of salmon and are 
fished by seines mainly for the reds that come to these streams, Port Walter and 
Patterson Bay being the exceptions. Fishing at these localities and at Port Herbert 
and Falls Creek Bay began much later than it did at Gut Bay and Ked Bluff Bay, 
and it was apparently very irregular as the catch data show intervals of two and three 
years in which no salmon were taken. Even if these bays were fished each year and 
the catches allocated only to Chatham Strait, the fact remains indisputable that 
there are no important fisheries on the east shore of Baranof Island south of Kelp 
Bay. On the other hand the west shore of Kuiu Island constitutes the most produc- 
tive field in the southern section of the strait, especially the north shore of Tebenkof 
Bay and the shore between Washington Bay and Kingsmill Point; but with this 
difference that the runs at these places are not necessarily local whereas those on the 
opposite side of Chatham Strait are strictly so. The large catches in both places 
were made by traps, and while the catches at Tebenkof Bay probably include some 
salmon that were bound to the streams of that bay, they also with equal probability 
contained large numbers of salmon that were destined to more northerly waters. The 
configuration of the shore at this point is such as to lead the runs into the bay before 
they round Point Ellis and continue their northward journey. It was possible, there- 
fore, for traps on this shore to reach these deflected bodies of salmon and make large 
catches before the migrating fish left the bay. Salmon taken along Kingsmill beach 
are also largely moving to more distant localities, chiefly in the Frederick Sound 
district, a fact that was fully demonstrated by tagging experiments in 1924 and 1925. 
The Bay of Pillars runs, of course, are not touched by traps at Kingsmill as it is not 
likely that they compose any part of the migration north of Point Sullivan. 
Bay of Pillars was one of the first localities to be fished in the southern part of 
Chatham Strait, solely for the reason that a tributary of the south arm supported a 
run of red salmon. It was a steady producer from 1892 to 1924, but after this arm 
was closed in 1925, due to the evident exhaustion of the run, very few red salmon 
have since been reported from Bay of Pillars. The falling off in the catch of other 
species, except chums, is also very evident as the total take of salmon in this bay in 
1927 was only 4,455. 
