HOMING INSTINCT OF PINK SALMON 
33 
total number of salmon lifted out of the trap, and 54 equals the number of marked 
salmon in the entire run of pink salmon. 
Out of the 132,351 pink salmon that composed the 1932 run at Anan Creek, 
13,965 were examined as they were lifted out of the trap. Two individuals hearing 
fin scars were found among the salmon caught in the trap, one had only a dorsal 
fin scar and the other had only an adipose fin scar. No marked salmon were either 
caught in the trap or observed in the stream. Although the salmon with only the 
dorsal fin scar might have originated from a fry whose adipose fin was missed when 
marked, the absence of any marked individuals (those with both dorsal and adipose 
fin scars) in the salmon caught in the trap makes it rather improbable that this 
individual was a Snake Creek salmon. 
The stream located nearest to Snake Creek in which pink salmon spawn in large 
numbers is Thoms Place Creek, approximately 10 miles south of Olive Cove and on 
the opposite side of Zimovia Strait. This stream was visited a number of times dur- 
ing the summer when the pink salmon were migrating into Snake Creek. During 
these visits only a few pink salmon were seen in the stream and none of these was 
marked. Toward the last of August when the run in Snake Creek was practically 
completed, a large run of pink salmon migrated into Thoms Place Creek. A survey of 
this stream was again made at this time, but no marked pink salmon were observed 
in the run. 
A survey was made of the streams along Eastern Passage, the streams in Brad- 
field Canal other than Anan Creek, and the streams along the shores of Stikine Strait. 
Owing to the unusual lateness of the pink-salmon runs throughout the whole district it 
was not until late in August that the pink salmon collected in the shallower regions of 
the streams where they could be observed. About 200,000 pink salmon in all were 
observed as they swam about in the shallow waters of the streams. No individuals 
with both their dorsal and adipose fins missing were found in any of the streams that 
were visited. 
In view of the returns from these marking experiments, it is conceivable that the 
extent to which the pink salmon return to their parent streams to spawn may be 
dependent upon the proximity of other pink-salmon streams in the vicinity. That is to 
say, the pink salmon composing the runs in streams that are more or less isolated from 
other pink-salmon streams may show very little or no tendency towards straying 
whereas the pink salmon in streams flowing into bays and in close proximity to other 
pink-salmon streams may stray more or less into the neighboring streams. The stray- 
ing of the marked Duckabush River pink salmon into the Hamma Hamma and Dose- 
wallips Rivers may be due to the close proximity of these streams to the Duckabush 
River. The Dosewallips River is located 4 miles north of the Duckabush River and 
the Hamma Hamma River is located 9 miles south of the Duckabush River. 
The apparent lack of straying of the Snake Creek marked pink salmon might 
well be due to its isolation from the other large pink salmon streams in the district. 
Although Thoms Place Creek is only 10 miles south of Olive Cove, the main run of pink 
salmon in this stream is usually much later than the run in Snake Creek. No marked 
individuals were found either among the pink salmon in this stream during the early 
part of the summer or in the large run that occurred late in August. The other pink- 
salmon streams in the district are all located more than 20 miles distant from Olive 
Cove. Anan Creek, in which a weir trap was operated in a manner similar to the 
operation of the trap in Snake Creek, is 25 miles distant from Olive Cove. The 
54190—34 2 
