HOMING INSTINCT OF PINK SALMON 
29 
caught in the stream during their seaward migration from the spawning grounds 
where they had hatched under natural conditions. The fry were caught with a 
small screen wire trap net (see fig. 6) that was anchored in the middle of the stream 
where the current was the swiftest. Although thousands of fry were caught in the 
trap, not one was caught during the day, the fry migrating only at night. Many 
trips were made to the spawning grounds throughout the period the fry were migrat- 
ing, but invariably no pink-salmon fry could be found during the daytime, either in 
the stream or in the pools along the sides of the stream. With the aid of a lantern 
they could he found in the stream any time during the night, but at dawn they dis- 
appeared, and those that did not reach the bay hid under the rocks or in the gravel. 
The actual marking of the fry did not involve any difficulties for the method of 
marking had been worked out at Duckabush the year before. The only change in 
the routine from that used at Duckabush was that the marking was done at night so 
that the fry could be liberated at the time they normally migrated. 
INTERPRETATION OF RESULTS FROM MARKING EXPERIMENTS 
HOMING INSTINCT 
The 36,000 pink-salmon fry that were marked at the Duckabush River hatchery 
in the spring of 1930 came from the spawn of 1929. Working upon the supposition 
that the pink salmon mature at 2 years of age, 5 the 1931 run of mature salmon in 
the Duckabush River was examined for individuals bearing the marks. In order to 
determine the extent to which the Duckabush River pink salmon returned to their 
parent stream to spawn, the 1931 runs in all the streams along the south shore of the 
Strait of Juan de Fuca, east of Port Angeles (see fig. 3) were likewise examined for 
individuals bearing the marks. 
The hatchery operations on the Duckabush River made it possible to examine each 
one of the 3,800 mature pink salmon that composed the 1931 run. This was done at 
the time they were removed from the river and put in a retaining pond where they 
were held until fully mature and ready to spawn. The runs in the other streams along 
the canal and the south shore of the strait were examined by frequent observations of 
the schools of salmon as they migrated into the shallow waters on the spawning 
grounds. Where possible the salmon were collected in the shallow areas by means of 
a large net and carefully examined for individuals bearing the marks. The hatchery 
attendants at the State hatchery on the Dungeness River examined the 40,000 pink 
salmon in that stream that were spawned artificially. The observation of the schools 
of salmon on the spawning grounds began in the latter part of August and continued 
until late in September. 
Out of the 3,800 pink salmon examined in the Duckabush River 5 females and 3 
males were found bearing distinct adipose and dorsal scars. One female bearing both * * 
a dorsal and adipose scar was observed in the Hamma Hamma River in which approx- 
imately 1,500 pink salmon were spawning. One male bearing both a dorsal and adi- 
pose scar was found dead along the banks of the Dosewallips River in which approxi- 
mately 5,000 pink salmon were spawning. 6 The dorsal scars on most of the marked 
5 Gilbert (1913) from a study of the markings on the scales of pink salmon taken in various localities came to the conclu- 
sion that the pink salmon invariably mature at 2 years of age. Furthermore, since the runs of pink salmon in the Ducka- 
bush River and neighboring streams occur only on alternate years, the pink salmon in these streams must either mature at 2 or 4 
years of age. A further discussion of Gilbert’s work is given in this paper under the section “Age at maturity of pink salmon.” 
• The Hamma Hamma, Duckabush, and Dosewallips Rivers are the only streams on Hood Canal in which pink salmon spawn 
in any numbers. The pink-salmon runs in these streams usually occur at the same time during the season. 
