HOMING INSTINCT OF PINK SALMON 
31 
River, 1 in the Dosewallips River, 3 in the Dungeness River, and 1 in Morse Creek. In 
a controlled experiment wherein 500 fry were kept for a number of months after being 
marked it was found that their dorsal fins had a greater tendency to regenerate than 
did their adipose fins. It may be that some of the 13 salmon with only adipose scars 
that were found in the Duckabush River were marked individuals whose dorsal fins 
had regenera ted. However, owing to the frequent occurrence of salmon whose adipose 
fins are missing due to natural causes, 7 it is exceedingly illogical as well as hazardous 
to consider any of the individuals with only adipose scars as marked salmon whose 
dorsal fins regenerated. Especially is this true of the 3 pink salmon with only adipose 
scars that were found in the Dungeness River, for out of the 40,000 pink salmon that 
were examined in that stream when they were spawned artificially not one was found 
with both its dorsal and adipose fins missing. 
The 50,000 pink-salmon fry, marked in Snake Creek at Olive Cove, Alaska, in the 
spring of 1931, hatched from the eggs spawned in the stream in the fall of 1930. Since 
marked pink salmon returned to the Duckabush River at 2 years of age, it was be- 
lieved that they would likewise return to Snake Creek at 2 years of age. Hence the run 
of mature pink salmon in Snake Creek and the runs in the neighboring streams were 
examined during the summer of 1932 for individuals bearing both dorsal and adipose 
fin scars. 
The runs of pink salmon in Snake Creek are usually rather large, ranging be- 
tween 50,000 to 100,000, so that it was impractical to try to catch and observe each 
fish in the run. By making use of the counting weir that is operated in the creek 
it was possible to devise a method whereby a large portion of the individuals com- 
posing the run could be examined. An enclosure or trap was built above the weir, 
as shown in figure 7, into which the salmon could pass through a V-shaped opening 
in the weir. The opening into the enclosure was never blocked so that the salmon 
could pass through it at all times. The salmon that were trapped in the enclosure 
were removed by means of large dip nets, examined for missing fins, and then 
thrown into the stream above the weir so that they could continue on their way to 
the spawning grounds. At times during the day the salmon collected in such large 
numbers in the stream that it was necessary to open the gates of the weir (see 
fig. 7) and count them as they passed through on their way up stream. An accu- 
rate account was kept each day of the number of pink salmon lifted from the trap 
and the number counted through the weir. Although only a portion of the entire 
run of pink salmon in the stream was examined for marked individuals as it passed 
through the trap, it was possible, as will be shown later, to calculate the number of 
marked individuals in the entire run. The 1932 run of pink salmon in Anan Creek, 
a neighboring stream (see fig. 8), was examined for marked individuals in a similar 
manner by the operation of a trap that was built in its weir. 
In addition to the examination of the 1932 pink-salmon runs in Snake Creek 
and Anan Creek a survey, similar to that made in the streams along Hood Canal in 
1931, was made of the other pink-salmon streams in the vicinity of Olive Cove. 
An examination of the 1932 run at Olive Cove shows that out of a total of 
7,944 pink salmon lifted out of the trap, 23 (10 males and 13 females) had both dorsal 
and adipose fin scars, 5 had only dorsal fin scars, and 10 had only adipose fin scars. 
Although the marked individuals (those with dorsal and adipose fin scars) are the 
7 In both marking experiments it was not uncommon to find a fry whose adipose fin had never developed. By marking the fry 
under the reading lenses it was not difficult to determine the presence or absence of the fins. 
