1909.] The Life History of the Pacific Salmon. 3y 



migrations are annual and not quadrennial and the old birds do not all 

 die off when the eggs for the next generation are laid, but take part in the 

 migration. If it really exists the homing instinct of the Pacific salmon 

 is more wonderful than that of any birds. 



To obtain direct proof that the fish return to the very streams in 

 which they were spawned is naturally very difficult, and the evidence in 

 favour of this idea is entirely circumstantial. Thus it is said that the 

 Indians can distinguish between the fish which belong to the various side 

 streams when they are taken in the main Fraser River, but it is not stated 

 how the identifications of the Indians were checked. Further it is stated 

 that there is a marked difference in the size of the eggs deposited by the 

 fish frequenting different tributaries of the Fraser, those of Silver Creek, 

 for example, laying eggs that average seven thousand to the quart, the 

 eggs from Morris Creek averaging eight thousand, and those from Har- 

 rison River six thousand, and although the latter two streams are close 

 together it is said that the fish with the larger eggs are never taken in 

 Morris Creek and vice versa. In other words it is held that there are 

 different local races of Sockeye which never commingle on the spawning 

 grounds. 



Rutter gives an interesting observation bearing upon this question. 

 It is known that before 1897 no spring salmon bred in Paper Mill Creek, 

 which opens into Tomales Bay, California. In that year 855,000 fry 

 were set free in the creek and in 1898 2,000,000 alevins were added. In 

 1900 a few salmon were seen in the stream and in 1901 they were abundant. 

 This seems a very pertinent observation, but it does not by any means 

 prove the existence of a homing instinct. And on the opposite side of the 

 account may be set an observation by Babcock. In 1898 a dam was built 

 across the outlet of Quesnel Lake which effectually prevented access to 

 the lake by fish; but in 1904 a fishway was built on the dam and in the 

 following year large numbers of fish spawned in the lake. In this case 

 the occurrence of fish in the lake cannot be assigned to a homing instinct 

 since no fish had spawned in the lake for seven years. 



President Jordan, who has the right to speak authoritatively, ex- 

 presses himself very sceptically concerning this theory. He believes that 

 there is no definite homing instinct, but that the fish when they reach 

 the sea never wander very far away from the mouths of the rivers in which 

 they were spawned and hence when the stimulus to seek fresh water 

 seizes them, they are apt to encounter their native rivers rather than any 

 others. It is a question of accident rather than instinct. This seems to 

 be the most plausible view to take of the matter and may be accepted 



