FURTHER PROOF OF THE PARENT STREAM THEORY 



By Alexander Robertson 

 Harrison Hot Springs, British Columbia 



Considerable attention has been paid by fish culturists to the 

 so-called parent stream theory, that is, the assumption that anadro- 

 mous fish return to the stream in which they were bred, to reproduce 

 their species. In the course of time the consensus of opinion was 

 that the salmon of the larger rivers, such as the Columbia and Fraser, 

 certainly returned to these rivers, and the controversy narrowed down 

 to whether the fish returned to the individual tributary in which they 

 were hatched. By scale readings of the Pacific Coast salmon Dr. 

 Charles H. Gilbert came to the conclusion that they returned to the 

 actual creek in which they were bred. Further light on the subject, 

 by one who has made a study of it on the spawning grounds for 

 nearly twenty years, may be of interest to those engaged in fish 

 culture. 



Ever since the first hatchery was built on the Fraser River it has 

 been common knowledge among hatchery men that the runs of sock- 

 eye salmon to its various tributaries differed in many ways, chief 

 among which was the time of arrival at the spawning grounds. The 

 latter could be depended upon to such an extent that one crew of 

 spawntakers could operate several stations, one after another, from 

 September to January, with no variation in the sequence from year 

 to year. 



Another characteristic, and one which as far as the writer is aware 

 has never been given much prominence, is that the difference in the 

 size of the sockeye eggs at the different creeks has been so apparent 

 and constant in the course of time as to pass without comment. 



In 1914 the writer began a series of measurements of sockeye 

 eggs taken at Morris Creek, Harrison Rapids, and Cultus Lake, 

 which has been continued each year since. Morris Creek enters Har- 

 rison River two miles above the rapids and Cultus Lake lies twelve 

 miles southwest. The three stations are thus comparatively close 

 together, which fact adds interest to the subject. 



To obviate the inaccuracies of imperfect chambering such as occurs 

 when a graduated glass or similar measure is used, a new method of 

 measurement was devised. A light V-trough one meter in length 

 was set up at an inclination of twenty-five degrees and the eggs 

 allowed to roll down until the trough was filled from end to end with 



87 



