PACIFIC SALMON 



63 



with the food supply present in given local- 

 ities, lead this species to remain in a lake 

 for a variable period. It has been found 

 that very few sockeye fry start immediately 

 for the ocean ; but that most of them remain 

 for a time in a lake where some of them stay 

 one year, others two and a few even three 

 years in fresh water as the extensive records 

 of W. A. and L. S. Clemens (1926-1937) 

 show. Such data indicate that variations in 

 environmental conditions modify the activ- 

 ities of salmon within the parent waters. 

 Foerster (1937) gives valuable data on this 

 point. 



The evidence which has been secured for 

 the return of salmon to their own native 

 waters in the opinion of some is not more 

 than sufficient to indicate a general compli- 

 ance with this practice whereas in other 

 minds it constitutes a habit little short of 

 absolute, involving a uniform return to the 

 tributary and even to the particular spawn- 

 ing ground involved. Dismissing accounts 

 by untrained observers, let us examine 

 briefly the evidence for the return obtained 

 by scientific students of the problem. These 

 data deal chiefly with physical facts in two 

 lines, (1) scale markings of the adults and 

 (2) the return of fish marked as young. The 

 array of data in these two fields is large and 

 impressive. It represents an important con- 

 tribution to the solution of this question and 

 is regarded as conclusive by many who have 

 devoted themselves to these experiments. 

 Nevertheless the data are statistical and 

 mechanical, and necessarily subject to the 

 limitations of all similar data. One is justi- 

 fied in asking how far the results obtained 

 for a few fish may correctly be extended to 

 cover the entire groups involved and whether 

 further evidence from a diverse point of 

 attack may not be needed to establish con- 

 clusions so far reaching and so heavily loaded 

 with implications concerning the charac- 

 teristics of living matter. I have myself felt 

 that more extended field studies on the 

 habits of the salmon and more definitely 

 controlled experiments completely followed 

 up would throw valuable light on the situa- 

 tion. The dangers of extrapolation in biol- 

 ogy are well recognized, and after all the 



fish is a living organism and not to be 

 handled merely as a simple mathematical 

 unit. 



Scale Markings 



In considering the implicit reliance of 

 some on scale markings, we must not over- 

 look certain other facts. It is well known 

 that in the past good investigators have 

 made mistakes in their interpretation of 

 such markings and that markings are not 

 uniform in scales taken from the same 

 population group in different years. Wliile 

 some of these differences have been ex- 

 plained, and perhaps rightly, there still 

 remain details unadjusted and differences 

 of opinion which cannot be overlooked and 

 should not be forgotten. In the hands of 

 some (Clemens and Clemens) the careful 

 discrimination of age groups has yielded 

 most valuable information on the life history 

 and migration of population groups in 

 British Columbia streams. Nevertheless I 

 think it fair to include in the category of 

 doubtful items also the fact that the extent 

 to which such scale markings furnish de- 

 tailed evidence concerning the origin of 

 population groups has not yet been fully 

 demonstrated. 



Some able investigators who have devoted 

 much time and care to the study of scale 

 markings are convinced that they furnish 

 thoroughly dependable means of determin- 

 ing not only the age of a salmon and the 

 varied experiences during life but also the 

 precise region from which it came. The 

 latter depends on minute characteristics 

 which, according to their studies, are shared 

 by scales from fish belonging to a particular 

 spawning ground, like individual finger 

 prints, and are not found in scales of salmon 

 from any other region. This was the view of 

 Gilbert regarding salmon from the Fraser 

 River region in which are numerous spawn- 

 ing areas and it implies great if not abso- 

 lute fixity of populations in these areas. 

 If this view be accepted, it is at once ap- 

 parent that no appreciable addition can be 

 made to the population of an area from any 

 other region, since, if foreign adults should 

 creep in, the scale markings would become 



