240 KNUT DAHL. 



Just as suddenly the characters sharply and concisely 

 distinguishing grownup salmon from trout asserted themselves 

 from 45 cms. length and upwards. 



Surely, the systematical works, especially the swedish 

 ones, ewerywhere contained remarks upon the difficulty in di- 

 stinguishing between salmon and trout in just these middle 

 stages; but this circumstance did by no means satisfy me. 



How could the young of one species, so different from 

 those of another species, that even children easily may learn to 

 distinguish them, how could the young, upon reaching a certain 

 length suddenly become like those of another species and then 

 again after some lapse of time and upon reaching a more ad- 

 vanced stage of growth become like itself. 



The contradictions contained in this chain of reasoning 

 seemed to me to point to the possibility that previous authors 

 on this subject might have confounded trout, especially shiny 

 slender and rounded trout with young salmon, and that other 

 investigators no more than myself had been able to procure for 

 examination salmon between 16 and 45 cm. in length or salmon 

 between the smolt stage and the grilse stage. 



In other words I was compelled to adopt for preliminary 

 use the working hypothesis, that salmon of the above mentioned 

 sizes were unknovn or at least undescribed. 



To solve this problem two things have been necessary. 



First I have had to catch and further rear in seawater the 

 young salmon (smolts) when emigrating from the rivers, and 

 further during their developement compare their characters with 

 those of trout of corresponding size. 



Secondly I have had to examine those collections of young 

 salmon on which the systematical descriptions of salmon in the 

 abovementioned stages (16 — 45 cms.) are founded. Most of these 

 are as far as I know to be found in the "Riksmuseum" in 

 Stockholm, and the late direktor of this museum, professor F. A. 



