EXPERIMENTS AT HOWIETOUN 209 



from Teith salmon, which parents had, of course, 

 never descended to the sea. 



Apart from such experiments we have in the 

 so-called land-locked salmon of Scandinavia and 

 America races of fish, indistinguishable struc- 

 turally from our British salmon, which spend 

 their whole lives in fresh water. The term 

 1 land-locked ' is inaccurate ; for there is nothing 

 to prevent these fish from descending to the 

 sea, although there are certain obstructions in 

 the way of their return. They appear to find 

 the vast areas of fresh water to which they have 

 access sufficient for all their requirements. 



An important point to be remembered in the 

 consideration of the question of relationship is 

 the fact that in the youthful 'par' stage of every 

 race we find an almost absolute uniformity, so 

 that it is with difficulty, if at all, that they can 

 be separated with certainty. The cross bands 

 or ' par-marks ' are common to all, being lost in 

 most cases on reaching maturity, although per- 

 sisting throughout life in the case of the brook 

 trout of some smaller streams ; an instance, 

 apparently, of arrested development through 

 less favourable environment. 



All the evidence, so far considered, points to 

 the fact that the different races of our Sal- 

 monidae have certainly descended from one 



