200 A SCOTTISH FLY-FISHER 



possess is based on inference rather than on observa- 

 tion. Not all the causes which determine the trout's 

 position in the flowing water are in operation on the 

 loch, but there is reason to believe that in the latter, as 

 surely as in the former, he holds a fixed and definite 

 place — until, driven by the most compelling instinct of 

 his nature, he leaves it for the stream. How far he 

 wanders from it in search of food we do not know, but 

 it is, I think, certain that he returns to it when his 

 hunger has been appeased, and remains in it con- 

 tinuously while at rest. To every loch-fisher it must 

 have occurred to raise and miss a fish, and take him 

 later in the very same spot. It is, of course, possible 

 to contend, and the contention is difficult to meet, that 

 the fish caught on the second visit to the water is not 

 that originally seen. It is so easy to make a mistake ; 

 fish of a size are hard to distinguish, and although the 

 angler may be perfectly satisfied that the trout he has 

 just consigned to his creel is the very trout he failed to 

 hook when last he fished the loch, he will rarely be in a 

 position to produce convincing evidence that he is right. 

 I have, myself, taken from beneath a stone, two trout 

 so exactly alike that, had I not had both in actual pos- 

 session, I should have been absolutely sure they were 

 the same. On the other hand, I have killed a trout in 



