COLOUR VARIATION 53 



the three streams, and in a little while that part of 

 the bodies which presses against the others will 

 exhibit the same appearance, whilst the other parts 

 will remain as before ; -and hence the clouded aspect 

 they exhibit. I once threw a trout, by accident, 

 from a clear channel stream over my head into a 

 peat -moss pool behind me, which had no com- 

 munication with the running water ; and after a 

 few months I caught him as black and portly as 

 possible. Such facts certainly prove, to my own 

 satisfaction at least, that trouts do not vary in 

 original and indelible type so much as is generally 

 imagined. In regard to what follows upon the 

 changing colours of fish when in the act of dying, 

 I cannot speak with the same certainty ; but either 

 my eyes deceived me very much (and at the period 

 of life to which I refer they were pretty good), or I 

 observed the following phenomena : I usually 

 killed my fish, not by breaking their necks, as is 

 now generally the method adopted, but by slapping 

 their heads against a stone, the edge of my shoe, or 

 the butt of my fishing-rod ; and even when a boy 

 I was sensible of some change which took place in 

 the colour of the dying victim. A kind of streamer, 

 or phosphorus light, seemed to shoot along the quiver- 

 ing flesh, and only ceased with the life of the trout. 

 In salmon I should think the fact is still more 

 manifest. The salmon fishery at the Eden afforded 

 me an accidental proof of this. Some summers 

 ago I was in the habit of bathing near the stakes 

 at ebb tide, when the salmon were removed from 

 the nets. I had a pleasure in walking into the 

 inside of the nets, and seeing the finely -shaped 



