NATURAL HISTORY OP THE SALMON. 83 



mistake in such experiments, the question perhaps could 

 not fairly be held as demonstrated. 



There is, however, one experiment which might 

 settle, and we thought had settled the matter beyond all 

 doubt — namely, the marking of grilse-kelts on their 

 descent, and their capture on their reascent. Mr. Young 

 of Invershin, writes that (besides marking salmon-smolts 

 that have returned as grilse) he has often marked grilse, 

 and that they have returned as salmon ; Mr. Mackenzie 

 says he has done so also, if not as often, with an opposite 

 result, the fish returning as grilse. There have been many 

 experiments conducted by marking on the Tweed ; but 

 these not having been conducted with any special refer- 

 ence to this particular point, the results are meagre 

 almost to uselessness. There is there no case at aU in 

 favour of Mr. Mackenzie's theory, but there is only one in 

 favour of ours. A grilse-kelt of two pounds, marked on 

 the 31st March 1858, was caught, on the 2d August of 

 the same year, as a salmon of eight pounds. Here again 

 there is a possibility of a mistake, and the matter really 

 remained to be demonstrated by experiment. That de- 

 monstration seems to have been supplied by the Stor- 

 montfield experimenters, by whom many grilses, marked 

 when grilse-kelts, have been recaptured on their reascent 

 as salmon ; but it might be well, now that the Tweed 

 Commissioners and Mr. Mackenzie have raised the ques- 

 tion, if experiments directed specially to this point were 

 repeated with increased care and in greater number. 



There is, however, considerable difficulty in the way 

 of obtaining conclusive evidence from experiments made 

 on the fish after it has assumed its migratory habits, and 



