STORMONTFIELD EXPERIMENT. 95 



water, their ova developing until ready to spawn, v 

 and although they get very much coloured by 

 being so long in the fresh water, and not a very 

 good-looking fish to send to the market, it is a 

 right good fish to eat/' We were so fortunate as 

 to capture one of these fish with the fly, in one of 

 the Sutherlandshire rivers, and, as we had never 

 seen a salmon so much coloured in the summer 

 season, we imagined it was a fish near the spawn- 

 ing. We were about to return it to the river, 

 when the keeper, who was with us, said it was a 

 fine clean fish, and put it in the creel It turned 

 out so, and, as Mr Ffennell says, " right good to 

 eat." Mr Ffennell also remarks, " I think it as 

 clear as possible, that the object of the law that 

 governs them is to cause the fish to distribute 

 themselves throughout the whole length and 

 breadth of the river/' These fish, by ascending \ 

 the river so early to. spawn, and having the spawn \ 

 so slightly developed, are strong, and able to I 

 overcome falls, and penetrate to the extreme \ 

 feeders of the river, whereas those fish that do 

 not leave the sea until heavy with spawn, could 

 never overcome falls, or ascend any considerable^/ 

 distance. Now, we think that the anomaly ascer- 

 tained to occur in the migration of the fry ex- 



