APPENDIX. 307 



have been opened at different times, by various persons, and 

 nothing could be discovered in them. 



According to my experience, the case stands thus. When they 

 first ascend the rivers they will eat greedily enough jump at 

 flies of every description without hesitation devour worms, 

 grasshoppers, and even small fish. In the Lakes of Killarney they 

 are caught under these circumstances by trolling with both natural 

 and artificial minnow. At this period, as 'every salmon fisher 

 \nows, they will rise at his fly with eagerness. I have myself 

 found, in at least a dozen instances, the larva of insects, remains 

 of earth-worms, grasshoppers and various kinds of flies, in the 

 stomachs of salmon caught soon after quitting the sea. But, after 

 a month or six weeks' residence in the rivers, when the sexual 

 propensities and organs begin to receive their development, the 

 fish cease to eat, and then appear to be able to live for several 

 weeks without any food whatever. 



Even before this time, and when they first run up the rivers, 

 salmon are capable of bearing a long fast without injury. At 

 Dayree's bridge on the Jacques Cartier river, nine leagues above 

 Quebec, there is a tank, or reservoir, fed by a copious spring 

 gushing out of the bank of the picturesque dell through which 

 that fine stream runs. In this receptacle the fish which are not 

 injured in being caught, are sometimes kept three weeks or a 

 month, until a sufficient number are collected to be sent to the 

 Quebec market. Under these circumstances, they continue in 

 good health, and do not appear to lose flesh. 



There is a ford on the river Esk, about a mile to the eastward 

 of the town of Donegal in the north of Ireland, which in my 

 young days was a favourite resort of salmon in the breeding 

 season. The lower part of this ford, just above the commence- 

 ment of a small rapid, was generally the chosen spot. Here the 

 bottom consisted of loose gravel, the stream flowed gently, and, 

 in ordinary states of the river, the water was about twelve or 



X 2 



