A BOOK OF FISHING STORIES 



be prohibitive ; but I have in view that type of glen which 

 occurs so frequently on our western coasts, deeply cut by small 

 streams through boulder clay, or " till," as they call it in the 

 north. The sides of such a glen are usually steep, but not 

 so steep as to offer any difficulty in cutting a new channel 

 along them, through the grass, fern, and gorse with which 

 they are generally clothed. Farther inland, in moorland or 

 hilly districts, the plan of campaign must be modified ; but 

 the cardinal principle holds good, namely, that the main 

 channel of the stream must be diverted, and the pools to be 

 formed must be fed by a runner from the stream, not by the 

 stream itself. 



There remain two questions to be considered : (i) What 

 quality and quantity of trout may be expected in fisheries 

 formed upon this system } and (2) What purpose is there 

 in forming them, supposing the owner of the ground to be 

 indifferent to the attractions of angling ? 



(i) It is an established fact that the size and quality of trout 

 which any water is capable of producing are in inverse ratio 

 to their number. The food supply in every stream or lake 

 is a fixed average quantity. Trout, unlike the higher verte- 

 brates, such as men or mice, have no average standard of 

 dimension ; they are of so plastic a nature that the weight and 

 size of individuals conforms in proportion to the amount of 

 nutriment each can obtain. Assuming, therefore, that in any 

 water there is a good or fair supply of food, the average size 

 of the trout therein will vary according to the spawning 

 accommodation. In swift streams with extensive gravelly 



136 



