THE LYON 85 



In seeking to elucidate the bearing which temperature has upon 

 the salmon's ascent of tributaries, it seems clear that we are dealing 

 with quite different conditions from those which govern the ascent 

 of salmon from the sea to fresh water. Once the salmon has 

 entered fresh water, the question of temperature seems to play an 

 important part. The steady ascent of the main river is continued, 

 though early fish certainly pause for considerable periods, and not 

 infrequently descend again, when met by exceptionally cold condi- 

 tions of water. Tributaries entering the lower portions of large 

 rivers are as a rule passed, by all early-running salmon, but upper 

 tributaries if supplied with good flows of water are entered when 

 their temperatures become approximate to the temperature of the 

 main river, or when their thermal conditions are already relatively 

 high. But it is clear that temperature must be treated in close 

 association with water-flow, and it may be with certain peculiarities 

 of water. Hence it follows that if the cause prompting fish to enter 

 any particular tributary at a particular time is sought (and we are 

 dealing exclusively with fish in which the spawning instinct is still 

 absent), we must study the temperature plus the conditions of water- 

 flow. We find salmon in large tributaries when the temperature is 

 that generally associated with winter and spring ; we find fish in 

 smaller tributaries only when the water has lost its wintry character. 

 A certain ratio between temperature and volume would therefore 

 seem to exist, which must be determined in every case ; and, given 

 the infinite varieties of the infinite number of tributaries in Scotland, 

 we are not surprised that different localities should present such 

 different results in times and seasons of successful fishing. 



This deduction has a most important application with regard to 

 the results which may be expected to follow the opening up of 

 natural and artificial obstructions, since such operations influence 

 neither temperature nor volume. On comparatively small tributaries, 

 while most valuable to enable fish to be more widely distributed over 

 suitable spawning grounds towards the spawning season, such works 

 would not influence the ascent of early fish. Also, in large 

 tributaries, such, for instance, as the Tummel, it seems to follow that 

 if any early fish are to ascend to the upper waters, the pass on the 

 falls must be of extremely easy gradient, since even a comparatively 

 slight obstruction, when the water is cold, is sufficient to check early 

 fish. 



