Hours for Angling 137 



pool, just above which is a very swift, glassy, and 

 shallow rapid flowing into it over smooth rock. 

 The salmon will not rise in the deep water 

 where they stay; but the fly has to be cast at 

 the top of the rapid, down which it goes so 

 swiftly that it seems impossible a salmon could 

 get it, but it is taken by the fish coming up 

 from the pool and always before it reaches 

 there. Another place I know where a gradually 

 shelving rock goes down into the river. The 

 fish lie six or eight yards from the shore, but 

 take the fly close by the rock and wiU not rise 

 elsewhere. I recollect once hooking a big fish 

 veiy near the rock in low water. I saw him 

 coming for some distance, and as he drew near 

 the fly he turned on his side before seizing it. 

 I had seen this done before when I had hap- 

 pened to be over fish in clear, shallow water, 

 and incorrectly assumed that this turning was 

 habitual with salmon, as I have since had the 

 chance of seeing them take a fly as one nat- 

 urally supposes they would. Salmon often lie 

 on the comparatively smooth beds of river flats, 

 where there is no shelter in the way of big rocks 

 or bars. In such places, and indeed in others 

 with rougher bottoms, the fish are liable to 



