CHOICE OF PATTERN 109 



three patterns comprised in the series, and the rising 

 fish that cannot be tempted by one of them is either 

 preternaturally shy, or has reached so high a standard 

 of education that I must retreat beaten, but I hope 

 not disgraced. The fisherman who desires to take 

 full advantage of these patterns must take the trouble 

 to study and understand the entomological portion 

 of this book. He must take from the surface of the 

 stream specimens of the insects floating down, and 

 with the assistance of his magnifying-glass identify 

 them, so that he may select patterns which are 

 imitations of the flies on which the rising trout has 

 at least a chance of feeding at the moment. No 

 casual glance at a passing insect whether on the 

 water or in the air is a safe basis for identification. 

 Perhaps readers will scarcely credit it, but with a 

 number of iron-blue and pale watery duns rising off 

 the water I have times out of number found it im- 

 possible to differentiate them in the air. 



The reader may well ask what he should do if he 

 cannot see any flies on the water when fish are 

 rising. He should hold his collecting-net pardy 

 immersed in the stream in places similar to those 

 in which he sees fish feeding. If on a periodical 

 examination he finds no flies (spinners or smuts are 

 the most likely ones) in the net his only chance is 

 to think over the insects he has seen on previous 

 days or in previous seasons at the same time of year, 

 and fish the patterns of these. If he is still un- 

 successful he must keep on changing his flies, and 



