226 THE PRACTICAL FISHERMAN. 



using those colours only which represent the prevailing tints in the 

 selected families. What, then, are the most favourite families of flies 

 most favourite, that is, in the eyes of the trout ? Without question, the 

 ephemeridce and phryganidce, and for a very good reason, as, with hardly 

 an exception, they are all bred in the beds, banks, and reeds of the waters 

 over which they afterwards fly. To the first-named family belong, roughly 

 speaking, the whole collection of the ' duns ' and ' spinners,' the drakes or 

 May flies, the dark mackerel, the sand fly, and the March brown ; whilst 

 the latter includes the cinnamon, the grannom, or green fly, the willow fly, 

 and more important than any the stone fly, or ' water cricket,' which, 



in the early part of the year, is so plentiful on many rivers 



" As regards form or shape no question can arise, as the selected 

 families are all unmistakably and characteristically flies in the proper 

 sense of the term, having wings, legs, and, I think, without an exception, 

 'whisks,' or hair- like appendages at the tail. These whisks are not 

 only very ' fly-like ' and distinctive features, but are also easily imitated, 

 and assist materially to disguise the hook, as well as to make the fly 

 swim straight. This last is an important point, as the effe ;fc of the extra 

 weight at the bend of the hook, unless counteracted by some additional 

 * float/ is to make the fly swim tail downwards. The great majority of the 

 most favourite river flies belong to the order Neuroptera, or nerve- 

 winged insects, the wings of which, being filmy and transparent, cannot 

 be really imitated by feathers, or by any other available material. Wings 

 are, therefore, merely an encumbrance to the artificial trout fly, and 

 should be entirely rejected. 



" The next point is colour. On examining the fresh caught ephemeridce, 

 or phryganidce (for those in entomologists' collections are generally faded), 

 it will be found, in the first place, that there is almost always a general 

 similarity in colour, though not in the exact tint, between the wings, the 

 bodies, and legs, and that the colours which predominate indeed, almost 

 monopolise are greens, yellows, and browns. Our typical flies should 

 evidently, therefore, be of these colours. 



" Moreover, the colours of the bodies of the ordinarily imitated flies 

 made of silk dubbing, &c., generally change when wet, and thus lose 

 another important item of the exact imitation, whilst, as a rule, they 

 always lack the glossy, semi-transparent appearance of the real 



insects 



" Size, a most important point in artificial flies, demands the next con- 

 sideration. As we have no longer imitations of individual species, size is 

 a matter of no moment as regards the flies themselves, though of the 

 utmost consequence in another point of view. And this is one of the 

 .greatest advantages which those who may act on the principles here 



