THE SYSTEM OF ARTIFICIAL FLIES. 67 
tinctly unsound ; and if my reader will follow me in the 
next few pages, calling to mind, also, his own fly-fishing 
experiences, I have little doubt that he will arrive at a 
similar conclusion. In fact the arguments of the two 
schools are mutually destructive. 
The position of the ‘‘formalists” is as follows :— 
Trout take artificial flies only because they in some 
sort resemble the natural flies which they are in the 
habit of seeing; if this be not so, and if colour is the 
only point of importance, why does not the ‘colourist’ 
fish with a bunch of feathers tied on the hook ‘ pro- 
miscuously ? why adhere to the form of the natural 
fly at all? Evidently because it is found, as a matter of 
fact, that such a bunch of feathers will not kill; in other 
words, because the fish do take the artificial for the natural 
insect. If this be so, it follows that the more minutely 
the artificial imitates the natural fly the better it will kill; 
and also, by a legitimate deduction, that the imitation of - 
the fly on the water at any given time is that which the - 
fish will take best. 
To the above argument the “colourists” reply -— 
Your theory supposes that Trout can detect the 
nicest shades of distinction between species of flies 
which in a summer's afternoon may be numbered 
actually by hundreds, thus crediting them with an, 
amount of entomological knowledge which even a pro- 
fessed naturalist, to say nothing of the angler himself, 
very rarely possesses ; whilst at the same time you draw 
your flies up and across stream in a way in which no 
natural insect is ever seen, not only adding to the impos-’ 
F 2 
