Though it would not be true to say that hitherto writers on fly-dressing have shown any lack of power of observation, still it is unfortunately true that their energy seems, strangely enough, to have stopped short at observing the natural fly, and has not been sufficient to carry them on to making even passable imitations, except of EphemeridÆ. With the exception of this family of flies, no one could possibly recognise the artificial through knowing the natural fly which it is supposed to represent. Yet the fisherman who knows the natural fly well by sight will go on using these imitations year after year unquestioningly; and though he himself would certainly not have known, unless he had been told, what natural fly the imitation he is using is meant to represent, he expects the trout to do so at once.

There has been much discussion recently as to whether trout have the power of discriminating between different colours, but no one has ever cast a doubt on their power of discriminating between different shapes; yet in most of these imitations it is not the colour that is wrong, but the shape. The wings of a fly undoubtedly play a most important part in forming the outline, and consequently the general appearance of the fly. Therefore, if they are not put in the natural position, the whole contour of the imitation must be entirely different from that of the natural fly.

It seems, however, judging by the standard works on the subject, that there is practically but one recognised position for the wings of the artificial fly, as the difference between the position of divided wings and wings dressed flat together is, after all, but slight. No one seems yet to have realised the fact that the wings of a May-fly do not lie in the same relative position to the body as do those of the Blue-bottle, whilst in the case of the Alder there is a further marked distinction from both.

The wings, in the different families of flies upon which trout and grayling feed, lie when at rest in three distinct positions in relation to their bodies.

In the EphemeridÆ they lie in planes approaching the vertical, slightly diverging from each other towards their extremities. Fig. 1 gives a sketch of one of the EphemeridÆ, and Fig. 2 a transverse section through the line a of Fig. 1. These drawings show the relation of the wings to the body. All flies have so far been treated by writers on fly-dressing as though their wings were in this position.

In the Caddis-flies (Trichoptera) and the Alder-fly (Sialis lutaria) the wings lie on each side of the body, meeting at their upper edges in front, gradually diverging towards their lower edges and posterior extremities.

Fig. 3 gives a sketch of an Alder, and Fig. 4 a transverse section through the fly, showing the position of the wings.

In the Diptera (Blue-bottle, Cow-dung, &c.), and PerlidÆ (Stone-fly, Yellow Sally, &c.), the wings lie in a horizontal plane. In some Diptera the wings diverge from each other towards their extremities, as in the Blue-bottle, shown in Figs. 5 and 6. In some other Diptera and in the PerlidÆ, the wings lie over each other, as shown in Figs. 7 and 8,. It will be seen that the wings in both these cases lie in a horizontal plane.

In Figs. 2, 4, 6 and 8 represents the section of the body, a and ? the section of the wings.

I wish it to be thoroughly understood that these positions are the positions of the wings of the natural fly when at rest.

Many flies when they fall on the water buzz round in circles periodically, apparently with the object of disengaging themselves from the surface. Between these efforts, however, their wings generally assume the normal position of rest. The only way to imitate the fly when it is buzzing is by dressing it without wings, and with extra hackle; and this is, after all, but a poor imitation. In most cases it is better to imitate the wings at rest; and if this is done accurately, it will present to the trout an accurate imitation of the natural fly as it appears to him when not trying to raise itself from the water.

I have on many occasions watched the behaviour of an Alder when it has fallen on the water. At first it moves its wings rapidly, but soon stops, to begin again, however, when it has rested. This is repeated time after time, but after each succeeding struggle, the interval of rest becomes longer. In many cases, however, the fly hardly struggles at all.

In observing many other flies which had fallen on the water, I have seen the same sequence of events occur, though some flies struggle to raise themselves from the surface much more than others, as in the case of the Blue-bottle.

The first trial that I made of a fly dressed with the wings in the natural position was with an Alder. To make this trial complete, I purchased some Alders, dressed according to the most approved patterns, from three well-known firms of tackle makers. When I got to the water-side the trout were rising freely, and the banks were literally swarming with Alders. I saw a trout take one which had fallen on the water, so it was evident that the Alder was the fly to use. I began with the flies I had purchased, and cast over a trout which was rising under a tree. He would not look at it, and the same happened with the flies of the other two makers when I cast over two other trout. I then tried one of my own, and got a fish at once. He did not take it in a half-hearted manner, but was hooked right in the back of the tongue. I then tried the other flies again without success. When, however, I went back to my own fly I hooked the first fish I cast over.

Imitations of other flies made with the wings in the natural position have served me as well as did my imitation of the Alder, though I was not inclined to try the ordinary patterns so freely on every occasion as I was at the first trial. I have, however, several times caught a rising fish on one of my imitations when he had refused the ordinary imitation not two minutes before.

My reader will of course think that these experiments, being carried out by myself, are hardly a conclusive proof of my theory, as, however impartial I might wish and believe myself to be, I must be naturally biased in my own favour. I quite realise that this is a natural doubt, but fortunately others besides myself have tried my flies.

Mr. Herbert Ash put them to an even more severe test than I did myself, and has kindly permitted me to give his experience. I give an extract from a letter written by him and published in Land and Water on October 23rd, 1897, as I think it is a very pertinent testimonial to the practical success of my theory.

“I put up a cast of three Alders, two being the shop-tied patterns which I usually used, and the third, which I put on as a first dropper, being Mr. Walker’s. I landed eight trout in about an hour and a half, and each of those fish took Mr. Walker’s fly.”

“Now, although I used three flies, I was fishing up stream and dry, my object being to test the new mode of tying the Alder, and I found that while the fish rose boldly at the first dropper, not one took any notice of the other flies.”

Colonel Walker also had much greater success with flies dressed with the wings in the natural position than with any others. In fact, for several consecutive days, on different occasions he caught no fish except with my flies, though he did not use them more than flies dressed in the ordinary way.

Several other fishermen have told me that their experiments with my imitations have produced similar results.

Mr. H. H. Brown, of the Piscatorial Society, after I had read a paper to that Society on my theory of the right way to dress trout flies, described a very interesting experience which he had one day when out fishing, and which bears directly on this theory. While out fishing some time ago, he rested on a bridge over the river in which he was fishing. There were a great number of Alders about, and on observing some fish in the water some little distance below the bridge, he caught some Alders, pinched their heads slightly in order to either kill them outright or at any rate stop them struggling, and threw them on the water. He was in such a position that he could observe each fly individually until it either floated past or was taken by the fish. What he observed was, that when in killing the fly he had disturbed the natural position of the wings, not one of the fish would look at it; while, if the wings remained in the normal position of rest, the fly was always taken. This occurred time after time, and not once was the fly with the wings in an unnatural position taken, but, on the other hand, not a single fly with its wings in the natural position of rest was allowed to pass. He also observed that once or twice the fish came up to look at a fly whose wings had been disarranged, but on getting close to it they always drew back.

This is, I think, an extremely strong argument in favour of my theory.

I do not propose in this work to deal with EphemeridÆ, as the wings in the imitations now sold are in the natural position. The families I do propose dealing with are the SialidÆ, Trichoptera, Diptera, and PerlidÆ, as no one has yet, to my knowledge, described the position in which the wings of the imitations of these flies should be put.