Some years ago, on the occasion before alluded to, when I was fishing the River Clady, in Donegal, the nets having been removed for that year, the river was full of fresh-run fish—it was in July. There was a pool in which the fish lay in serried rows in the stream, which at that point ran under a steep, high bank. I lay down on the bank overlooking and a little behind the rows of salmon, and some twenty feet above them. By shading my eyes I could make out all the fish as clearly as if I were looking at them in an aquarium. I arranged a code of signals with my fishing friend, and he went some thirty yards or so up the river to fish the pool. As soon as his fly began to work over the first line I signalled that he had got the length; there was, however, no movement among the fish. I then signalled to cast again with the same length of line. As the fly worked over the fish for the second time they all seemed to shun it, dropping down stream a foot or so, with the exception of one fish, which, separating from the others, came up some three feet to follow the fly, eventually leaving it and dropping back into his former position. A third passage of the fly produced similar results, the same fish moving again. He made a break in the water, which my friend saw, but he had come short. A fourth cast secured him.

I could come to no other conclusion but that the fish had been bored into taking that fly. His curiosity had been excited at first, and in ordinary circumstances the fisherman would have known nothing and passed on. Does not this tend to show that many a fish may be moved without our knowledge, and that a subsequent fly might secure him?

It is often thought that the first fly over a pool stands the best chance, provided, of course, that it is properly offered. Personally, I would just as soon follow a good angler down a pool as precede him. Unless a fish breaks the water in his rise, the fisherman can tell little of what is happening below the water level, except when, by chance, a glimpse of a silver flash is accorded him. But he may have moved a fish with his fly, and, knowing nothing, will have moved a yard down stream, his next cast being a yard below the fish. The next fly, suitably offered, if it be about the same size, may lure our friend to his destruction. Could we all know exactly what is going on under the water out of our sight, many more fish would doubtless be brought to bank. Of course, on those days when the temperature of both air and water have attained that precise relative proportion that seems to cause a simultaneous rise of fish in every pool, the first fly will pay best, for on such happy occasions that fly, however ill delivered, may secure the best fish. And what fisherman cannot recall instances of "duffer's luck," the veriest tyro catching, perhaps, the fish of the season? I remember once trying to teach a would be angler how to cast, and in a most unlikely spot—the river being dead low—was endeavouring to instil into him the rhythm of the cast, and trying to make him get his line out well behind him. Holding the rod with him, I kept the same length of line, steadily flogging the water to the tune of "one, two," when, at about the ninth or tenth cast, a travelling fish seized our fly, and eventually came to the gaff, a clean-run salmon of 18 lb.

But surely the precise pattern of the fly, within limits, is of small moment; the size, coupled with the proper working of the fibres, is the main thing. Every angler has, naturally, his own favourite shibboleth, mainly, in my opinion, because he has succeeded with it, and therefore perseveres with it far more steadily than with any other pattern. In the same way local fetishes are set up, and when once adopted are hard to shift. On the Beauly, years ago, fishing on that lovely water in the spring, we were using the orthodox spring fly, a sort of exaggerated Alexandra, and were mainly catching kelts. When one of us suggested a Gordon (having lately used it on the Dee) the fishermen laughed us to scorn, and said we might as well fish with it on the high road. Nevertheless, the fly was tried, and nearly all the clean fish we got that week were secured by it. When our time was up our gillies begged for our worn specimens of the goodly Gordon, and the next lessee caught all his fish upon flies of that pattern; and, for aught I know, that fly may now be reckoned as one of the standard flies of the river.

To revert to the original query. Can it be answered satisfactorily? Surely it must represent some food taken whilst the salmon are in their sea home; and yet, if this be the only probable answer, how comes it that on some rivers, as is the case in Canada, salmon cannot be persuaded to rise at any fly of the kind? After all, whether the question is unanswerable or no, the glorious uncertainty of salmon fishing forms one of its most potent fascinations. If every bungling cast hooked a salmon, few people would care for the sport.

All this said, then, what form of fly are we to use? Here we get upon very debatable ground, and whatever conclusion we arrive at will probably be strenuously opposed. The patterns of salmon flies are legion, many differing but slightly from others. Are we to credit salmon with such extraordinary intelligence as to believe them able to differentiate between varieties of almost similar flies, and to have such a correct eye for colour as to refuse a fly because the colour of the body or hackle is a shade unorthodox? The size of the fly, no doubt, is a most important factor, both as regards the size and volume of the river and the time of the year. It would be the height of absurdity to use in fine run water in the summer a three inch fly that would be a suitable lure on the brawling Thurso in the spring, and vice versÂ. The finer the water the smaller the fly—within reason.

So far, I think, we are all agreed. It is when we attempt to reduce the vast number of flies now in vogue that differences of opinion will begin to assert themselves.

On the whole, perhaps, there will be less divergence of opinion about that singularly fortunate combination of fur, feather, and tinsel, termed the Jock Scott. It seems, to an extraordinary degree, to be effective on most rivers where the artificial fly is used. The combination of colour is most happy, and the fibres of its mixed wing give it, in the water, a most life-like appearance. Few anglers would care to be without Jock Scotts of sizes. Similarly, in bright water the Silver Doctor is a universal favourite, and justly so. As a direct contrast the Thunder and Lightning is bad to beat, and I should be sorry to be without a Blue Doctor.

Eagles, grey and yellow, hold their sway on the Dee, and the play of the feathers seems to be alluring in the quick waters of that river. How would such a fly suit the quiet waters of the Avon? You would imagine that you might as well fish with a mop-head! The fibres of Eagles require fast, fleet water to make them work, and to use an Eagle as your lure in slow-running rivers would appear to be most inappropriate. The play of the rod point may, however, be substituted for the play of the water, and a tempting opening and closing of fibrous and mixed winged flies can be obtained by a judicious rhythmical raising and lowering of your rod point. Indeed, if you watch an experienced salmon fisherman from a distance, you can tell at once the kind of water his fly is working through. If the stream be sufficiently broken and rapid to work his fly automatically, his rod point will be still. If the water should be sluggish, you will note the work of the rod top. It would, therefore, be folly to dogmatise on such a matter, and I should be sorry to attempt to do so.

Gordons, Butchers, Wilkinsons, and a host of others have their staunch advocates.

It is, however, unnecessary to run through the whole gamut; suffice it to say that in my opinion, a good selection of, say four or five, would be as effective as twenty or thirty. The main difficulty is local prejudice, and the uncertain kind of feeling—that if you had not discarded local favourites your blank day might have been fruitful. Once, however, you have shaken yourself free from this feeling, you will very soon gain full confidence in your theory. The blank day that you are mourning would probably have been equally blank if you had been equipped with all that local fancy could suggest. Can it be seriously suggested that salmon can be credited with sufficient intelligence to refuse a Silver Doctor or Silver Grey and to accept only a Wilkinson? Is it not rather that the fly that was accepted was presented in a most alluring manner, whilst the others which were rejected did not come within the salmon's ken in such a way as to tempt him? Are we not all too prone to change our flies on the slightest provocation, and are we not all inclined to have our own favourite fetish—a fly that succeeds with us simply because we give it ten chances to one of any other? The vagaries of salmon are universally admitted; at one time they will allow all lures to pass them unnoticed, and in the next half hour may take any fly, of the proper size, suitably offered. The relative temperatures of air and water have, I feel convinced, much to say with regard to this. The fly in which an angler believes, and with which, therefore, he perseveres most, will bring him more fish to bank than any other.

It goes without saying that the fly that is most in the water, in the fishable parts of the pools, of course, will catch most fish. The patient, persistent angler has that great advantage over his less energetic brother of the angle. What angler is there, who ties his own flies, who has not built up a combination of fur, feathers, and silk by the river side, and, on trying the novelty, perhaps after days of disappointment, has found it unexpectedly to succeed, and who has thereupon fondly imagined that he has found a "medicine," only to be equally disappointed the next time it is tried? Scrope, in his day, seems to have been satisfied with five patterns. To come to later times and later writers, Sir Edward Grey and Mr. Gathorne Hardy both advocate four only. The colour of the bottom of the river, of the sky, the brightness of the day, or its cloudiness, all these will affect our choice of fly, whilst the size and volume of the water will affect our choice of size.