"Wherever the natural fly can float, " there the artificial can float if properly " made, and oiled, and used. It is most "interesting to watch your fly coming "down dancing on the waves, and then "disappear when the brown head of a " trout breaks the surface; also to see it " pulled under when a trout takes one of "the wet flies." Once I was witness of the same feeling expressed without aid from the literary art. I was casting into a pool in a Hampshire stream. " See! see!" some one behind me exclaimed. I turned ; and there, in his shirt-sleeves, was the landlord of the little inn at which I was staying. With outstretched arm he was pointing to something in the blue air athwart the copse bordering the water, and his eyes were gleaming with some sudden joy. I looked towards where he pointed. It was the first Mayfly of the year that had moved him. The innkeeper was a reserved, shy man, who at ordinary times could scarcely be induced to talk at all ; but that fluttering Mayfly, symbol of summer at the noon and all the green world at the freshest intensity of throbbing life, had stirred him to a panic happiness. Now, something of the same joyous emotion comes at the sight of a fly, with cocked wings floating, that has been lightly cast where a great fish is known to be on the lookout. This I know full well. Still, as we are endeavouring to treat the whole subject in a scientific spirit, it is necessary to point out that the delight is not always unmitigated. Sometimes a trout takes the floating fly; but how often does he rise and miss ? In my own experience missing is the rule. Up comes the trout, and down he goes ; but the fly is where it was, on the surface. It is not that I have missed the fish. It is that the fish has missed the fly. This is very often what happens on a river, and it almost invariably happens on a lake. Were it not for an astonishing fact, which I will mention immediately, it would argue wariness on the part of the trout. One might believe that when a trout has risen at a floating fly and gone down without it he has detected or suspected the thing to be a lure. It is pleasant to think so, for much of the fascination of the sport is derived from the feeling that human skill is matched against astuteness in the fish ; but now I fear that another tradition must be sacrificed, or at least modified. If the trout suspects the artificial fly, he is equally suspicious of the natural I Day by day, as I write, I have been watching his behaviour carefully. It is not only my fly that he usually misses. He misses the real insect as well!

Has it been generally noticed that there are at least three different kinds of rises ? There is the rise, in leisurely manner, which is, as it were, finished off by a slow wave of the trout's tail above the water. That, though I cannot make out why it should be so, is when, early in the day, there are on the water myriads of minute black-and-white gnats upon which the fish are feeding. Then, there is the ordinary rise, when, if it be at a natural fly, the trout just tips the surface and retires without anything like a somersault, and when, if it be at an artificial fly, it is a business-like swift action without fuss. That, I think, is when the fish are feeding on insects slightly below the surface. Again, there is a rise which is hard to describe but beautiful to see. No part of the trout is visible; but he must have been very active for an instant. Swiftly the water breaks, swirling as the ripples rapidly expand, in a manner quite different from that of the ordinary rise, which is usually but a slowly-spreading dimple. That is a few minutes after a hatch of the larger insects. The trout do not move when the single spies appear; but when the battalions are abroad the movements are rapid and exhilarating.

Surely this, if ever there be one, is the time for the floating fly ! Of course it is, and I do not neglect it; but it is necessary to confess that my best efforts are almost always of little avail. Not only do the trout miss my own flies : they constantly miss the real flies. Sometimes I see one taken; but much more frequently the insect is still afloat on the swirl as the trout goes down. By and by, when the rise of flies has gone on for a time, or when the whole hatch has been on the water for half an hour or so, I find fish with a cast of flies slightly sunk. Why ? The obvious explanation seems to be that, although the trout begin to rise soon after the first risings of the fly, they do not begin to feed in earnest until many of the flies have been drenched.

Reasoning thus after many days of observation on lake and stream, I thought it would be well to examine methodically the literature of the Dry Ely. Surely, I felt, there must be some scientific consideration, which I had completely overlooked, to account for the practically unanimous enthusiasm with which the anglers of England had accepted the theory that artificial flies should float ? Well, I obeyed the judicial impulse ; and, after diligent search, I came upon relevant evidence which was surprising. The passages presenting it were these :—

"No doubt the Salmonidae in rivers " will at times take, and take freely, " winged flies on the surface ; but, besides "minnows and other small fish, Crustaceans and Molluscs, their staple food " consists of Caddis or larvae of Trichop-"tera, and the larvae of Ephemeridae, "Perlidae, Sialidae, Diptera, and many " other land and water-bred insects.

"As one of the few fishermen who "have for many years consistently "studied the food of the trout and " grayling by the only available and practical means, i.e. autopsy, may I be " allowed to tender my evidence ? I have "invariably found that the undigested " insect food has consisted of masses of " larvae and nymphs, with a few occasional " specimens of the winged insects. This "has been the universal result, whether "the trout or grayling have been taken "in waters fished daily, or in compara-" tively wild parts where they seldom see " an artificial fly. In rivers where in the " memory of man no stocking had taken " place, or in others, which, from neglect "or other causes, had been depopulated, " and where, therefore, a fresh generation "of trout had been turned in from the " pisciculturist's ponds, the experience "has ever been the same. The earliest "autopsies taken do not differ at all in "this respect from those of the latest " date.