After the beginning of July the angler may have many a pleasant day's fly fishing though his basket will not often be heavy. The natural flies continue to hatch out in July and August, and the trout feed upon them, but in dilettante ways, and in all sorts of water—in deep still water, as well as in streams and pools. The rivers will as a rule be very small and clear, and the fish partly for this reason, and partly because they are now less intent at any given time of the day upon feeding, will be more shy and particular. It becomes desirable to cover a larger extent of water than is at all necessary earlier in the season, fishing one bit of water because the light summer breeze happens to be making a fair ripple on it; another because a fish is seen to rise; and a third because it is broken water with good sheltering stones, amongst which fish may be lying: but neglecting or passing lightly over many a stretch of water, where at the height of the rise earlier in the season dozens of trout would have been hooked. In fact, at this time of the season one has a roaming day, trying many places and many individual fish, succeeding only now and then, pleased with difficulties that are overcome rather than proud of the total, and half inclined to look upon all success as unexpected. Personally on such days I am apt to spend some of the afternoon very quietly, and to fall to remembering how the river looked in the spring and what happened then. Very small and gentle are the best streams of many north country trout rivers in July and August, and have then but a tinkling sound.

" Like to the noise of a hidden brook In the leafy month of June, That to the sleeping woods all night Singeth a quiet tune".

And the woods are thick and silent at this time.

In September the trout take better, but their condition is becoming suspect, the days are shorter and the glory of the trout fishing season does not revive.

A perfect trout fishing river is not very large. There is fine sport to be had in great rivers, such as the Tweed, but I would rather fish for trout in a smaller river, where the whole of the water can be covered by wading, and where trout can have the main stream to themselves as well as the shallower sides and eddies. On the broad part of the Tweed the question will occur as to whether it would not be better to use a boat, and thoughts of salmon continually intrude. The most famous trout fishing river in Northumberland used to be the Coquet, and any one who is curious as to its reputation and merit will find these set forth in the "Coquet-dale Fishers' Garland." In size and character and variety of water the Coquet is a perfect river for trout fishing, but the average size of the fish is small, much smaller nowadays we are told than it used to be, and smaller, it seems to me, than it ought to be. Why this should be so I cannot tell, nor why this change, if change it be, in the average size of the trout has taken place. There are plenty of trout, it is the size alone that is complained of, and this is the sort of complaint that is very frequent on north country rivers. Whether it is really founded upon fact, or whether it is only an impression, I cannot say. I have seen no actual records of average weight in other years which enable me to make comparisons between them and those of the present day, and we are apt to remember the larger and forget the smaller trout of our youth, just as we so often retain an isolated memory of very hot days in summer, or very cold days and deep snows in winter, and take these to be typical of what the respective seasons used to be in earlier years. Another river of which I often think, though it is about twenty years since I have seen it, is the Dart. Here too the average weight of the trout is small, as it seems to be in all west country rivers, but I did not hear upon the Dart the same complaints of a falling off in size. It must not, however, be supposed that there are not large trout in these rivers. I have seen a yellow trout, which weighed several ounces over two pounds, landed from a clear stream in the Coquet on fine tackle in July, but such things are exceptional; and in my own experience, even half-pound trout in these rivers are not very commonly hooked by fly fishers. Apart from this, the Coquet may be taken as typical of what an angler might wish a north country trout river to be. What then should be the average weight of trout caught with a wet fly? and how much should a good basket weigh ? Probably answers to these questions will differ. In my own opinion, any north country river of size corresponding with the Coquet, in which the trout averaged three to the pound, would be first-rate trout water, and as in dry fly fishing ten pounds' weight of trout landed by fly fishing would be a good day; whilst fifteen pounds would be a very good day, and twenty pounds or upwards exceptionally good. It has been my fortune to fish such rivers occasionally, but I am not sure that I have ever attained on a river to fifteen pounds' weight of trout in one day with a wet fly. It will perhaps be interesting to compare this estimate of what may be expected or hoped for in a day's fishing with that given by Stewart in his " Practical Angler." Stewart probably knew the rivers of the Tweed district as well as any one, and there was no doubt about either his skill or his success as an angler. He says, " There are not many days from May to October in which an angler, thoroughly versed in all the mysteries of the craft, should not kill at least twelve pounds' weight of trout in any country in the south of Scotland, not excepting Edinburghshire itself." In the same volume, in another passage, which refers to worm fishing in July, it is asserted that "he is not worthy of the name of angler who cannot in any day of the month, when the water is clear, kill from fifteen to twenty pounds' weight of trout in any county in the south of Scotland." It must, however, be added that a note, at the end of the first chapter of the seventh edition of "The Practical Angler," tells us that before his death Mr. Stewart confessed to a necessity for lowering this estimate, and we are warned to take into account his "ideas and habits as to a day; which a jealous gamekeeper whom he had always utterly beaten described as 'twenty-four hours of creeping and crawling.'"