If we could watch salmon more in the water, as we can so often watch trout when feeding, we should learn much that would be of great practical advantage in angling, both in working the fly and in choosing size and pattern of fly for each day. After fishing for a few hours without a rise we get the impression that the salmon are not to be caught, and are taking no notice of the fly at all, but the latter is probably much less often correct than is supposed. Such opportunities as I have had of observing the behaviour of salmon at rest in the water lead me to think, that the fish are continually taking notice of the fly and following it when we do not see them. I was once fishing with a friend on a beat of the Spean in June when the river was very low. We came to one of the best pools and found it so low and clear that we felt sure that it was not worth fishing, but when standing on a high rock above the pool we saw one good salmon of nearly twenty pounds' weight, and four or five small ones, lying together on a patch of smooth flat stones in the middle of the bed of the river. It was agreed that one of us should go down and fish the pool, while the other remained above to observe what happened. My friend went first, and as soon as the fly reached the fish, one of the smaller ones followed it without breaking the water. Time after time the fly was cast in the same place, and one or other of the smaller fish continually noticed it by some movement, or followed it to the bank, but there was no rise, nor was the fly actually touched. Then I went down and my friend reported from above. I succeeded in moving the big fish; he followed my fly two or three times, but none of the smaller fish made any movement. Then my friend tried again and moved more than one of the smaller fish, but without getting a visible rise from any of them or stirring the big fish. When my turn came again the smaller fish never moved, but the big fish followed the fly right round, and at last made a rise at it with a visible boil at the end of the cast, but without being touched by the hook. That was our nearest approach to hooking a fish, but we had enjoyed half-an-hour's very exciting sport. It was impossible for the person fishing to see these salmon while casting over them, and had either of us been alone, we should no doubt never have persevered long enough to get the one visible rise, which we did get, and should have asserted afterwards with perfect confidence that we had never stirred a fish. One curious point was, that though we changed patterns and sizes of flies, and interchanged them with each other, I could not move one of the smaller salmon, but only the big one, while my friend at different times moved every one of the smaller fish and never the big one. We had also on this day a very good illustration of the value of knowing a river. We had often fished this pool before, when it was in better order and the fish were not visible, and we now saw that the fish were lying in exactly that part of the pool where we had most often risen or hooked a fish. The reason seemed to be in these particularly comfortable looking flat stones, on which the salmon rested, but till we had once seen this, we had never realised the special virtue of that one spot in the whole pool. An old ghillie, who had known a Highland salmon river for very many years, once told me of a similar experience. There was a deep black stretch of the river, about a quarter of a mile long, where the water flowed with a smooth even current between high wooded banks. This part was fished from a boat, and the old ghillie told me that for years he had known that in all heights of water there was only one particular bit of some six yards in the whole of it that was worth fishing; but he neither knew nor could guess the reason, till there came an unprecedented drought, and for the first time in his life he saw the bottom of this part of the river. Then going quietly down it all in a boat he saw the salmon lying together at this one place on some stones which were more flat and smooth than the surrounding rock and gravel.

This sort of knowledge must be learnt, in the first place, from those who know; we must in salmon fishing at first profit by the experience of past generations on each river and take it on trust. It would need a lifetime to discover the best places of a river afresh for ourselves. A salmon angler of great experience may make some good guesses at the favourite spots of a strange river, but even he is sure to mistake some useless places for good ones, and to pass lightly over some of the best. Now the most essential thing in salmon fishing is concentration on the best places; it is not enough to be told which are the good streams or pools, and to fish them all with evenly distributed care. The angler should fish all that he is told is good water, but he should concentrate his care and skill and perseverance on the best spots of the good streams and pools. The most successful salmon angler is one who feels expectation—it is more than expectation, it is almost faith, founded on previous experience—stir within him as he approaches certain well-known places. It is as if there was some magnetic influence in the angler's confidence, which predisposed the salmon to take his fly, and an angler who knows that he is fishing a good pool, but does not know exactly where to expect a rise in it, has not so good a chance of rising a fish as the man who has hooked salmon in that pool before, and knows not only that it is a good pool, but what is the best spot in it.

And yet salmon fishing is more lucky than any of the other sorts of angling discussed in this book. Luck does perform the most extraordinary feats on salmon rivers, and plays all sorts of tricks sometimes, but none the less is it true, that the angler who throws the longest line well, and knows the river best, will hook most fish in the season.

I know nothing which raises anticipation to such a pitch as salmon fishing, and nothing which so often wears it down by sheer unrewarded toil. There is much monotony about it; each cast down a long even stream is very much a repetition of the one before it, and when there is no result the angler first loses expectation, and then hope, and falls into a dull mechanical state. In summer and autumn salmon and grilse are often jumping and showing themselves, but in spring there are no grilse, the salmon do not jump, and you sometimes cast all day without seeing any sign of a fish, even when there are plenty of them in the river. I must admit that, after casting for hours without a rise, great despondency comes upon me, when it is a question of fishing a second time over a long piece of water that has already been tried with one fly without success. How often have I sat on the bank and looked at the unsuccessful fly, and wondered whether it was too big or too little, and then at my other flies, feeling that there was no reason why any one of their patterns should succeed better than the one already tried; till at last I have looked at the unconscious water and doubted whether there were any salmon there that day at all!