The Yellow Breeches is a limestone trout stream not far from Carlisle, Pa., and about two hours north of Washington. From the village of Boiling Springs for a mile downstream to Allenberry, fishing is restricted to the use of artificial flies, the loftiest level of the fisherman's art. This way of taking fish was first mentioned in the third century by the philosopher Aelian. Dame Juliana Berners and Isaak Walton both enlarged on it in the 15th and 17th centuries respectively. Hundreds of others since have continued the noble work.

The stream is high and the air is cold today. There are few flying insects about. Under such conditions trout seldom come to the surface but feed instead underwater on the larval and pupal stages of insects and also on small fish and freshwater shrimp. It is the kind of day when a fisherman ignores his dry flies and uses those which sink below the surface to imitate underwater creatures. This is precisely what most of two or three dozen fisherman spread out along the stream are doing.

I find it a difficult and unsatisfying way to fly fish. With a dry fly I know when a fish has struck because I see it happen. But with a wet fly the fisherman must sense the slightest motion through the line or in the rod tip. Or through sheer intuition. Some fishermen become quite expert at it and even use wet flies later in the season when natural flies are plentiful. They know that even when some trout are rising to the surface to capture floating insects, many more, and larger, trout are feeding under the surface.

So, having satisfied the canonical requirements of the sport for four hours, during which I have tried dozens of fly patterns designed by generations of learned men to give the impression of being any of several hundred species of insects in various of their underwater forms, I have reached a self-serving conclusion: I would rather be skunked dry than wet. Any day.

There is poetry in dry-fly fishing. The sight of the rise -- the bulge in the water and the ring that rolls outward, the flash of the fish's colorful side as he arches over, and sometimes, some glorious times, the whole fish leaping from water to air and back. Where is the poetry in "feeling" that a trout has taken the fly in some unseen area below the surface? Touch works effectively for some, but the eye creates a work of art in the mind, a painting that can be recalled and re-savored many times, can even become more sharply delineated and take on signifcant details in retelling.

My mood improves immediately when I switch. The casting itself is more meticulous and precise. The line and leader must fall so that the float of the fly is unimpaired and natural. The exact spot the fly will fall is planned in the mind in advance. A system of rewards and punishments is established. For a good cast, the sight of the fly floating freely, not dragged across the current by leader or line, the attitude of the fly -- wings upright, hook below, hackle resting on the surface film of the water. Or the punishment for haste and inattention -- fly upside down or sunken, the line landing and floating in front of the fly and spooking the quarry, the leader failing to turn over and "present" the fly properly. Casting a dry fly can be an end in itself. All too frequently it has to be.

In England, dry-fly fishermen do not cast until they spot rising fish. This makes the game a hunt for rises. The technique is known as fishing the fish rather than fishing the water. Some Americans, myself included, tend to fish the water out of sheer impatience.

Now, having switched to a dry fly, I move upstream to where the water runs flat and rather deep. I locate a stretch empty of fishermen and wade to midstream. I make several casts toward the far bank, each succeeding cast putting the fly about a yard further downstream. The fly is a royal coachman Wulff, size 14, a relative large, pretty fly that I can see well, selected with little thought for the preferences of the assumed fish. I am using a fairly short (seven-foot) nylon leader tapering down to a 2X tip, on the heavy side with a breaking point of about four pounds.

After 15 or 20 minutes of fine casting, I think I notice a slight rise out from the far bank. More alert now, I concentrate on placing the fly four or five feet upstream of the spot where the rise appeared. After several minutes it appears again, a tiny ring of water that dissipates itself quickly. The sort of rise a three-inch chub might make. Or a very small trout. But it is the only evidence of fish I have witnessed in five hours of fishing with half my body in 53-degree water. Okay, so I'll fish the fish. And by the book.

Still standing in midstream, I first remove the coachman. I remember having seen a few flying insects which looked in size, color and shape something like a size 18 (the larger the number, the smaller) beige caddis fly.

But first I tie on a finer tip to make the leader less visible and longer, reducing the possibility of the line spooking the fish. Eventually the leader has been extended to 10 feet and the tip is fine enough to match up with the smaller fly.

Now check all pockets. If anything can fall out, it will. Make sure the wooden staff is floating in back of me and not where it can tangle the line. Look over the shoulder to guage how far the back cast can go without reaching the bushes. How close are the overhead branches of the sycamore? Too close -- maybe 10 feet -- but I have little choice because the rise was just opposite the tree. Now false-cast in the air to fluff out the fly. Let it drop on the water away from the rise to check the float. Lengthen the false casts to about 30 feet. Now drop the fly five feet upstream of the tree trunk and about four feet out from the bank.

Not good. The current is slower near the bank and the fly begins to float unnaturally, pulled askew by the line, which has fallen into faster water. I pick the line off the water carefully, false-cast once to check the distance, then let the fly drop. This time a loose loop of line falls behind the fly, permitting it to float freely.

Suddenly the water bulges under the fly. A fraction of a second later my left hand pulls the line down a few inches as my right wrist bends the tip of the rod backwards, "tightening" the hook into the fish's mouth. Just fast enough and just firmly enough. It is easy to break off a leader.

The fish responds by diving straight down, bending the rod sharply. I release line, letting it slip through my left hand until the rod tip has only a slight bend off the vertical. This is no chub. And small trout do not bend the rod down. This is a fish that has to be played skillfully, the fish that all the brainy designers for more than 500 years were leading up to. All the special tolderances, nuances and subtleties tested and built into the equipment by thousands of skilled craftsmen were put there to permit me to land this fish. But only if I do it properly and with full respect for the fish.

Too much pressure can break the leader. Too much slack might permit him to shake out the hook. He swims in a slow arc, well down but without taking line. I swivel to follow, the rod vertical with a minimum of bend, only slight pressure. Then he stops, 15 feet downstream and swims straight down, holding near the bottom.

Unnerving. The ploy of a larger rather than a smaller fish. The correct response is not to respond. Too often a fisherman, charged with adrendaline, cannot resist trying to stir the fish into action. Too often have I felt the sudden, dismal parting of leader and fly. Not this time.

He turns and runs downstream; I let line slip through the guides, bending the rod back to the vertical. With the fish 40 feet out and moving fast, the angle formed by line and water begins to lessen. He is rising quickly. I dip the rod to slacken the line. If his head breaks the surface, I will lose the resilience of the water which adds several pounds to the leader's breaking point. I know he will not jump, though he may thrash on top, because his actions of driving down initially and staying well under have established him clearly as a brown trout, not a rainbow or a brookie. Either of them would have surfaced, perhaps jumped, swum faster and by now be about worn out. The brown is a tough canny fighter, carefully husbanding his strength and prepared to outlast an unwary opponent: the fly fisherman's choice.

Now he drives directly toward me and down. This is another unnerving maneuver. The closer he comes the more difficult it is for me to maintain even slight pressure with the rod. And there is always the possibility he might swim between my legs, rendering useless centuries of lore and legend on the handling of trout. He veers off upstream and toward the bank, again forcing me to release line and again he starts to rise about 30 feet out. Then he turns. And turns and turns. With each maneuver I am gaining line on him, stripping it in and collecting it on the water in front of me, trying to avoid letting it get involved with the staff, the net, the nail clippers, the buttons, the zippers and all such stuff hanging from me.

Any distraction now could be disastrous. He still has strength; for every five feet I gain, he takes back three. He circles slowly, staying down and taking advantage of holding water behind boulders. Fortunately there are no underwater logs in this spot.

He stops about 15 feet out, then starts to circle again, now gradually coming up toward the surface. He is wearing down. I must be especially careful now not to try prematurely to bring him to the net, to "horse" him in.

Then he is only 10 feet out and near the top, showing flashes of his yellow-white undersides. Closer. Now all the line is in and 10-foot leader is slipping through the guides of the seven-foot rod. I can make out his deep green back heavily covered with black spots. I transfer the line to my rod hand and move the net with my left toward the fish, gently pressuring him toward me. His head rises slowly toward the surface and I slide the net under him. It is over. He is caught. Tension and doubt drain out of me.

Keeping the fish in the net but upright and underwater, I slide my left hand down the leader and grasp his lower jaw. With my right hand I unhook the fly. I hold the butt end of the rod along the fish to have a means of establishing his length. With him still upright in the water, I draw the net out from under him. He hovers by my right leg for a few seconds, then slowly swims down and disappears.

My marks on the rod butt measure only 14 1/2 inches. But he was thick through the middle, probably weighing close to two pounds. It has taken me 20 minutes to land him. I leave the stream and head for the car. In five hours I have caught one fish. A fish that rose near me. A fish I thought about and planned for, that bulged the water, took the fly and did all the magnificent and frightened things a hooked trout does. Now I can go home and do it all over again. Many times. Thank you, Dame Juliana.