Milling swarms of desperate fish, frantic scientists working in the heat and smoke of a forest fire, waters that would not run cold and waterfalls that could not be jumped—these things combined to produce the exciting climax to an eight-year experiment intended to restore an almost extinct race of salmon to this remote river that for untold centuries counted them in millions
October 22, 1961

As any alert detective-story reader will immediately recognize, the illustration on the opposite page looks like a thumbprint. It contains those whorls, ridges and irregular patterns that, we are told, are never the same for two human thumbs, and thus are the definitive means of identifying human beings. Actually, this seeming thumbprint can do a good deal more: it is one of the scales of an adult sockeye salmon (below), as seen under a microscope, and it not only can identify the fish but can tell where it was spawned and where it is going.

In a startling ichthyological melodrama rivaling any paperback thriller, these scales have just helped establish a point that has kept biologists and fisheries experts on the edge of their laboratory chairs—a point that is momentous to sport and commercial fishing everywhere. On the Horsefly River of British Columbia, where scientists watched and worked and waited through tense days, the scales proved beyond question that salmon can be restored in immense numbers—by the millions, in fact—to barren but once-abundant streams. Identification by scale has made it possible to regulate salmon fishing in a systematic and selective way, and assure that the fish returning to a depleted stream to spawn will get there and rebuild their former populations. A quarter century ago the Horsefly, once one of the great salmon streams of the fabulously rich Fraser River system, numbered its returning salmon barely in the hundreds. But after what happened there last month, the culmination of an experiment hopefully begun eight years ago, fish biologists are prepared to say that runs of salmon and other migrating fish, given proper pollution control and a way of circumventing barriers such as dams, can be preserved and built up again on any river anywhere. As Clarence Pautzke, Commissioner of the U.S. Fish and Wildlife Service, put it succinctly, "No stream can be written off." What it could mean, said one eminent authority, is the return of game fish to all the streams where they were once abundant.

The sockeye is not a game fish, but salmon are so closely related that what applies to one species is generally applicable to all in matters having to do with conservation and control. In the center of each sockeye-salmon scale the rings have grown close together; these are growth rings formed in a salmon's first year in a lake, and they differ for every lake. The Horsefly salmon, for example, after being hatched swim 30-odd miles down the river and into Quesnel Lake. They spend their first year there and acquire a scale pattern different from that of the salmon spawned in the Stuart River, the Chilko, the Raft, or any of the other streams that make British Columbia the greatest salmon-producing region of the continent, if not the globe.

These young salmon from the Horsefly are three-inch fingerlings after a year in Quesnel Lake. They swim the Fraser to its mouth and in the first week of May join millions of salmon from other rivers before heading out to sea. Fisheries scientists trap them there and take samples of their scales. Plastic replicas are made and filed in small loose-leaf notebooks, about 4 inches by 7 inches, each page containing 115 replicas of scales. Thus, when the fish return three years later as adults, they can be identified as surely as though they were tagged by species and home addresses, and their return upstream can be regulated accordingly.

That circumstance and the history of the Horsefly made this fall one of the most dramatic ever known in northwestern fishing. The story began in a narrow valley deep in the heart of the Cariboo country of interior British Columbia during a few days of perfect Indian-summer weather last month. The Horsefly River is a bright, short, shallow, crystal-clear little stream that springs from two branches on the slopes of Mount Perseus, an 8,361-foot cone that the natives call Haycock Mountain. In its extreme upper reaches, the Horsefly flows 30 miles or so through deep wilderness and into a rocky chasm, heavily wooded on steep slopes that rise 800-odd feet above the water. The Horsefly at this point is 30 feet to 50 feet wide, and at the end of a dry summer perhaps four feet deep; it is a noisy, wild, turbulent stream with beautiful coiling rapids that seem to wind and unwind as the river is thrown from one canyon will to the other. As it emerges from its gorge it drops over a series of falls. Below the falls the river widens to 100 feet, slows down, warms up and flows with unhurried speed over miles of tranquil little rapids and riffles.

Centuries ago the sockeye or red salmon, Oncorhynchus nerka, bright-red fish weighing on the average six pounds and measuring 24 inches in length, selected this particular stretch of the river for its spawning beds. Every four years, 25 times each century, from time beyond reckoning, these salmon came from the North Pacific, forced their way up the Fraser River (swimming from 17 to 33 miles a day), passed hundreds of thousands of places that you or I might think would be fine for spawning, and unerringly reached this one little section of the Horsefly. If anything prevented their reaching it, they died without spawning.

Salmon live 12 days on their spawning grounds. Five days are spent preparing to spawn. The fish pair and select the right kind of riffle, with the precise kind of gravel and the right stream flow. Another five days are spent preparing the redd, the nest in the gravel. The female scoops this out, lying on her side and flapping her tail, the current of the river carrying the sand away. Two days are spent spawning. The female hovers, suspended over the sand of the excavation she has dug, the male pressing against her, and as she deposits some of the eggs she carries, the male at the same instant fertilizes them with his seminal fluid. The female now digs another nest upstream, the sand and gravel from this covering the eggs she has just laid, and so on with several hundred eggs in each nest until 3,000 to 5,000 have been laid. Both male and female die after spawning, reduced to only one-third what they weighed when they entered fresh water. The eggs develop under the gravel, where they are safe from predators, and the fry work to the surface of the gravel when the water warms in the spring. Then they are swept downstream by the current to quiet water, collect in schools, spend their first year in lake water and as three-inch fingerlings eventually disappear into the Pacific.

Other salmon runs, like the Adams River run, are larger than that of the Horsefly. But the Horsefly, or the Quesnel-Horsefly, to give it the name the scientists use, was once one of the major runs of the whole immense Fraser River system.

The first salmon run that brought the Horsefly to the attention of the outside world was that of 1857. So many salmon churned up so much gravel in their spawning operation that year that they uncovered gold. Five men exploring up the Horsefly found free gold in the gravel near the lower limits of the spawning grounds; they picked up 100 ounces of gold nuggets in a week and started the great gold rush into the interior. More gold was found in other rivers than on the Horsefly, however, and the river, already notorious for the large numbers of big, black horseflies that gave it its name, gained further ill repute for not containing more gold.

The Horsefly next came in for considerable public attention in 1888. It occurred to some forgotten promoter that mining, which was now widespread not only in the wilderness around the Horsefly but in the Quesnel River below Quesnel Lake, would be easier if the flow of the river could be entirely stopped and the gold picked up at the miner's convenience from the dry gravel that would thus be left. The Golden River Quesnel Company Limited accordingly began a dam, 763 feet long and 18 feet high, at the outlet of Quesnel Lake. Fortunately for the salmon—if not for investors—the work proceeded so slowly that the runs of 1889, 1893 and 1897 went on their way up the Quesnel, past the dam to the Horsefly without being impeded in the slightest by the structure. But in 1898 the dam was finished. The flow of the Quesnel was completely blocked. The next great salmon run was due in 1901 and in 1900 the mining operation was abandoned.

The dam was left standing, the gates still closed, in the hope that more investors would rush in and place their savings in the project. Meanwhile, the 1901 salmon run turned out to be the biggest in human memory. At least 11 million salmon on their way up the river were caught for the commercial canneries, and unknown millions more still swarmed below the dam, trying to get to the Horsefly. A small flume, no more than a foot wide according to some accounts, had been left for the run to go through; it was soon choked, and almost the entire run died below the dam without spawning. But a few fish did get through—even nature could respect odds of 1,000 to 1—and these kept the cycle alive.

Public sentiment meanwhile was changing from indifference to a concern for the survival of the salmon. In 1903 John Pease Babcock, an American who became British Columbia's first commissioner of fisheries, built a small fish ladder over the white elephant of the Golden River Quesnel Company. It cost $4,104. When the salmon arrived in 1905—the small run, offspring of the few survivors of the 1901 catastrophe—they easily got to the Horsefly. The natural reproductive capacity of even a few salmon is tremendous, and as there was no obstruction to the next run, some 4 million salmon came to the Horsefly in 1909. Eight years after the cycle was nearly destroyed, it was back to its prehistoric abundance.

The next run, that of 1913, was the largest ever recorded, with 35 million fish returning from the Pacific, 25 million caught by fishermen, and 10 million starting up the Fraser to spawn. But they encountered a new and ghastly hazard. One hundred and thirty miles from salt water the Fraser narrows to a defile called Hell's Gate, a gorge 110 feet across at the bottom of an 800-foot canyon, with the water 100 feet deep during dry seasons, as much as 200 feet deep during floods, and with a current of 25 feet per second. Salmon could get through by making short rushes from relatively quieter pools formed by rocks and crevices on both sides. But in 1913 the Canadian Northern Pacific Railway and the Grand Trunk Pacific Railway were racing each other laying track on opposite sides of the Fraser. Their blasting started slides that smoothed the banks of the Fraser to a sort of trough, creating hydraulic conditions that the fish could not overcome: there were no longer crevices and obstructions in which they could lodge before dashing a few feet against the rushing stream.

All sorts of improvised measures were attempted to get the fish through—a legend of the fisheries is that Babcock personally hurried up and down the gorge with buckets, carrying a few fish above the slides—but they failed. For 100 miles below Hell's Gate the banks of the Fraser were lined with rotting salmon. Some races of salmon were virtually destroyed, and among them was the Horsefly. By 1941, after a quarter century, there were only about 1,000 on the Horsefly spawning grounds.

About the time the Horsefly run was believed to be destroyed, the method of identifying salmon by their scales (discovered by Dr. Charles Gilbert of Stanford University) gave rise to some hope for its restoration. The formation of the International Pacific Salmon Fisheries Commission in 1937 provided the machinery to regulate commercial catches so that depleted races like the Horsefly could be permitted to escape and spawn, thus opening the possibility that, little by little, the run might be built up again.

And to bring back this run, to build up runs of hundreds to the abundance of millions, was the stupendous task that scientists and fishing men set themselves in the years that followed. "Imagine a race of fish destroyed not once, but twice," said Clarence Pautzke recently, "and still making a comeback." This is the consideration that made the fate of the salmon on the Horsefly so consequential. If the Horsefly could be brought back to produce the millions it once produced, it could well become, in the words of Dr. William Royce of the University of Washington, a model "that may be followed with any species on any river." Hence the tension of fish biologists and fishermen on the Horsefly last month, when it came down to a matter of days to determine whether the salmon could spawn and whether they would ever return.

The opening scenes of this momentous conservation drama lived up to expectations. Early last summer the salmon collected off the coast and began their eerie progress down the west shore of Vancouver Island, swimming as if directed by compass. Commercial fishermen caught them in the ocean, and samples of the scales were hurried to the microscopes to determine which fish they were and where they were bound. During their ocean years, their scales had grown larger, growth rings like the rings of a tree being added at the rate of seven a year, but only the tightly woven rings in the center of a scale, comparable to the growth rings in the heart of a tree, were used to identify the salmon, for these were the rings formed in a lake, and the patterns differed for each lake. By means of that small fresh-water nucleus on the scales they were identified—these from the Stuart, from Bivouac Creek, Driftwood River, Forfar Creek, Felix Creek, these from the Nechako River, from the Endako or the Nadina, these from the Thompson, from Seymour River and Scotch Creek, from Big Silver Creek on the Harrison or Birkenhead River on the Lillooet, along with salmon from the Horsefly and Quesnel Lake.

The identification process was complex enough, but the mechanics of control that followed were excruciatingly complicated. The proportionate number offish indicates their relative abundance, and the ideal arrangement would be to close the fishing grounds periodically so that the best and strongest fish at the peak of each run could escape to their home stream. The mouth of the Fraser is like a funnel toward which the fish move, one run after another at fairly well-defined intervals; the fishermen, with some 2,000 fishing boats, are ranged in the waters between the U.S. and Canada. To stop all fishing would be uneconomical, because some streams produce all the fish they have room for. If all their population returned, the fish would overcrowd the spawning ground and many would have to be destroyed to prevent them from destroying the beds already spawned. So the regulations specify such things as that the High Contracting Parties—meaning the U.S. and Canada—are agreed that taking sockeye salmon shall be prohibited, say from 5 o'clock in the afternoon of Wednesday to 5 o'clock in the forenoon of Monday following, between Angeles Point in Washington and across Race Rocks to William Head in British Columbia; in other words, free passage is permitted there during these hours for some race of salmon to make it to the mouth of the Fraser and so get to its home stream.

The fish identified from their scales as coming from the Horsefly were in superb condition. Loyd Royal, director of the international commission, said he had never seen fish in better condition. "No parasites," he said admiringly, "and no sea lice." These sterling representatives of their species swam smoothly through the Strait of Juan de Fuca during the periods of no-fishing set by the commission, unmindful of the yearning glances that fishermen cast in their direction. They were practically taken by the arm and escorted to the Fraser. Outside the mouth of that great river they congregated, awaiting a flood tide before entering fresh water. Oddly enough, they would only enter the grayish-green water of the Fraser with the tide, though thereafter they would be swimming against the current throughout the few days left to them.

No part of the drama of the wilderness ever runs according to man's scenario; that is an essential part of its endless appeal. The salmon swam about 500 miles up the Fraser at the unprecedented speed of 40 miles a day, and came to the Horsefly a little early, into warm, dry weather, with the river lower than anyone could remember. Salmon spawn when the water is between 55° and 45° with most spawning at a water temperature of 50° When they reached the Horsefly the water was 66°. Almost the entire early run, perhaps 130,000 salmon, died without spawning. When the main run arrived about the first of September dead fish were everywhere along the banks, in the bottoms of the deep pools or lodged in riffles or in the limbs of trees that had fallen in the river.

Usually fish at the spawning grounds are active, darting in great lines, a hundred fish across, in formations so orderly they seem to have been drilled. Or they may race with what appears to be exuberance through the shallows, hundreds of them in a pack, sending up spray five feet in the air, and pushing before them a wall of water so high that they may be washed several feet up the bank. But these fish were torpid, turning and moving ceaselessly but languidly. Occasionally a female would begin halfheartedly to prepare a redd, then give it up.

In desperation, the fish now began to try to reach cold water above the normal spawning grounds by leaping the falls. Above the 20-foot falls was a 50-foot falls, and above that the Horsefly careened through a rocky flume in turbulent and chaotic disorder. But the salmon did not know that; they moved onto the falls and threw themselves into the air.

The canyon walls are steep at this point, rising straight up for 50 feet. The fish could be seen 100 feet downstream, moving slowly into the pool, and then they could be seen again, emerging with terrific velocity to pose for a flash against the falls before they vanished. Biologists say that these fish can leap only 5½ to six feet. But at the scene the strongest impression was of the differences in the height of their jumps. Every few minutes some particularly gifted specimen took off in a great soaring lunge into the atmosphere, his body flailing powerfully as he left the water and settling into an upward glide twice as high as the majority of his leaping companions, seeming to watchers at the base of the pool to reach at least eight feet.

By the end of their first day at the falls, they were leaping into its waters at the rate of 65 a minute. They fell back stunned, drifted downstream, and came back to leap again. Early in the morning of the next day—these salmon do not travel at night—they were jumping at the rate of 150 a minute. There was no visible pattern in their movements. For several seconds there would be no fish in sight, then a dozen at once, crisscrossing each other, or even colliding in the air. A big gray boulder the size of a freight car divides the falls, and one salmon in 10 struck the boulder. Its top half was dry and hot in the sunlight, and the bottom half drenched with spray from the falls and the water left by the fish striking sideways against it. They hit with a sound like the crack of a .22 rifle, clearly audible above the throb and roar of the falls. Occasionally a salmon missed the falls entirely, sailing at right angles with it, hit the rock and remained partially lodged on a tiny bench high above the water for several seconds—plainly outlined, big, misplaced, eerie. And underneath, the fish were leaping tirelessly, a dozen at a time, all day long and day after day.

They were still arriving; some 303,000 were at the spawning grounds. So many fish create a hypnotic condition; it becomes as difficult to see them as it is to watch a fluid. They moved slowly up the rapids to the pool at the base of the falls, in clusters close to the banks, with 10 to 30 in each shallow pool as they entered and left. Their bright vermilion bodies, rose-colored under the rushing water, seemed to have the texture of rich, wrinkled Chinese silk. They poised lightly in the current, only a few inches below the surface. It was possible to stand in the scoured gravel within a foot of them. From time to time one swung out into the current, braced against it for a few moments, and returned or dropped back to a lower pool, or swam to one higher. Or a pair of salmon burst from a pool lower down, sending up a fine sheen of spray as they rocketed against the rapids.

Except for the sound of the falls the wilderness was quiet. The occasional riflelike sound of a salmon hitting the rock was loud. There were no birds. Tracks of bear were everywhere, but the bears were gorged and had vanished, and there were no eagles, though these birds are said to be fond of the eyes of dead salmon. The hot, late-summer sun fell heavily on the motionless air, into a world that seemed drained and emptied of all life except that of the salmon moving steadily against the current to the highest point they could reach.

The temperature of the water was 59°. There was still a margin of safety; if it continued to drop, enough time remained so that most of the run could spawn. But now another unexpected crisis was added to this wilderness melodrama: the immense forest at the headwaters of the Horsefly burst into flame. There had been fires burning elsewhere in British Columbia, supposedly started by lightning, but this one was the great showpiece of them all. Frank Jones, who was sitting on the porch of his farmhouse on the riverbank when it started, said he had seen nothing like it during his 48 years on the Horsefly. "At one o'clock I saw it start at the base of Haycock Mountain," Jones said, "and by 1:45 it was two-thirds of the way to the top." By 3 that afternoon the smoke was so thick that the trees across the river were barely visible in the dense brown haze.

The roads—or the road, for there was really only one—became a thick coil of standing dust from the trucks loaded with bulldozers headed toward the mountain. Soon 30,000 acres were burning. The flames enclosed the upper branches of the Horsefly, from which the colder water had been flowing. By the time the fire was a quarter of a mile from the falls, the salmon had reacted to the warming water; they stopped trying to jump. In the fisheries camp the atmosphere resembled that of an army that has suffered its final defeat. Because of smoke the planes could not come down on the nearby lakes. The distinguished visitors invited to watch this climactic phase of the experiment could not arrive. The bright-colored rubber boats drawn up on the banks looked grotesque A reporter tactlessly asked about the outlook for the next run, four years in the future. Loyd Royal winced visibly and said he could make no predictions of any kind. After a silence he seemed to feel that this answer was inadequate, and added stoically, "The 1965 run is impaired. It may be so badly impaired that there will be no fishing...." The elementary fact was that unless the temperature dropped in the next four days there would be no 1965 run and the salmon cycle on the river might be forever ended.

A story went around that the salmon were heading downstream. Fishermen and scientists walked down the river a mile or so where there was a deeper pool of quiet water in which the fish could be clearly seen. There was a grassy bank at a bend of the Horsefly, opposite a white-gravel stream bed where, in higher water, the river created another channel around a little island. A ribbon of dead fish, five feet to 10 feet across and 100 feet long, was piled up on the bar, and the men of the fisheries crew with long poles were still lifting the dead fish from the water and tossing them into piles, in a methodical and mechanical effort that seemed in the smoke and haze to have lost all purpose and meaning. There were about 110 salmon in the pool directly below the knoll, and most of these were headed downstream, a startling testimony of the increasing warmth of the water that was now coming through the burning woods, and startling too because they had been seen since they entered fresh water only swimming against the current. But they did not swim downstream far; they turned and swam back, ceaselessly, and headed downstream again in a dazed and confused manner.

Late that afternoon, as the watchers were driving back to camp, the first drops of rain fell. There was an odd, unfamiliar patter, and enough water to warrant starting up windshield wipers. The dust ceased to explode under the wheels. Then the rain stopped. But that night the watchers awoke with a reaction from a sensation so unexpected that for a time they did not know what it was: it was cold. Outside the sky was light with gray, luminous rain clouds. Touching the needles of a pine tree could send shivers up the spine: there were drops of water on them. The wind began to blow hard, driving the fire back over the land it had burned. Then the rains began in earnest. The salmon spawned in such numbers that the whole surface was riffled with them.

But the margin of time had come down to hours, and the narrow escape pointed up the old paradox that nature can always frustrate man's most careful calculation based upon his closest observation of her ways. In the deepest sense, of course, there would have been no real defeat even had the whole run been lost. The fact that it had been restored from almost nothing in 1941 to nearly half a million fish in 20 years was in itself an epic achievement. But now what did this triumph ant conclusion really mean? Clarence Pautzke summed it up: "It has a terrific impact on today's thoughts with regard to bringing back runs of fish.... It points the way to the restoration of all species."

TWO ILLUSTRATIONSJACK KUNZ PHOTOIn a spectacular leap, salmon at right outjumps other fish at spawning grounds in effort to clear falls of Horsefly and reach cold headwaters. MAPSalmon arrowed through Juan de Fuca Strait, swam unerringly to spawning grounds at phenomenal speed of 40 miles a day.
Quesnel River
Horsefly River
Mount Perseus
Fraser River
Thompson River
PHOTOForest fire added final hazard to ordeal of salmon when heated water threatened to destroy run.

Eagle (-2)
Birdie (-1)
Bogey (+1)
Double Bogey (+2)