Search just our sites by using our customised site search engine


Click here to get a Printer Friendly PageSmiley

Click here to learn more about MyHeritage and get free genealogy resources

The Glittering Mountains of Canada
Chapter XIV. In the Shadow of Mount Robson


“No game was ever worth a rap For a rational man to play,
Into which no accident, no mishap Could possibly find a way ”

Everest Expedition, 1924

It was during my first season of real mountaineering in the Northwest that I saw Mount Robson. Our climbing had been in widely separated regions that summer—in the Rockies, along the Canadian Pacific; in the Selkirk Range. We had tramped through the Cascades, voyaged up the coast in a tiny steamer to the islands of southeastern Alaska, and had camped by the glaciers that come down to Taku and Atlin. It was late in the fall when we started eastward from Prince Rupert, on the Grand Trunk road; there was a delightful crispness in the clear September air, sharpening the outlines of jagged little peaks of the Coast Range that lifted above the dim violet of distant, forested slopes, and the nearer brilliance of yellowed birch and poplar banking the Skeena River.

And then, for a day, our train had followed the canyon of the Fraser, past the hilly mining country near Prince George; and, at evening, we were on the bend near Tete Jaune, with the glorious massif of Robson towering afar. Clouds were gathering, and the vision was soon gone; but we had one splendid moment when the mountain’s crest was golden, above a mist-wreathed base.

Of course I had read about the mountain before. It was first described by Lord Milton and Dr. Cheadle, early travellers through the North Thompson Valley, in their book, The North-West Passage by Land. “On every side,” we are told,1 “the snowy heads of mighty hills crowded round, whilst, immediately behind us, a giant among giants, and immeasurably supreme, rose Robsons’ peak. This magnificent mountain is of conical form, glacier-clothed, and rugged. When we first caught sight of it, a shroud of mist partially enveloped the summit, but this presently rolled away, and we saw its upper portion dimmed by a necklace of light feathery clouds, beyond which its pointed apex of ice, glittering in the morning sun, shot up far into the blue heaven above, to a height of probably 10,000 or 15,000 feet. It was a glorious sight, and one which the Shuswaps of The Cache assured us had rarely been seen by human eyes, the summit being generally hidden by clouds.”

The very origin of the mountain’s name is lost in the past; but the Indians had their own names for it, long before the arrival of white men.

Mr. H. J. Moberly, factor of the Hudson’s Bay Company, gives the following information in regard to the name of the mountain: “Years before the Hudson’s Bay Co. and the Nor’-West Co. joined (1821), it was the custom for the Nor’-West Co. to outfit a party for a two years’ trip, hunting and trading. They went west and north, even as far as the border of California. One party, under the charge of Peter S. Ogden, some two hundred men, chiefly Iroquois and French Canadians. When west of the Rockies, he scattered his hunters in different parties under the charge of a foreman, to hunt for the season. One of his camps, under the charge of a man named Robson, was somewhere in the vicinity of this mountain, and it was the rallying point where all other parties came together for their return east.” wyn,1 Director of the Geological Survey of Canada, as early as 1871, reported that the Indians told him their name for the mountain signified “The lines in the rocks.” As Dr. Dawson, in a paper presented before the Royal Society of Canada in 1891, informs us,2 “The Kamloops Indians affirm, that the very highest mountain they know is on the north side of the valley at the Tete Jaune Cache, about ten miles from the valley. This is named Y uh-hai-has-kun, from the appearance of a spiral road running up it. No one has ever been known to reach the top, though a former chief of Tsuk-tsuk-kwalk, on the North Thompson was near the top once while hunting goat. When he realized how high he was he became frightened and returned.” The Cree Indians call Robson simply “The Big Mountain,” but this seems to be a modernism; old men, with whom I have talked, say that their tribe never had a special name for the peak.

Mount Robson is the highest summit in the Rockies of Canada; but, like many a lesser peak, its height has diminished with recent measurements. The first triangulation, that of McEvoy, resulted in a figure of 13,700 feet; but the more recent determination of the Interprovincial Boundary Commission has brought this down to 12,972 feet. Thus an old illusion is shattered, and no peak of the Rockies of Canada attains thirteen thousand feet.

Dark was the night when the train pulled through Yellowhead Pass and down to Jasper. This was in the days before hotels, and one obtained lodging in rooms above the grocery-store; or, if the vacation was an extended one, in the “tent city” on Lake Beauvert, where the attractive Lodge of the Canadian National is now situated. The Grand Trunk was then running a through train to Prince Rupert only on alternate days; so in order to go back next day for a better view of Mount Robson—as I had decided to do—it was necessary to get permission to ride on the freight. There was less trouble about it at that time than there is now, and I know of no more delightful way of seeing the mountains—I refer of course to the railway zone— than from the caboose of a slow-moving “side-door pullman.” If the engineer happened to be very good-natured, you rode up front on the cow-catcher; with every opportunity for photographs, and, perhaps, if there was a delay in waiting for an express to pass, the chance to go fishing with the train-crew.

Roadbeds of the mountain area, recently constructed, were not yet equal to the rolling-stock which passed over them. We started off early in the morning, up the valley of the sun-flecked Miette, and on through Yellowhead Pass. Standing on the rear platform, looking across at the heights of Mount Fitz-william,5 we were startled by a crashing and grinding.

Jumping off, we saw that a spread rail and a buckled truck had wrecked the whole train ahead, and that several huge loads of groceries had spilled down the embankment of the Fraser River, with more than one car slanted out on a precarious angle and threatening to follow.

So the engineer wired for a wrecking crew, and the trainmen got out their fishing-tackle. We were only half way to Robson, and I walked back along the tracks to Lucerne to await another train that was due on the Canadian Northern road. I have never regretted this pleasant stroll; there was no reason for hurrying—a small boy informed me that my train would be four hours late—and the time was passed on the lake shore, where ducks were swimming and the placid water mirrored the peaks of Yellowhead.

It was again dark when we started westward. In a few minutes, across the Fraser, we could see the fires lit by the train crew to assist them in clearing the wreck of our freight. The night was calm and clear, and a red, waning moon rose over the hilltops as we neared Moose Lake and the Rainbow Range. I had no very accurate idea as to the exact location of Mount Robson, or just how one would set about getting near to it. We had seen it from Tete Jaune, and realized that the mountain lay north of the Fraser; but of the trails through the intervening valleys and forests we knew nothing. Still I knew that I could again see it from the railroad, and under better weather conditions; I thought it might be possible to find a way across the river. I was quite determined to go as close as possible to that loftiest of mountains.

It is a strange and interesting little adventure to look back upon. I knew practically nothing of the ways of the trail; I had no equipment save a camera and a rucksack filled with provisions; it was late in the month of September. The train slowed down to let me off by the box-car filled with hay, which then served as Robson Station. It was midnight; I was alone, and a cold wind blowing from the west and the rising moon were my sole companions when the train had passed around a bend and out of sight. I could not see that much was to be gained in waiting for morning to come; it was almost impossible to restrain my eagerness for a sight of the mountain. And so I started off along the tracks in the semi-darkness. Youth does such things.

And suddenly the vision appeared! Mist and vapour play strange tricks with one’s judgment of size and distance—moonlight affects it to perhaps an even greater degree. I had walked several hundred yards along the tracks, and then I stopped, scarcely believing what I saw. There it was; Robson, “the mountain of the spiral road,” seeming to touch the very heavens, flooded with soft light and gleaming like molten silver. The light of the moon seemed to become tangible, as if one were swimming in a luminescent haze that altered and exaggerated refraction. I had seen higher peaks before, and have since stood on the summits of many, but nothing has ever equalled the impression of stupendous height that Robson gave on that starlit night of years ago. It seemed to me then as if I were gazing up to the throne of some Divinity; although maturer years have somewhat tempered this idea to the thought that there is merely a feminine quality in some mountains which makes them best seen at night.

But certainly that did not enter my head at the time. I sat down on a trestle and dangled my legs over the side, and looked and looked. But contemplation of the sublime cannot maintain the body at an even temperature on a freezing night. I got up and moved slowly along, rubbing my eyes and still half afraid they were deceiving me. And then a piece of luck: a bobbing twinkling light along the rails ahead. It was carried by Bert Wilkins, an outfitter then working for Donald Phillips, who had come out looking for me. I had met him for a moment when the train came through a few days before and had told him that I would try to come back. He still had some horses down in the Grand Forks and in a few minutes we had arranged for a pack-train trip to Robson Pass. I was glad at the thought of having company who knew the way.

The keeper of a section-house put us up for the night, and next morning we went down for the horses. There was then no bridge over the Fraser; but a logjam, long since departed, afforded a place of crossing. We rounded up the cayuses, and finished a tremendous breakfast of bear meat, potatoes and steaming coffee. On the preceding day Bert had shot a large black bear on a nearby berry-slide, and the hide was now nailed up on the door of the shack. A “homesteader,” whose section is in the angle of the Grand Forks, came over and joined us as cook. We were off shortly after seven o’clock, three riders and two pack-horses, not long after the first sunbeams reached us across the high hills bordering the Fraser. One crosses the old freight road and, entering the woods, leaves civilization behind.

The trail is unforgetable in its beauty, with spruce and cedar trees straight and perfect above a carpet of berries, fern-brakes, and devil’s club; tropically luxuriant. The stream descends in cascades and rapids, with the southern cliffs of Robson almost above one’s head. Far behind, in the direction of Tete Jaune, rises the multicolored ridge of Mica Mountain in the Cariboos. We are soon in Robson’s shadow and the top is no longer in sight. Turning a corner we come out on the shore of Kinney Lake, with the slopes of Little Grizzly and the pinnacle of Whitehorn far above. Just now there is not a cloud to relieve the deep blue of the sky, nor a ripple on the lake to disturb the images of tall trees and soaring peaks. Some day there will be a hotel here and perhaps a funicular; how much the worse!

Rounding the northern shore of the lake, through the trees along the water’s edge, we cross the expansive delta of glacial silt at its western end and take up the trail again in the Valley of a Thousand Falls. There is a beautiful glacier and a rock spire at the valley head, and if not quite a thousand falls come streaming down from the cliffs on either side, the number is at all events most satisfactory and surpassed only by the beauty of their unbroken height. We cross through rushing streams, and slowly climb up the thousand feet or more of zigzag trail, cleverly engineered with wooden trestles, to the upper levels where the roar of Emperor Falls is heard—dissonant to our vocal efforts in urging the horses along.

Across the deep valley-trench, Whitehorn is magnificent with its icy arete and hanging glaciers, above a black precipice streaked with threadlike, silvery waterfalls. Beyond the misty rainbows formed in the cauldron of Emperor Falls, above tier on tier of horizontal strata and cliff-belts, rises Mount Robson, steep and snowless, into an enormous wedge. Skirting a burned-over, level area, and emerging from the woods, we reach the marshy flats at the western end of Berg Lake, with distant views to Robson Pass. The north side of Robson is sheer, but snow again appears. The little basin at the foot of the snowy Helmet gives rise to the five thousand feet of icefall known as the Blue, or Tumbling Glacier. There are few places in the world where lake scenery can equal this prospect; as we ride along, bits of the ice-front break off with a crash and the fragments add to the number of floating bergs already sparkling in the dark-blue water.

Just east of Tumbling Glacier is the rocky promontory of Rearguard, with the tongue of the Robson Glacier protruding beyond. The eastern end of the lake, Robson Pass, is the end of our day’s ride. Stiff and tired, we slip down from our sweating ponies, fatigue somewhat lessened by thoughts of a square meal soon to be ours. The outfit was very sporty and had packed along a portable stove, on which bear steak and potatoes were soon sizzling and sending up delectable odours that distracted one’s attention from the business of putting up tents. It was bitterly cold when the sun went down, but we moved close to the fire, while Bert regaled us with tales of sheep-hunting in the north country.

When I crawled out of my blankets in the morning there was a film of ice on the nearby brook. The northern face of Robson was clear and tinged with a rosy pink; and promising noises from the cook-fly suggested a successful day ahead. Breakfast over, we set out to explore the lower reaches of Robson Glacier. Clambering over the morainal debris and up the tongue to the glacial surface, we made our way over the parallel ridges of ice which run in the longitudinal direction of the glacier. The tongue splits on Robson Pass, a Pacific-Arctic watershed, some of the water flowing to Berg Lake and reaching the Fraser, while another brook runs northward to Lake Adolphus and Smoky River—a Mackenzie headwater. The glacier originates in the high saddle and extensive neve fields between Robson and Mount Resplendent, an extensive area of serac and crevasse running eastward to the base of a curious buttress, resembling a candle-snuffer and known as the Extinguisher, whence the level and unbroken glacier runs northward for more than three miles to the pass. To the west, just seen along the cliffs of Rearguard, Whitehorn lifts to a needle-point; northward, the view is closed by Mount Mumm and peaks near Moose Pass that border the valley of the Smoky.

During the afternoon we sauntered over to Lake Adolphus and took a nap on the mossy banks. There were superb views of the dazzling cone of Resplendent, and the glacier; but Robson itself was hidden by the cliffs of Rearguard. Day ended in glorious sunset, with afterglow on the snowy peaks. Even the cook could not draw us into the tent until the last bit of gold and purple colouring had merged into the green and dull grey of nightfall.

We made an early start on the down-trail next morning and were back at the corral in the Grand Forks by eleven o’clock. I spent a couple of hours fishing in the Fraser and landed several small trout for lunch, rather opportune, for by this time there was not much left of the bear steaks. It was the off day for train service, so we decided on a hunting trip in the cabins at the mouth of Grand Forks, and successfully bagged a choice assortment of pack-rats that had grown sleek and fat from their forays on our provisions. On the day following, we packed several fifty-pound sacks of potatoes from the homestead’s garden to the railroad, and had still energy enough to ride horses over to the old freight-road where we shot three coyote that we had noticed in the edge of the grass. It was a tame sport, but the fur was in good condition and the hides were soon tacked up beside the bearskin.

In the evening we rode bareback for a little way along the Fraser trail, driving the other horses ahead toward Jasper.

It was not for a number of years—the Great War had come and gone—that we could come back again to Robson; not until the summer of 1924. In that season we were completing our exploration of the Continental Divide, northward from Lake Louise; we had been to Athabaska Pass and to the peaks of Tonquin and on to Yellowhead. It was a July day, the 17th and cloudy to be exact, when Conrad Kain, Alfred Ostheimer, and I unloaded our packs before a crowd of curious tourists at Robson Station.

This time we had come to climb; there would be more of hard work and less of sentiment than on my first visit. But how changed things were! Cabins had sprung up like mushrooms; there was a broad trail, almost a road, leading to a well-engineered bridge spanning the Fraser canyon; permanent camps on the summit of Robson Pass made it unnecessary to use horses or even carry provisions. It was getting altogether too civilized! We put our packs on our backs and started off, arriving in due course at Kinney Lake. There we met the Oberland guides employed by the Canadian National, Hans Kohler and Alfred Streich. They had been of the party that came looking for us on Mount Hooker, and were now making themselves acquainted with the Robson district prior to the camp of the Alpine Club of Canada.

We had a pleasant walk together, next morning, when we all went up to the cabins at Robson Pass. We had not yet seen the top of Mount Robson, weather seemed to be getting worse instead of better, and we were quite ready to believe the Indians’ statement that it was, after all, rarely beheld by human eyes. Fog hung in the valley, blowing in from the Fraser; then, in a change of wind, coming back again from the Smoky. The mountain rises so much above its immediate surroundings—scarcely a peak nearby approaches within two thousand feet of its elevation—that, by its very isolation, it becomes a storm centre. I thought longingly of the cloudless September days in another year.

Enforced inactivity was making us jumpy; so, on July 20th, although it was cloudy and a high wind blowing, we all decided that something must be done. It occurred to us to try Resplendent (11,240 feet), and go up as far as we could. So we started out at halfpast six, and made our way to the glacier, wandering up to the seracs and through a portion of them, killing time in the hope that the wind would die down. It seemed amusing to try the rocks of the north arete, the crest of which had been but twice followed throughout. We roped below a little schrund, as well-guided a party as has ever tackled a Canadian mountain. Ostheimer, with Kohler and Streich, made one rope; while Conrad and I followed behind, showing wisdom therein, as we could use them for a wind-break while they cut the steps. Streich had quite a job of it; it was terribly cold, and the slopes below the rocks were steep and hard. However, in an hour we were in the lee of a rocky pinnacle and enjoying a second breakfast of bread and sardines. Conrad and I then went ahead and found some quite delightful climbing in a short stretch of chimneys and slanting slabs, where handholds were few and body-friction alone kept one from swinging sideways on the rope. At one o’clock we were on the upper snow-level below the peak; everything was enveloped in swirls of mist, but the wind had lessened in force and we could occasionally see for a short distance ahead. Resplendent is not an easy mountain on which to lose the way, and though there was no view to be had, Conrad led us through the fog to the steep-corniced summit in another ninety minutes. There was still enough of a gale so that the last portion had to be done carefully; Streich cut up to the cornice, while the rest of us crouched down in the driving snow and anchored. Each of us had a look over the edge, and then we beat a retreat to the western snow-col at the head of Robson Glacier.

The fog-level had risen to 10,000 feet, and under the edge of its grey blanket we looked far out across the sunlit Fraser Valley to the borders of the Cariboos and peaks beyond. Still no sign of a real break in the weather. As we descended with long glissades into the main basin, the Bess Group was visible in the north. There was just one momentary glimpse of Robson’s peak, rising like a sword-point, vanishing again in the veil of billowing cloud. We were back in camp, with good appetites, just twelve hours after our start, feeling that we had accomplished something even under unfavorable conditions.

Camp was being put up for the annual activities of the Alpine Club of Canada, and the first hikers arrived next afternoon. On the 22d, although the clouds hung low, a large party on three ropes ascended Lynx Mountain (10,471 feet), an attractive peak of the Robson cirque, commanding a widespread view of the Robson and Coleman Glaciers and of Resplendent Valley. One ascends snow-slopes, with a few steps to be cut, nearly to the southwestern saddle whence a broad highway of rising shale leads to the summit. It took six hours to reach the highest point, and but half that time for returning, with many a boisterous glissade in the softening snow carrying us almost to the glacier. Resplendent and Robson appeared several times; never quite clear, but moist and shadowy in the mist. Down Resplendent Valley we could see the slender needle known as the “Finger of Kain,” and far to the south, momentarily of course, we thought we could discern the jagged outlines of the Rampart Group.

Professor Coleman may be considered the first to approach Mount Robson with the idea of climbing it. As early as 1907, he and his companions had come over the Saskatchewan and Athabaska trails from Laggan, and had reached the head of Grand Forks Valley. They returned in the year following, going in from Edmonton and gaining Robson Pass by way of Moose River and the Smoky. After several attempts in bad weather, a final climb from the glacier broke down at 11,000 feet.

In August, 1909, Rev. G. B. Kinney and Donald Phillips attained the summit crest by way of the northwest arete and western face. It was a fine, sporting effort and deserves the credit of a first-ascent. Later in the summer a party of distinguished British mountaineers—Messrs. Amery, Hastings and Mumm, under the guidance of Moritz Inderbinen—had a further try at the eastern face, but desisted after a narrow escape from an avalanche. The very highest point was not attained until A. H. MacCarthy and W. W. Foster, with Conrad Kain, reached it during the summer of 1913. Their route was also by the dangerous eastern slope, but descent was made in a southwesterly direction, with a night out, to Kinney Lake.

Writing for the Alpine Journal, Conrad said of the southwestern ridge, “There is no doubt that this ridge will be the future route to ascend to the summit of Mt. Robson. But the climb cannot be done from Lake Kinney in one day. It will be necessary to build a hut at the head of the Lake Kinney Valley. The snow conditions on the highest peaks in the Canadian Rockies can never be compared with those in the Alps, as there are more avalanches in the Rockies on account of the dryness of the atmosphere, which leaves the snow powdery and unpacked. And so I may say that Mt. Robson will always be a risky climb, even on the easiest side, on account of avalanches.”

In an article for the Canadian Alpine Journal Conrad was no less emphatic, writing, “In all my mountaineering in various countries, I have climbed only a few mountains that were hemmed in with more difficulties. Mt. Robson is one of the most dangerous expeditions I have made. The dangers consist in snow and ice, stone avalanches, and treacherous weather.” During the ten years since, Conrad no doubt modified this opinion of the mountain; but his view in regard to the length of the climb was unchanged. He had so carefully prospected the southwestern slopes of the mountain that the real climbing difficulties were reduced to a minimum; but the danger from falling ice could never be ignored. And so, although we had come to the very foot of the all-highest, I had very little idea of doing more than look up at it from lesser height. But the wretched weather had most of the time effectually prevented even that much. Under the direction of Phillips, a high camp (at about 6500 feet) had been placed near the last trees in the gully above Kinney Lake; we had seen the white speck of a tent, high on the green point, when we had come by the lake a number of days before. We were rather pleased to be chosen to make an attempt with the first official party from the main camp; as a matter of fact, our vacation time was drawing near a close and the trial must be made now or postponed for a long while.

And so, on July 23d, with Messrs. Geddes, Moffat, and Pollard, Ostheimer and I—all members of the Alpine Club of Canada—packed down to Kinney Lake. Conrad, of course, led the way, adding considerably to our remarks about the unpromising weather. As this would prevent serious climbing on the morrow we decided to spend the night at the lake rather than go up to the higher level.

Mount Robson may be considered as a gigantic wedge, rising—although structurally the lowest point of a syncline—in buttressed heights to the summit icecap, ten-thousand feet above the Grand Forks Valley. On its northern slopes, exposed for only seven thousand feet, it presents a spectacle of snow and ice; but the western and southern slopes, above timber-line, are comparatively bare and rocky. It was this southern aspect that we had beheld from the mountains of the Whirlpool, during our journey to Athabaska Pass, when we marveled at the lonely isolation of the great peak. From that point of view, the precipitous southeastern shoulder had been so foreshortened as to be almost indistinguishable; but now, from the shore of Kinney Lake, the lower cliffs and couloirs, with their lines of horizontal strata, attracted our attention.

Next day it took five hours to mount the steep trail through the woods to the climbing-camp. Conrad had the heaviest pack of all—only slightly smaller than himself—and was forced to “build a fence” of willow-twigs in order to accommodate a pail and several loaves of bread on top. The afternoon was spent in camp-work: chopping wood, carrying water, and in constructing a well in the nearby gully.

Above the cliffs, a little to the north of our tents, we could see rolling clouds which hid the crest of Robson, but which lifted enough to show us the green seracs of a lower icefall, from which two crashing avalanches came down just before we started supper. Sitting on the limb of an ancient, storm-gnarled tree, one felt that it would be quite possible to throw a stone into the grey waters of Kinney Lake, three thousand feet below. The lake was now in shadow, but the sun, breaking through the upper levels, flooded Whitehorn with a luminous red-gold light. To the southwest we could see across the bordering hills of Fraser River, almost to the head of Canoe and North Thompson Rivers, and beyond to the Cariboos, whose winding central glaciers were steeped in lavender, and heliotrope—last pale colours of evening.

During the night the clouds rolled back, and we started out at four o’clock on the finest of clear mornings. In two hours we had climbed over long slopes of shale and scree to the limits of vegetation, in the top of a small cirque near the lower icefalls. These ice-falls, two in number and separated by a narrow partition of cliff, owe their formation to a reconsolidation in the avalanche ice that breaks off from the summit cap. Walled in on one side by the southeastern shoulder it is forced, for the most part, into the couloirs bounding the head of Kinney Lake Valley.

Early in the morning, it was quite safe to cross close below these falls; we were quickly through the short distance, floored with shattered blocks, without a sign of anything giving way above. Then up and up the crest of a long, rocky ridge, where the sun met us, to a flat ledge with a trickle of water that met the requirements of a breakfasting place. On such a level, Conrad told us, a hut should be constructed. It would necessitate the carrying up of fuel, since timber-line is more than a thousand feet below, but it would immensely facilitate the ascent.

We sat there, eating bread and jam, a little below the first snow, and looked across to the shoulder. Above the icefalls, under which we had recently crossed, is a level of hardened snow swept by tracks of immense avalanches that had come from the great furrowed icecap that sparkled above us. The cap itself, from the western crest of the mountain to the top of the southeastern shoulder is guarded by a veritable barrier of ice, some hundred feet high. This is the real danger of the southwestern route: one must work up through this upper fall, not often feasible, or traverse under it, or a portion of it, toward the western arete.

We were soon on the snow; Conrad, Ostheimer and I on one rope, striking up a sharp snow-crest that connects the rock ridge from the southwest with the base of the ice-cap. We stopped to reconnoitre, while the other rope came up. In the ice-cliff there was a choice between a frozen chimney, nearby, blue and steep, which Conrad pronounced hazardous for the leading man, and a lateral traverse on horizontal, snow-covered ledges, below the seracs, to a break that seemed to afford access to higher slopes. The traverse seemed the only course; but it looked nasty. It meant an exposed crossing through the head of the great southwest couloir, so conspicuous from the Grand Forks Valley. It was past the noon hour, the ice was in the full light of a hot sun for the first time in a fortnight; and the summit of the mountain, although less than two thousand feet above us, showed us plainly enough that to go on meant a night bivouac. Not very much more than two weeks before we had had two long nights of shivering in the caves of Mount Hooker, and Ostheimer and I were not keen for an immediate repetition of that mode of existence.

Just then there was an ominous cracking, and Conrad shouted: “It’s coming down,” and we all ducked under the nearest ledge. Fortunately only a few small cakes fell, and these not near us. Still it does not take a very big piece to put one hors de combat, and the business might not yet be done with. I made up my mind that the amateurs on our rope must turn back. Everyone has his own standard about what is to be done under such circumstances—mine is that there is plenty of good mountaineering to be done without knowingly placing oneself in an exposed position, requiring time for its passage, where ice or rock may come down. I sometimes subject this to a very liberal interpretation; but on this day, toward the end of a long and successful season, I was not willing to take a chance on the good behaviour of those ice-pinnacles.

The others felt differently about it and inclined to go on. Conrad said, “Gentlemen, it is risky. I am willing to go on if you wish.” So we decided that Conrad should rope with them and continue, while Ostheimer and I on the rope remaining should descend the ridge below, and continue to the high camp. So we parted, wishing each other the best of luck.

The four were immediately lost to sight behind a hummock of snow, while we descended in the steps cut on the way up. We were near the level of Whitehorn, with a widespread view across the Fraser Valley. It seemed to us worth while to ascend the little rock-point which forms the very apex of a buttress just south of the main couloir. From near Kinney Lake it seems to rise as a sharp spire; but from below one does not see the snow that extends behind and toward the ice-cap. We built a little cairn, and sat down to watch the climber’s progress. All at once there was a grinding crash in the direction of the couloir, and some large blocks of ice came tumbing down. The men were still out of sight, but that shower of pieces must have been uncommonly close to them. They were untouched, however, and a little later we saw them gain the icecap through the break in the seracs. Still later, as we descended, we saw them high up on the snow, half hidden at times by veils of mist.

We had come down nearly to the lower icefalls; we stopped to finish off some sardines and coffee, seating ourselves on a broad ledge that seemed almost to overhang Kinney Lake. Then something made us turn our gaze to the lower ice. There was not a sound, but as we looked the entire front of pinnacles began to move. Slowly the green wall tottered and sank, splintering laterally and sweeping the path through which we had come in the early morning. Then came the crashes. We sat as if petrified until the last echoes died away. Conrad heard the noise, on top of Mount Robson. He told me, afterward, that they spent the night near where we had been sitting and thought that the avalanche had caught us. But no such misfortune overtook us, and before dark we were in the blankets beside the campfire.

Conrad and his successful party came in at four o’clock next morning, all rather tired—a night on the rocks is never restful. I got up, did what I could to help get breakfast ready, and then packed down-trail to Kinney Lake, eventually making Robson Pass in time for a belated lunch. The last pull was a hard one. And so, after more than a thousand miles of trail-riding through the Canadian Alps, with success on many high peaks, in more than one long season, I have not conquered Robson—yet.

It is not fair to the mountain to say that conditions are always such as have been described. Later on, within only a couple of weeks, in 1924, a total of twenty people had reached the summit without mishap. Perhaps after a few days of sunlight the ice conditions were better. But to my mind there is always a potential menace in those westerly-exposed seracs. If enough parties try it, there will at some time be an accident from avalanche. I have not yet forgotten whit Conrad wrote after the first-ascent, in 1913, “I do not know whether my Herren contemplated with a keen alpine eye the dangers to which we were exposed.”

Mountaineering includes a philosophy too optimistic for one ever to dwell on defeat; there is always happiness in having tried with good comrades. Perhaps it were best that I should never attain that height; I might think the less of it. For, to me at least, it would be nothing short of sacrilege to stand on the very summit of the majestic mountain that, when little more than a boy, I went hunting for—and found—in the pale splendour of northern moonlight.

7I have at least made climbing contact with the first three of the 12,000 ft. peaks of the Rockies of Canada. I wonder if some energetic climber will ever have all four peaks—Robson, Columbia, North Twin, Clemenceau—on his list?


Return to Book Index Page

This comment system requires you to be logged in through either a Disqus account or an account you already have with Google, Twitter, Facebook or Yahoo. In the event you don't have an account with any of these companies then you can create an account with Disqus. All comments are moderated so they won't display until the moderator has approved your comment.

comments powered by Disqus