Home A - Z FAQ Online Library Discussion Forum Muir Weather Maps About Search
Online Library: Title Author California Geology History Indians Muir Mountaineering Nature Management

Next: Tuolumne MeadowsContentsPrevious: Devil’s Postpile

François Matthes and the Marks of Time: Yosemite and the High Sierra by François E. Matthes (1962)


Last summer, while roaming over the High Sierra with the Scout Naturalist Expedition, it was my good fortune to become acquainted with a piece of mountain sculpture of a very exceptional sort.1 Though presumably not without parallel in the Sierra Nevada, it is nevertheless of a type that from the very nature of things cannot be represented by more than a few examples. The feature in question is on the top of the mountain known as Shepherd’s Crest, which stands forth prominently on the east side of Virginia Canyon, a mile or more above the McCabe Lakes. To many members of the Sierra Club, doubtless, this mountain is a familiar landmark; for all I know, it has been climbed and explored from end to end; but to me it was new and its summit sculpture a revelation, the more unexpected since the small-scale topographic map, which I had duly scanned in advance, gave scarcely a hint of its unusualness.

Viewed from any low point to the southwest, Shepherd’s Crest appears surmounted by a row of blunt pinnacles, all curved in the same direction, and rising from a sheer wall that is cleft at almost regular intervals. Not having seen the mountain before, one might readily suppose these jagged teeth to constitute the main summit crest; but on viewing it from other directions and from higher vantage-points, one perceives that there is a second crest, higher and smoother, some distance to the north of the first. Between them lies a bit of rolling upland that seems wholly unrelated to the sheer glacier-trimmed sides of the mountain, and, what is most remarkable, this bit of upland consists of a V-shaped valley instead of a convexly moulded summit. From each of the two confining crests the surface slopes inward to an old stream-channel that drains out at the western point of the mountain. This channel is, however, much nearer to the low southern crest than to the high northern crest, which culminates in a summit almost 400 feet above the valley, and so the feature as a whole is strikingly asymmetric.

The accompanying photographs, taken by members of the Scout Naturalist

Shepherd’s Crest from the southwest. By Robert Branstead
[click to enlarge]
Shepherd’s Crest from the southwest. By Robert Branstead

Expedition, will help to make clear the odd configuration of the little valley. There are added a sketch map and a bird’s-eye view designed to bring out its character more completely; these are based merely on the topographic map, the photographs, and my own observations, not on instrumental surveys of any sort. The little upland area, it will be seen, is roughly triangular in outline, and measures three-quarters of a mile in length from northwest to southeast, and one-quarter of a mile in greatest breadth. Its lowest point, at the lip of the valley, is just above the 11,500-foot contour line; the highest point on the northern crest reaches an altitude of 11,860 feet. The floor of Virginia Canyon, near by, is somewhat below 9000 feet.

In the view from Mount Conness (below), Shepherd’s Crest is discernible in the left middle distance, mainly by the gentle slope that leads up to its high northern crest. The little valley itself is not visible; being masked by the pinnacled southern crest, nor is its actual extent apparent, yet its isolated position amidst the titanic environment of craggy peaks and profound canyons is almost dramatically revealed. It seems like a little secluded skyland realm, cut off from the fierce world around it by impregnable cliffs.

Looking north from Mount Conness. In the lower part of the view is the nivated west slope of North Peak. A bit of the upland surface on Shepherd’s Crest is visible in the left middle distance, at the far end of a long arête.  By Richard M. Leonard
[click to enlarge]
Looking north from Mount Conness. In the lower part of the view is the
nivated west slope of North Peak. A bit of the upland surface on Shepherd’s
Crest is visible in the left middle distance, at the far end of a long arête.
By Richard M. Leonard

That this little “lost valley,” as the boys called it, is a lone remnant, a surviving bit of an ancient landscape of moderate relief that once had wide extent, but that has been largely consumed by the incision and widening of the deep newer canyons, readily suggests itself to one who observes it critically. Certainly to a geologist trained in the interpretation of topographic forms the fact is at once manifest from the very contrast between the flowing contours of the little upland valley and the stark sculpture of the canyon walls below. Moreover, in the foreground of the view from Mount Conness one beholds the smooth westerly slope of North Peak, which is in the same general range of altitudes as the valley on Shepherd’s Crest and represents another remnant of the same ancient landscape. On the west it connects with still other smoothly curving remnants on Sheep Peak (not visible in photo on page 111). To the southeast of Mount Conness, again, one looks down upon a gently sloping tableland that exhibits the same subdued style of topography at the same general level. Farther to the southeast is the long flattish top of White Mountain, and beyond that the nearly level Dana Plateau, the largest tabular summit of this type. To the east

Figure 9.—Sketch map of Shepherd’s Crest
[click to enlarge]
Figure 9.—Sketch map of Shepherd’s Crest

and the northeast of Mount Conness, finally, are the smoothly rounded summits of the Tioga Crest, about three miles in aggregate length.

Though these different fragments of the ancient landscape (or erosion surface, as geomorphologists would term it) lie so far apart that the missing portions between them can hardly be reconstructed in imagination, it is possible, nevertheless, to make local restorations and to visualize to some extent the progressive destruction of the old topography by the development of the new. There can be no reasonable doubt, for instance, that the long attentuated arête which ties Shepherd’s Crest to the main divide of the Sierra Nevada was once a massive ridge broad enough to bear a strip of the ancient topography throughout its entire length. By the glacial enlargement of the deep canyons on both sides to capacious cirques it has been gradually reduced in width until now there is left only a thin, sharp knife-ridge, a cleaver,2 as such a feature would be termed in the Mount Rainier country. By the divergence of the two cirque glaciers Shepherd’s Crest and its little upland valley happily were saved from a similar fate, but the broadening of the cirques nevertheless has progressed far enough to destroy in large part the two spurs of the upland topography that originally flanked the little valley. The two crests that now enclose it are not the tops of those ancient spurs—they are merely the sharp edges in which the encroaching cirque walls without cut the gentle slopes of the valley within.

Figure 10.—Bird’s-eye view of the Little Lost Valley on Shepherd’s Crest
[click to enlarge]
Figure 10.—Bird’s-eye view of the Little Lost Valley
on Shepherd’s Crest

But has not the little valley itself been glaciated? you will ask. No, it exhibits none of the characteristic signs of glaciation—that is, of erosion by a moving ice mass shod with rocks. According to the report of Scoutmaster Richard M. Leonard, who with several of the boys climbed up to the little valley by way of the spur that leads to its lip, polished and striated, or even simply smoothed rock-surfaces and rounded ledges, such as are common features of glacier-beds, are wholly absent from it; neither are there any accumulations of rock debris resembling moraines. On the other hand, he found its slopes encumbered throughout with angular blocks, large and small, loosened and heaved by the freezing of water in joints and crevices; and most of these blocks, he observed, lie on or near their places of origin—no forces other than those of frost, snow-pressure, and gravity, apparently, have acted upon them. Such a mantle of frost-riven fragments is a characteristic feature of high mountain slopes that have borne no active glaciers, but only inert drifts or fields of snow. It is the product of that slow and unspectacular rock-shattering process due to oft-repeated alternations of frost and thaw, unaccompanied by any adequate transporting agency, for which some years ago I proposed the term nivation, in contradistinction to glaciation.3

While alternating freezings and thawings occur almost everywhere at high altitudes, the special combination of conditions that results in nivation occurs typically only on high summits and slopes that annually bear snowdrifts for long periods. For both the recurring drifts and the porous rock mantle tend to prevent the melt-water from gathering into vigorous transporting and eroding streams, and instead to distribute it into many feeble rills. Nivated slopes, accordingly, not only are mantled with rock debris that remains in situ (except as it is affected by local creeping movements known as “soil flow”),4 but they are devoid of sharply cut stream-channels as well.

The little valley on Shepherd’s Crest exhibits both of these effects of nivation. Its sides are rock-strewn throughout, and also unfurrowed by stream-worn ravines. Nevertheless, these facts alone cannot be accepted as absolute proof of its non-glaciation, for it is conceivable that the little valley was glaciated at a very early date in the Ice Age—so long ago that the nivation process has since had time to obliterate all traces of ice wear. At least three, and possibly four epoch5 of glaciation have been recognized in the Sierra Nevada, and the earliest of these occurred presumably not less than half a million years ago. Such a span of time might have been long enough to give the little valley a thoroughly nivated aspect. However, it is to be observed that the valley retains the V-shape characteristic of stream erosion as well as remnants of a stream-channel, now apparently no longer functional, at the bottom of the V. These facts constitute almost irrefutable proof of non-glaciation, for even moderate glacial action would have sufficed, considering the jointed structure of the granite of Shepherd’s Crest, to remodel the valley into a fairly smooth U-shape and to wipe the central stream-channel out of existence; and no amount of nivation would have transformed a glacial U-shape back to a V-shape, or would have produced a new central channel. Its distinct V-shape, therefore, together with its nivated aspect, proves conclusively that the little valley on Shepherd’s Crest has remained unglaciated.

Perhaps it will seem as though this conclusion had been reached with needless caution; but it is to be borne in mind that a hollow feature such as a valley is inherently well-adapted for the catchment of large quantities of snow and for the generation of a glacier—much better adapted than a tabular or convex

Ancient landscape on North Peak, Mount Conness in right background. Taken from the top
of the avalanche chute at the head of Little Lost Valley. The nivated slope of North Peak
contrasts strikingly with the glacial sculpture roundabout. By Alfred Dole

[click to enlarge]
Ancient landscape on North Peak, Mount Conness in right background. Taken from the top of the avalanche chute at the head of Little Lost Valley. The nivated slope of North Peak contrasts strikingly with the glacial sculpture roundabout. By Alfred Dole

summit. The non-glaciation of the little valley on Shepherd’s Crest therefore seemed rather unexpected, and it called for particularly convincing proof.

Such proof having been found, there opens at once a new vista of thought on the subject of the non-glaciation of the high tabular summits of the Sierra Nevada in general. All the tabular summits I have been able to examine bear the earmarks of prolonged nivation, yet corroborative evidence of their non-glaciation is not in every instance afforded by their topography. However, if the valley on Shepherd’s Crest has definitely escaped glaciation, then the presumption is all the stronger that these tabular summits—or at least a large proportion of them—have escaped glaciation also.

Now, these summits, mark you, are situated in the highest parts of the range, whence emanated the mighty ice-streams of the glacial epoch—ice-streams that attained lengths of thirty to sixty miles and depths of 2000 to 4000 feet. Shepherd’s Crest itself stands between two large cirques that formerly held glaciers a thousand to fifteen hundred feet in thickness, and it fronts on Virginia Canyon, which was the pathway of a trunk glacier fourteen miles in length and 2000 feet in thickness. The unglaciated slope on North Peak and the gently sloping platform to the southeast of Mount Conness are both literally surrounded by deep cirques that sent forth good-sized ice-streams. The same is true of the level top of White Mountain, of the Dana Plateau, of the tabular summits of Mount Gibbs, Kuna, Koip, and Blacktop peaks, and amongst many others farther south, of Mount Darwin and Mount Whitney. How then, it may be asked, does it happen that all these high-level tracts have escaped the heavy hand of the ice which wrought destruction all around them?

One reason readily suggests itself from the fact that they are all so oriented as to be exposed to the heat of the midday sun as well as to the southwesterly winds—which are the prevailing winds in the High Sierra, as is so eloquently attested by the asymmetric and even recumbent forms of the timberline trees. Everyone of the tabular summits and slopes before mentioned is inclined to the southwest, the west, or the south. Even the little valley on Shepherd’s Crest, although its axis trends northwestward, has in the main southwesterly exposure, for the row of pinnacles on its southern edge is too low to create a “wind shadow” of any consequence. Moreover, any westerly air-currents that enter the little valley at its lip must in part be deflected by the high northern crest so as to turn directly up the valley.

Now, it is a fact of observation that the southwesterly winds blow the bulk of the snow, while it is still in a powdery state, from the exposed slopes up over the mountain crests, and fling it in great banners, as Muir aptly called them, out to the northeast, to let it swirl down at last in the sheltered valley below. Whatever snowdrifts remain untouched by the wind are later consumed by the rays of the sun, and so toward midsummer all southwesterly and southerly mountain sides are wholly bared, whereas the northeasterly and northerly sides are still generously flecked with snow, and in some places even retain perennial ice bodies.

In an article which he published in the Sierra Club Bulletin, as well as in the Journal of Geology, the late Dr. G. K. Gilbert6 pointed out that during the Ice Age this markedly unequal distribution of snow, due to the combined action of wind and sun, must have tended to minimize glacial action on the southwesterly and southerly sides of the mountain crests and to intensify it on their northeasterly and northerly sides. As a consequence, many of these crests are now decidedly asymmetric in form, their southwesterly and southerly sides sloping at moderate angles, and their northeasterly and northerly sides being very abrupt, in part composed of unscalable cliffs. Dr. Gilbert saw, furthermore, that this asymmetry becomes more pronounced toward the lower levels of the High Sierra, where the windswept and sunny slopes were only feebly glaciated, and that it reaches an optimum at what may be termed the lower limit of glacier generation, where small glaciers could exist only on the sheltered northerly and northeasterly sides of the ridges, and where the southerly and southwesterly slopes remained wholly unglaciated. The contrast there is between the hacked-in headwalls of small cirques, on the one hand, and the gentle contours due to normal weathering and stream erosion, on the other hand. But, curiously, Dr. Gilbert did not complete his analysis. He did not see that the asymmetry of the crests becomes more pronounced also toward the upper levels of the High Sierra, and reaches another optimum on the lofty, tabular summit peaks, where the contrast again is between intense glaciation, on the one hand, and complete non-glaciation, on the other.

Three circumstances account for the non-glaciation of the tabular summit peaks of the Sierra Nevada and the little valley on Shepherd’s Crest: First, the southwesterly winds attain much greater velocity and sweeping power at the crest of the range than at lower levels on its west slope; second, because of the cold and the dryness of the air at the higher altitudes, the snow there remains longer in a powdery state and susceptible of being shifted about by the wind; and third, less snow falls in the winter on the main summit peaks than at levels 2000 to 3000 feet lower down. The last statement, it is true, is not supported by actual measurements of snow at different elevations on the west slope, but it may

On extreme summit peaks. Southwest side nivated only.  Example, Dana Plateau Halfway up in High Sierra.  Both sides glaciated, but southwest side only moderately.  Example, Mount Hoffmann.  At lower level of glacier generation. Southwest side unglaciated.  Example, Horse Ridge, near Ostrander Lake. Figure 11.—Profiles of asymmetric crests in Sierra Nevada.
[click to enlarge]
Figure 11.—Profiles of asymmetric crests in
Sierra Nevada.

safely be inferred from the fact that the clouds which blow in from the Pacific Ocean discharge the bulk of their snow approximately at the level at which they strike the chilling body of the range—at altitudes between 8000 and 11,000 feet.7 Thence upward they inevitably discharge diminishing quantities as they rise toward the summit. In these respects the conditions in the Sierra Nevada are analogous to those that obtain in many other mountain ranges of great height, notably the Swiss Alps, the Pyrenees, the Caucasus, and the Andes of South America. In all these ranges the zone of maximum snow precipitation is known to lie several thousand feet below the summit peaks.

During the more severe climate of the glacial epoch, naturally, the snow-clouds hung even lower on the Sierra Nevada than they do today, and the zone of maximum snow precipitation was correspondingly lower on its west slope. The tabular summit peaks then received proportionately less snow than now, and rose into regions of relative aridity. In that wintry epoch too, no doubt, the southwesterly winds went roaring over the crest of the range with greater fury than at the present time, and so, for both of these reasons, the conditions were peculiarly favorable for the non-glaciation of the higher wind-swept slopes. Paradoxical though it may sound, then, it is because of their great height that the tabular summit peaks and the little valley on Shepherd’s Crest have remained unglaciated.

Remains the question: How old is the little valley on Shepherd’s Crest? Or, more generally, how old is the “ancient landscape” of which it and the numerous tabular summit-tracts in the High Sierra are the remnants? Is it possible to determine its age in any way? Yes, it is possible, though only roughly and by roundabout methods.

It will be remembered that the Sierra Nevada consists essentially of a vast block of the earth’s crust that lies tilted to the southwest, so that its eastern edge forms the crest line and its western edge lies deeply buried beneath the sediments in the great valley of California. This great earth-block gained its tilted attitude not at one bound but by successive hoists separated by long intervals of relative stability—intervals to be reckoned in millions of years. With each uplift the streams coursing down its west slope were tremendously accelerated and intrenched themselves in narrow steep-sided canyons. During each interval of repose their downward cutting slackened, the canyons widened out to valleys by the weathering and erosion of their sides, the tributary streams cut ramifying valleys, and there was developed a landscape or “erosion surface” with a topography of its own. Naturally the canyons and valleys of each new cycle of stream activity were cut into the topographic forms left by the preceding cycle, and so each new landscape was developed at the expense of the previous one.

On the west slope of the Sierra Nevada there can be distinguished four sets of topographic forms recording the work of as many cycles of stream erosion. The newest forms are the narrow V-shaped canyons in which the main streams now flow. They were carved in consequence of the last uptilting of the Sierra Block, which occurred probably early in the Pleistocene epoch.8 Less than a million years old, they are still being actively deepened by the streams and remain youthful in aspect.

To a close observer it is patent that these Pleistocene canyons were cut into the broad floors of mature valleys of an earlier cycle. The Big Meadow flat, which lies more than 2000 feet above the Merced River at El Portal, is a remnant of such an older valley. The gently sloping platform about Turtleback Dome, over which the new highway to the Yosemite Valley is laid, is another remnant, and so is the entire Illilouette Valley, which has never been trenched. Examples are plentiful also along the other canyons of the Sierra Nevada, notably along those of the Stanislaus and San Joaquin. These older valleys, which attain great breadth on the lower slope of the range, are the products of a much longer cycle of erosion—a cycle that comprised probably all of the Pliocene epoch.

Big Meadow, Turtleback Dome, and the Illilouette Valley in their turn lie 2000 to 2500 feet below the general level of the little valleys on the uplands that flank the Yosemite. These billowy uplands are, indeed, portions of a still earlier landscape—a landscape that was produced during a very long cycle of erosion comprising most of the Miocene epoch and probably large parts of the preceding Oligocene epoch. Its age cannot be determined in the Yosemite region for want of telltale fossils, but it is indicated as probably late Miocene by fossils found near the old mining town of Columbia, north of the Tuolumne Table Mountain.

High above the Miocene landscape, which remains preserved on many of the extensive intercanyon tracts, stand the peaks and ranges that give the High Sierra its alpine character; and it is on some of the loftiest of these peaks and ranges, 2000 to 3000 feet above the Miocene hills, that are found the gently sloping, tabular remnants of the ancient landscape to which the little valley on Shepherd’s Crest belongs. The age of this landscape is indicated approximately by the fact that in the northern parts of the Sierra Nevada remnants of it lie 1000 to nearly 2000 feet above the “fossil stream-beds” that contain the earlier gold-bearing gravel. These stream-beds, which were preserved by masses of indurated volcanic ash (rhyolite tuff), have yielded fossil plant remains of middle Eocene age. It follows that the ancient landscape in question goes back at least to early Eocene, possibly, even, to late Cretaceous time.

That any parts of a landscape so ancient could remain preserved in exposed mountaintops may at first seem incredible. Yet in the Sierra Nevada the fact is hardly open to doubt. Three circumstances, it would appear, have operated to preserve those bits of the early Eocene landscape that form the tabular summits of the highest peaks—namely, the resistant nature of the granitic rocks of which those peaks are made; the position of those peaks at the extreme heads of the rivers, where the streams are smallest and have the least cutting power; and their complete exemption from glacial erosion. Of course, it is not contended that these residual summit tracts have suffered no degradation whatever since early Eocene time; but the fact is stressed that they have suffered but very little change as compared with the deep canyons that surround them—so little, that they retain the gentle slopes and rounded contours that were imparted to them when the Sierra region still was a land of moderate elevation.

Of all the ancient summit-tracts in the High Sierra, certainly the little valley on Shepherd’s Crest seems most remarkable; for a valley, being the pathway of a stream, is inherently more likely to be cut away during the uplift of a mountain range than is a ridge or a summit. Only some special circumstance could have saved it. Perhaps the streamlet on Shepherd’s Crest was unable to compete with its neighbors because its water was entrapped by vertical fissures that developed across its path—the same fissures that separate the pinnacles of the south crest from one another. Again, the little valley seems remarkable because it has escaped glaciation, although valleys inherently afford good sites for glaciers. And, finally, to a student of the High Sierra it seems particularly precious because its non-glaciation, so well attested by its form, confirms the non-glaciation of many of the lofty tabular summits of the Sierra Nevada.

Reprinted from Sierra Club Bulletin, 1933, pages 68-80.

1The topographic features referred to in this essay are in the headwaters area of the Tuolumne River. See the topographic map of Yosemite National Park.—Ed.

2Might not the expressive English word cleaver be more generally adopted in our vocabulary of mountain terms, in place of the alien and often mispronounced arête, thus leading the way, perhaps, to the expulsion of terrible bergschrunds and fanciful roches moutonneés?—F. E. M.

3Matthes, F. E. Glacial sculpture in the Bighorn Mountains, Wyoming: U. S. Geological Survey, 21st Annual Report, part 2, 1900, pp. 167-190.

4Soil flow is relatively rare on the granitic peaks of the Sierra Nevada, but evidences of it were observed last summer on the Dana Plateau. Both nivation and soil flow are common phenomena in Alaska, and they occur on a large scale in northern Greenland, where, in spite of the high latitude, no glaciers ever existed.

5Blackwelder, Eliot. Pleistocene glaciation in the Sierra Nevada and Basin Ranges: Geological Society of America Bull., vol. 42, 1931, pp. 865-922.

Matthes, F. E. Geologic history of the Yosemite Valley: U. S. Geol. Survey Prof. Paper 160, 1930.

6Gilbert, G. K. Systematic asymmetry of crest lines in the High Sierra of California: Journal of Geology, vol. xii, no. 7, 1904, pp. 579—588; Sierra Club Bulletin, vol. v, 1905, pp. 279—286.

7These figures are for the latitude of the Yosemite region. They are rough approximations. More accurate data are desired.

8See footnote on page 97, and Appendix.—Ed.

Next: Tuolumne MeadowsContentsPrevious: Devil’s Postpile

Home A - Z FAQ Online Library Discussion Forum Muir Weather Maps About Search
Online Library: Title Author California Geology History Indians Muir Mountaineering Nature Management