USGS Logo Geological Survey Professional Paper 160
Geologic History of the Yosemite Valley




To one who has not extended his observations beyond the great series of lateral moraines that cling to the sides of the Little Yosemite it would seem entirely justifiable to conclude that the conspicuous embankments at the top of that series mark the highest level ever attained by the Merced Glacier. Indeed, no further moraines of a like nature are to be seen on the slopes above, nor do these slopes exhibit any of the sculptural effects characteristic of glacial action. On the contrary, they are in places corrugated by spurs and ravines or show other inequalities that are obviously products of normal erosion—that is, chiefly of the erosive action of running water in streamlets and rills.

Nevertheless, careful scrutiny of the slopes above the highest morainal embankments reveals the presence of scattered boulders of Cathedral Peak granite—the granite with the large feldspar crystals which is so readily identified among the materials that have been brought down from the High Sierra by the glaciers (p. 56). There are also scattered boulders of quartzite, schist, and other types of rock that are foreign to the Yosemite region and clearly derived from the crest of the range. Most of these boulders are, like those in the moraines described, imperfectly rounded, but unlike them they are invariably rust stained and weathered to a depth ranging from an eighth to half an inch. Many of them are so "rotten," to use the popular expression, that they are traversed by a network of cracks and fall apart at a moderate blow. Thus it becomes increasingly apparent, as the search is continued, that the ice of the glacial epoch did once, at a relatively remote date, attain a level much higher than that indicated by the prominent moraines.

An excellent locality for such a search is the slope on the north side of the Little Yosemite which is traversed by the trails that lead to Half Dome and Clouds Rest. This slope is remarkably smooth and easy to traverse, compared with the rough, stony belt of moraines below. It is surfaced for the most part with a thick layer of granite sand and forest soil, materials whose formation must have required considerable time. In some places the slope is cut by ravines such as occur in unglaciated regions. It also bears here and there crags and pinnacles of granite that evidently form part of the body of the mountain and owe their fantastic shapes to long continued disintegration under atmospheric influences.

In the deeper trail cuts and in holes left by uprooted trees, moreover, boulders and cobbles of typical glacial forms protrude from under the surficial blanket of sand and soil. When freshly exhumed these boulders and cobbles are enveloped almost invariably in a yellowish, rusty coating that masks their constituent minerals and makes all look alike, although they may belong to widely different rock types. Thus, as the traveler proceeds up the slope, he comes to realize that, though smooth, it is in reality veneered for the most part with old glacial materials and that the conspicuous boulders of Cathedral Peak granite which attract the eye are but a few of the larger fragments that are not entirely covered over with soil. What is more, he perceives that in many places the old glacial materials form indistinct swells a foot or two in height that extend across the slope. When traced out and plotted on a map these swells are found to lie roughly parallel to one another like the crests of the series of lateral moraines below. They are indeed old, dim moraines that have long since lost their crests and have been reduced by slow disintegration to a mere fraction of their original height. As may be seen on the map (pl. 29), they are on the average broader individually and spaced farther apart than the younger, sharp-crested moraines and would seem to record the fluctuations of a larger and more heavily loaded glacier.

The highest of these old moraines lies to the southeast of the upper of the Quarter Domes, at an altitude of about 8,250 feet, more than 1,000 feet above the level of the highest of the younger moraines. It terminates at the brink of Tenaya Canyon in such a way as to show that at one time the Merced Glacier there united with the Tenaya Glacier at a common level, the two forming one vast, continuous sea of ice above which only the summits of Clouds Rest and Half Dome rose like bold rocky isles and beneath which the saddle separating these two eminences lay submerged to a depth of fully 700 feet.

On the south side of the Little Yosemite similar vestiges of old moraines occur at corresponding levels. On the plateau-like upland to the northeast of Mount Starr King the conditions evidently were particularly favorable for their preservation, for there several of the old moraines still retain for considerable distances their characteristic ridge forms, though the sharp crests which doubtless they originally possessed are now reduced to rounded backs cloaked with granite sand derived from disintegrated boulders. They may be readily traced south of Helen Lake and in the vicinity of the Starr King Meadows, where they describe great arcs, showing how the glacier deployed broadly over the more level parts of the upland and again contracted when it met the obstructing masses of Mount Starr King and its companion domes.

It is on this upland to the south of the Little Yosemite, perhaps, that the student may best learn to recognize these older moraines by their constituent materials as well as by their indistinct forms. His eye, having become sensitive to these characters, will then discern dim vestiges of old morainal deposits in many parts of the Yosemite region where otherwise it would detect nothing indicative of the former presence of glaciers. Thus a little searching will reveal the remnants of a series of old lateral moraines that swing in sympathetic curves across the recess in the south side of the Little Yosemite, north of the Starr King group. Another, still dimmer series of moraines slope from the north shoulder of that group southwestward into the Illilouette Valley, outlining, as may be seen on Plate 29, the successive positions occupied by the retreating margin of a massive ice lobe of the Merced Glacier which penetrated the Illilouette Valley for about 2 miles.

No one who has traced these ancient moraines (one of them is a massive, fairly well preserved embankment that extends unbroken for a distance of more than a mile) can have the slightest doubt as to the reality of this ice lobe which pushed up the valley, nor is he likely to entertain the belief which so many writers on the Yosemite region appear to have shared, that the Illilouette Valley formerly contributed a large volume of ice to the Yosemite Glacier. The younger moraines show clearly that during the last glacial stage the Illilouette Valley contained no ice for a distance of 5 miles up from its mouth, and the older moraines show that during the earlier stages, when the ice reached its greatest extent and depth, the Illilouette Glacier was effectually imprisoned by a lobe from the trunk glacier itself.

The older moraines indicate, further, that at the time of maximum glaciation the Illilouette Glacier and the opposing lobe met and united, forming a broad sea of ice that eventually rose high enough to find an outlet westward into the Bridalveil Basin. Later, when the Illilouette Glacier and the lobe melted apart, there was formed between them a temporary lake—ancient Lake Illilouette it may be called. The gravel and sand deposited in this body of water remain to attest its former presence. (See pl. 32, A.) They cover a stretch of nearly 3 miles, in part bare, in part covered with manzanita bushes, whose brilliant green, contrasting with the somber tones of the surrounding pines, creates the illusion, at a distance, of a verdant meadow. Muir48 long ago recognized these deposits of sand and gravel as indicative of a temporary glacial lake, but he makes no mention of an ice lobe, nor is there anything in his writings to show that he apprehended the real nature of the dam by which the waters were impounded.

48Muir, John, Studies in the Sierra—Ancient glaciers and their pathways: Sierra Club Bull., vol. 10, p. 193, 1917.


Of more than academic interest are the glacial deposits that occur near Glacier Point, for they definitely answer the question so often asked by visitors, whether that high promontory was ever overtopped by the ice. The extreme point of the promontory is wholly bare, but in the immediate neighborhood of the Glacier Point Hotel, on the slopes below, in the hollow to the west, and, most significantly, on the wooded slope above, glacial material is abundant. This material is spread out in a sheet that scarcely suggests distinct moraine ridges, yet its glacial origin is definitely proved by the presence in it of rocks derived from the Little Yosemite and the High Sierra. As so much depends upon the testimony of these rocks it seems worth while here to describe some of them.

Most plentiful are rounded boulders and cobbles and angular fragments, all deeply weathered, of Half Dome quartz monzonite,49 the light-colored granite of which not only Half Dome but all of the Little Yosemite and its surrounding heights are composed. Clearly this rock would make up a large proportion of the load gathered by any glacier passing through the Little Yosemite, and naturally it would be abundant in the moraines left by such a glacier at Glacier Point. Some of the quartz monzonite in the deposits about Glacier Point might possibly be derived from localities nearer than the Little Yosemite—for instance, from the giant stairway, or from the walls of the Illilouette Gorge but if so, it obviously could not have been carried from those places up to Glacier Point, which is fully 2,000 feet higher, save through the agency of a glacier.

49The characteristics of this rock and those mentioned in the following paragraphs are described more fully in the appendix, p. 126.

There are also a few boulders of a coarse-grained, highly siliceous granite, light buff in general tone when fresh but vivid rose when weathered. They are of peculiar interest because there is only one place in the High Sierra above the Yosemite from which they can be derived—Mount Clark, the sharp-profiled peak which stands on the east side of the Illilouette Basin, 8 miles from Glacier Point. Being devoid of hornblende and biotite, the dark minerals that speckle all the granitic rocks of the Yosemite region, the Mount Clark granite is readily identified in the miscellaneous assortment of rocks that make up the moraines.

Finally, there are fragments of yellowish quartzite and gray schist whose places of origin have been located by Calkins on the long northern, spur of Mount Clark. It is evident that these two highly distinctive materials and the Mount Clark granite were carried to Glacier Point on the southern margin of the ice stream.

On the slope above Glacier Point the morainal deposits, traced by these and other diagnostic materials, are found to extend up to an altitude of 7,700 feet, thereby affording indubitable proof that the ice once rose 500 feet above the promontory. Higher still, however, on the bare rock slope at the immediate base of Sentinel Dome, scattered boulders indicate that, perhaps at a very early stage in the glacial history of the region, the ice attained an altitude of over 7,900 feet, or fully 700 feet above Glacier Point. Of the significance of these scattered boulders more will be said on another page.


The rocky platform at Glacier Point, from whose edge the sightseer beholds the grand panorama of the Yosemite Valley and the High Sierra, presents some features that are commonly, though erroneously, regarded as evidence of the passage of a glacier over the promontory. Whether it was for these features that Glacier Point was named is uncertain—the origin of the name is obscure—but in any event it seems appropriate here briefly to digress in order to explain their true nature.

The features mentioned are basin-shaped cavities in the rock, measuring as a rule from 12 to 18 inches in diameter, and from a few inches to 6 inches in depth. A typical example is shown in Plate 33, A. They bear some resemblance to "potholes" such as are worn in the rocky beds of streams by cobbles whirled round and round by a swift current, and this is true especially of the deeper basins.

Now it happens that in the minds of many persons such potholes are associated with glaciers, being held to be characteristic products of "glacier mills" (called "moulins" by the Swiss)—that is, torrents of water descending through crevasses in the ice and impinging with great force on the rock bed below—and as a consequence the cavities at Glacier Point are generally linked with the former presence of a glacier. As a matter of fact, potholes are formed by whirling cobbles in the beds of open streams as well as in the beds of subglacial streams, and they do not, therefore, afford prima facie evidence of glaciation. Potholes of both kinds abound in the Sierra Nevada. A fine series of potholes that are unquestionably of subglacial origin is to be seen at the lower end of the Tuolumne Meadows, extending across the ice-smoothed slope of a low dome of granite where, manifestly, no stream could have flowed in the open. (See pl. 34, B.) On the other hand, potholes that clearly have been formed without the intervention of a glacier are plentiful in the lower Merced Canyon and in other canyons and gulches that have never been penetrated by glaciers.

The cavities at Glacier Point, however, upon close scrutiny, are found to be not stream-worn potholes but products of strongly localized weathering, a process that affects the granitic rocks of the Sierra Nevada in many places where they are directly exposed to the atmosphere. They belong to a class of features that will here be referred to as "weather pits." The development of cavities of this type is promoted as a rule by the presence in the rock of local aggregates of readily soluble minerals. A small initial hollow having been formed by the decomposition of such an aggregate of minerals, it becomes a receptacle for water from rains or melting snow and is enlarged gradually by both chemical and mechanical processes. Acids produced by decomposing pine needles, lichens, or other vegetal matter attack the weaker minerals, and in freezing weather the ice, expanding with force as it crystallizes, pries off flakes and grains of rock from the walls. Thus weather pits may expand from an initial diameter of about an inch to diameters of 2 or 3 feet, and those situated close together may eventually coalesce as their rims intersect. (See, pl. 33, C.) Their growth in depth, on the other hand, seldom keeps pace with their lateral expansion, for the less soluble particles of rock detached from the rims collect at the bottom and, although the finer particles are blown out by the wind in dry weather, usually enough of the coarser ones remain to form a protective pad that tends to retard downward excavation.

PLATE 33.—A (top, left), WEATHER PIT AT GLACIER PIT. Cavities of this type are commonly mistaken for potholes such as ar worn in stream beds by swiftly rotating cobbles. Those at Glacier Point have been pointed to as evidence of the passage of a glacier over the promontory. The earlier ice did pass over Glacier Point, but these cavities were produced long afterward by strongly localized weathering of the rock, promoted by pools of water. Such weather pits occur in the Sierra Nevada in many places where no glaciers have ever penetrated. Phogotraph by F. C. Calkins.

(B, top right), TYPICAL "PERCHED" BOULDER OF AN EARLIER GLACIATION. The boulder is on the divide south of the Starr King Meadows, about 3 miles from its probable place of origin on the Clark Range. It measures about 5 feet to the side and is perched on a pedestal 20 inches high. The pedestal is composed of local rock and has remained preserved as a result of the protection from the weather afforded by the boulder. Photograph by F. C. Calkins.

C (bottom), WEATHER PITS IN SLAB ON NORTH DOME. These weather pits have been formed entirely since the earlier ice passed over and smoothed the crown of the dome. The ice of the last glacial stage did not reach this level. Several of the weather pits here shown have expanded until they coalesce. Others are about to coalesce, the rock partitions between them being already broken down.

Some weather pits are difficult to distinguish from stream-worn potholes, but as a rule such pits possess characteristics that set them clearly apart, for the processes whereby they are enlarged, being dependent largely upon the presence of standing water, operate intensively only up to the level of the water surface and thus tend to undercut the sides, leaving the rims overhanging. This undercutting action is of course intensified in those weather pits in which the depth of water is definitely controlled by an outlet, giving rise to sharply overhanging rims. Plate 33, A, shows a characteristic example, and other examples are to be seen in Plate 33, C. The weather pit in Plate 33, A, moreover, has alcovelike recesses under the rim, which show that enlargement is proceeding more rapidly at some points than at others. The resultant scalloped form contrasts strikingly with the smoothly cylindrical forms characteristic of true potholes, produced by the grinding, boring action of whirling cobbles.

It is a significant fact that weather pits do not occur on freshly glaciated rock surfaces. Not a single one is to be found within the area that was covered by the later glaciers. Evidently they develop at an extremely slow rate, and not enough time even for their initiation has elapsed since the glacial epoch. On the other hand, weather pits do occur on rock surfaces that were overridden by the earlier glaciers, as well as on such surfaces as have remained wholly unglaciated, these two kinds being indistinguishable, so far as effects of weathering are concerned. Particularly fine examples of weather pits that have developed since the passage of the earlier ice are to be seen on the summit of North Dome (pl. 33, C); examples of weather pits that have developed in unglaciated localities are to be seen on the summits of Sentinel Dome and Illilouette Ridge.

The presence of weather pits on Glacier Point thus clearly affords no proof of the glaciation of that promontory but attests its long exposure to the weather since it was glaciated.


Along the south rim of the Yosemite Valley, west of Sentinel Dome, one looks in vain for older moraines. Evidently the earlier Yosemite Glacier in this part of its course declined rapidly westward, so that it lay entirely beneath the brink of the upland. The precise level of its surface at the time of maximum glaciation is not indicated on the cliffs, for these are too precipitous to retain glacial débris and have long since lost whatever glacial polish, striae, or sculptural effects were imparted to them by the ice. Fortunately, however, the Cathedral Rocks bear on their summits and slopes small deposits of glacial boulders that give some idea of the depth attained by the earlier Yosemite Glacier.

To Frank C. Calkins belongs the credit of discovering these boulders on the Cathedral Rocks and of identifying them beyond all possible doubt as ice-borne boulders. Among those on the main summit of the Cathedral Rocks he found several boulders composed of the porphyritic phase of the Half Dome quartz monzonite which occurs near Tenaya Lake and Merced Lake; also a fragment of limestone derived from the metamorphic belt of the High Sierra. On the slope immediately south of the main summit he found a boulder of the familiar Cathedral Peak granite.

So imposing are the Cathedral Rocks, especially when viewed from the floor of the Yosemite Valley (pl. 18), that it is difficult to imagine them as once having been completely overtopped by the ice, yet the testimony of the boulders leaves no doubt that they were. Even the highest of these boulders, at an altitude of 6,638 feet, do not mark the highest level reached by the ice flood, for on the brush-covered summits south of the Cathedral Rocks other remnants of moraines occur at altitudes ranging from 6,800 to 6,900 feet. These remnants show that the earlier Yosemite Glacier passed over the main summit of the Cathedral Rocks with a depth of not less than 300 feet.

Turner,50 who first noticed these higher remnants of moraines, thought that they might have been deposited by the tributary glacier which issued from the hanging valley of Bridalveil Creek. However, from their very position on the east, up-valley side of Bridalveil Creek, it would seem more likely that these moraines were deposited by the main valley glacier, and this inference is borne out by an examination of their constituent materials.

50Turner, R. W., The Pleistocene geology of the south-central Sierra Nevada, with especial reference to the origin of Yosemite Valley: California Acad. Sci. Proc., 3d ser., vol. 1, p. 304, 1900.

A survey of the Bridalveil Basin discloses that the tributary glacier which it sent forth was of small volume and carried but little débris. Older moraines are scarce in the basin and, on the whole, are difficult to trace. However, they indicate unmistakably that the Bridalveil Glacier, which had its sources in the cirque now occupied by Ostrander Lake and on the adjoining slopes of Horse Ridge, was reinforced in its lower course by a considerable flow of ice from the Illilouette Basin. This flow came through the saddle southwest of Mono Meadow.


As compared with the south side of the valley, the north side is rich in glacial deposits and affords an abundance of data that throw light on the earlier glacial history of the Yosemite region. The slope between North Dome and Basket Dome, for instance, is veneered with old glacial débris up to an altitude of about 7,600 feet. Higher still, on the southeast side of Indian Ridge, is a fairly distinct lateral moraine that marks the junction of the Snow Creek Glacier with the Tenaya Glacier, a short distance above the confluence of the Tenaya with the ice stream in the Yosemite Valley. This lateral moraine, which is comparable in size and state of preservation to the morainal embankments that are situated near the Starr King Meadows, can be traced up to an altitude of about 8,150 feet. However, even this is not, apparently, the highest level reached by the earlier ice, for in the vicinity of Indian Rock, at altitudes as high as 8,400 feet, there are scattered boulders, among which are several composed of porphyritic granite from Mount Hoffmann.

Large quantities of glacial débris, including several remnants of lateral and frontal moraines, occur in the valley of Snow Creek; but instead of describing these deposits it seems far more important to invite attention to the few isolated boulders and cobbles that lie on the bald, rounded summit of Mount Watkins. These clearly attest that this Capitan-like eminence, which stands 3,300 feet above the floor of Tenaya Canyon, was once completely overswept by the ice. The observer marvels that snow and ice could have accumulated in this part of the Yosemite region to so great a depth, yet the testimony of the boulders and cobbles is not to be doubted. Among them is a 3-foot boulder representing the porphyritic phase of the Half Dome quartz monzonite which occurs near Tenaya Lake; and some of the subangular fragments and rounded cobbles consist of yellowish, reddish, and banded quartzites from the High Sierra.

North of Mount Watkins, at several places on the upland, are other glacial boulders and patches of morainal material which show that all of this upland, as far north as Mount Hoffmann, was once enveloped by a continuous mantle of ice. West of Indian Ridge this glacial mantle extended over a considerable part of the upland north of the Yosemite Valley, as is shown by the occurrence of fragments of granite from Mount Hoffmann in the basins of Indian Creek and Yosemite Creek and by the presence of southward-trending moraines on the divide that separates those two basins. Even the summit immediately west of the head of the Yosemite Falls is crowned by glacial boulders.

Farther west, however, are eminences which, beyond a doubt, were never overridden by the ice. Such are Boundary Hill and the long crest that extends northward from that hill to the head of Yosemite Creek. They were high enough to stem the westward flow of the ice—high enough, even, to bear small glaciers of their own, as is attested by the cirques that scallop their sides.

Another eminence that stood above the highest ice flood is Eagle Peak, the sharp, craggy summit which surmounts the massif of the Three Brothers, on the north side of the Yosemite Valley. The highest moraine lies but a short distance north of the peak and at an altitude of somewhat over 7,500 feet, or only about 250 feet below its top. Northward and northeastward from Eagle Peak, on both sides of the Eagle Peak Meadows, extend dim old moraines that outline the front of an ice lobe which pushed southward up into this basin.

Of peculiar interest is the story told by the moraines in the little valley of Ribbon Creek. Muir51 did not fail to observe these moraines, but he was puzzled at the utter absence of glacier polish, of which there is so much in the neighboring valley of Yosemite Creek, and he therefore concluded that the little Ribbon Glacier, because of its southerly exposure and its small volume, "was one of the first to die"—a very reasonable conclusion considering the fact that in 1871, when Muir wrote this, the ice age was still conceived of as a single uninterrupted reign of ice, instead of a succession of protracted cold glacial stages separated by equally protracted mild interglacial stages, as it is now known to be. The fact is that the valley of Ribbon Creek contains only older moraines. It was filled with ice only during the earlier parts of the glacial epoch, of which no polish now remains preserved in the Sierra Nevada, save in a few exceptionally favored localities. Of this ice only a small proportion came from local cirques; most of it was supplied by overflow from the valley of Yosemite Creek, through the gap north of Boundary Hill. During the last glacial stage, which is so vividly recorded in other parts of the Yosemite region by sharp-crested moraines and polished rock pavements, the valley of Ribbon Creek remained wholly free of ice.

51Badè, W. F., The life and letters of John Muir, vol. 1, pp. 304-305, 1924.


Most important of all in this study are the older moraines in the lower part of the Yosemite region, for they indicate the farthest limits reached by the ice. Though covered almost throughout with forest and brush and though in places washed away, they are nevertheless not difficult to trace, and together they outline distinctly the margins of the earlier Yosemite Glacier within a short distance of its terminus. Indeed, they outline the course of the glacier so distinctly that it seems surprising that there should ever have been any controversy as to the maximum extent of the Yosemite Glacier.

The highest lateral moraine on the south side of the valley begins in the first hollow west of Old Inspiration Point and thence extends westward along the slope. It is crossed by the Pohono Trail at an altitude of 6,300 feet, or 2,400 feet above the floor of the valley. Below it are other relatively dim moraines, across which the Pohono Trail zigzags down to the Wawona Road. One of them bears a large angular block of Cathedral Peak granite.

The highest moraine, though gashed at intervals by ravines, and in places reduced to an inconspicuous terrace, is readily traced westward for a distance of nearly 2 miles. It reaches the Wawona Road near the point where that road turns southward at an altitude of 5,650 feet. On the precipitous rock slopes below the road, however, the moraine loses its identity, and the second highest moraine, situated on the back of a sloping spur, gains prominence. (See pl. 34, A.) Indeed, so well preserved is this ancient moraine for a stretch of a quarter of a mile that it constitutes one of the notable landmarks of glacial origin in the lower Yosemite region. About 500 feet below it, on Turtleback Dome, several lesser moraine ridges record successive phases in the recession of the glacier. In the hanging valley of Grouse Creek the highest moraine reappears as a strong and continuous embankment that effectually blocks the creek and deflects it southward for a distance of half a mile. Grouse Creek now finds its way to the canyon through a sharp notch which it has cut in the embankment south of a rocky knoll. Thence it cascades down over the sloping cliffs through a rough, unfinished rock channel. Evidently the stream has not followed its new path long enough to wear a smooth channel in the resistant granite.

PLATE 34.—A (top), OLD MORAINE NEAR WAWONA ROAD. One of the best preserved moraines of the older series. Most of the older moraines have long since lost their ridge forms and can be identified with certainty only by the character of their constituent rock materials. This moraine is situated below the Wawona Road, south of Turtleback Dome.

B (bottom), TYPICAL POTHOLE OF SUBGLACIAL ORIGIN, AT LOWER END OF TUOLUMNE MEADOWS. Above the pothole are remnants of a torrent channel across a steep rock slope where, manifestly, no stream could have flowed in the open. The rock slope and the channel are in part still veneered with glacier polish. Several of the cobbles that did the grinding lie in the pothole. A pine tree has taken root in the granite sand that has been washed into the hole. Phogograph by G. K. Gilbert.

Undiminished in strength the morainal embankment extends south-southwestward, over the next divide, to the mouth of the hanging valley of Avalanche Creek, outlining the sinuosities in the glacier's margin uninterruptedly for a distance of 2 miles and indicating the slope of its surface from a level of about 5,000 feet down to about 4,500 feet. Parallel to the embankment, between it and the edge of the Merced Gorge, lie several lesser moraines. All end at the gorge of Avalanche Creek. Beyond this gorge, on the prominent granite spur that marks the lower end of the Merced Gorge, none of them reappear, for want of a place on which to rest, but on the ravined sides of the lower Merced Canyon, west of the granite spur, a few disconnected patches of morainal material still remain in place. Naturally they are best preserved on the spurs between the ravines, the spurs having suffered least from erosion. Opposite El Portal, on the spur followed by the funicular logging railroad, such patches occur at altitudes ranging from 3,300 to 3,500 feet. A few cobbles and subangular fragments of granitic rocks (the local rocks are slate and quartzite) have been found west of the track at an altitude of only about 2,800 feet, but it is possible that these have rolled down from above.

On the north side of the Yosemite Valley the configuration of the cliffs and slopes does not permit the preservation of continuous embankments of great length, and only fragments of the right lateral moraines of the earlier Yosemite Glacier occur at intervals. The most distinct and perhaps most impressive fragment is situated at an altitude of 5,800 feet, on a knoll east of the valley of Cascade Creek, near the head of the main grade of the Big Oak Flat Road. Though this is doubtless one of the highest lateral moraines left by the glacier, it can not be the topmost, for it trends at right angles to the valley of Cascade Creek and evidently was deposited at a time when the tributary Cascade Glacier had already receded some distance up that valley. That the Cascade Glacier was confluent with the Yosemite Glacier during the culminating phases of the earlier glaciation is amply attested by the numerous lateral moraines which it has left behind. It attained a breadth of three-quarters of a mile and a depth of 500 feet just above the point of confluence.

On the slopes west of Cascade Creek the lateral moraines of the earlier Yosemite Glacier are relatively well preserved. The highest extend almost unbroken for a distance of 2-1/2 miles. No tributary glacier ever issued from the valley of Tamarack Creek to interrupt them. On the mountainous spur west of Wildcat Creek the moraines attain their fullest development, forming a parallel series analogous to the series of newer moraines that exists on the north side of the Little Yosemite, though, of course, much less distinct.

As the moraines curve around the brow of the spur most of them die out, but a few can be traced along its steep west side, where they decline rapidly to the Big Meadow Flat. Evidently this flat was invaded by a broad lobe of the earlier Yosemite Glacier, the hills on the west side of the Merced Gorge being too low to stem the ice flood. A series of concentric moraines, some of which are more massive than any other moraines in the Yosemite region, record the fluctuations of the front of this lobe on the north and west borders of the flat. Sandy terraces on the inner slope of the innermost of the concentric moraines show that during certain episodes in the recession of the ice front the waters of Crane Creek were imprisoned and formed a temporary lake along the north side of the ice lobe.

At the lower end of the Big Meadow Flat the moraines turn to the west and outline the margins of a narrow ice stream that flowed out from the flat. This ice stream cascaded abruptly into the deep gulch of Crane Creek and, following that gulch to its mouth, finally joined the Yosemite Glacier just above El Portal. The left lateral moraine of this Crane Creek Glacier is dimly visible on the crest of the spur east of the mouth of the gulch. Obscure remnants of the right lateral moraine lie near McCauley's ranch.

The farthest point reached by the earlier Yosemite Glacier is not marked by any terminal moraine, but that point can be determined, approximately at least, from the longitudinal profile of the glacier, as shown in Figure 22. The curve of the glacier's proffle is definitely determined by the lateral moraines as far west as El Portal, beyond which the curve can be built out—"extrapolated"—with some confidence, in accordance with the known laws governing the forms of glaciers. In any event, the absence of morainal material in the canyon west of El Portal does not imply that the glacier ended abruptly at that point. On the contrary, the very height at which the last patches of morainal material lie above the floor of the canyon—between 1,200 and 1,300 feet—shows rather definitely that the ice tongue extended some distance, perhaps fully a mile, farther west; for glaciers of the type to which the earlier Yosemite Glacier belonged ordinarily thin down to a depth of only a few hundred feet at their lower ends.

In this connection account must be taken, of course, of the fact that the lower Merced Canyon has been deepened somewhat by the river since the departure of the earlier ice. The depth of cutting is, however, only about 50 feet, as would appear from the position on both sides of the canyon of masses of boulders and coarse gravel that evidently are remnants of a "valley train" composed of outwash from the glacier. This outwash material is conspicuous in many cuts along the new automobile highway and also in some places along the railroad. It shows that the valley train extended down the canyon for about 30 miles and had a maximum depth, next to the ice front, of more than 100 feet. The base of this material, so far as can be ascertained, is about 50 feet above the present river bed; hence it may be concluded that the lower Merced Canyon had already been cut within 50 feet of its present depth by the time the glacier made its farthest advance. That it has been deepened no more than 50 feet since that time is explained in part by the fact that the river had to cut first through the accumulated outwash material, in part by the fact that owing to its low gradient on the metamorphic rocks below El Portal (70 feet to the mile, as compared with 350 feet to the mile in the granite gorge above El Portal) the river had but moderate cutting and transporting power and for a long time after the melting of the glacier it was too heavily loaded to cut.

As will be clear from the foregoing statements, the absence of a terminal moraine may possibly be accounted for by the erosive action of the Merced. A river that could remove an entire valley train of outwash material 30 miles in length might well have been able to remove also a terminal moraine or even several such moraines. On the other hand, it is possible that the glacier failed to pile up the débris in the form of a distinct moraine. Its front may have oscillated back and forth in such a way as to cause the débris to be spread out in an irregular sheet. That the second explanation is more probable than the first would seem to be indicated by the fact that no terminal moraines of the earlier glacial stage exist in any of the main canyons on the western flank of the Sierra Nevada that have been examined, not even in the canyons of the Tuolumne, San Joaquin, and Kings Rivers, which contained the mightiest trunk glaciers. Naturally all these trunk glaciers would have fluctuated in essentially the same way, being affected by the same climatic changes. Had they all deposited terminal moraines, it is more than likely that recognizable remnants of such moraines would still exist to-day in at least some of the canyons.

The absence of a massive terminal moraine in the lower Merced Canyon was one of the circumstances that led Prof. I. C. Russell and others among the earlier observers to doubt that the Yosemite Valley had been profoundly excavated by the ice. What has become, they asked, of the large quantities of boulders and fine débris which the glacier is supposed to have removed from the valley? However, these doubters overlooked the remnants of the long train of glacial outwash material in the lower Merced Canyon, which, when first laid down, must have had enormous volume. Even so, it contained but the coarser parts of the material excavated by the glacier. The finer parts had been swept away by the Merced River. Again, this train of outwash material represented the product of only one stage of glaciation. Farther down in the Merced Canyon are terraces composed of coarse gravel and boulders that are in all probability remnants of more extensive valley trains produced during earlier stages of glaciation. These terraces have not yet been studied in detail.


That the older and younger moraines differ considerably in age will be sufficiently clear from the foregoing paragraphs; still it is a pertinent question, how great the difference between them really is. Is it a matter of a few thousand years, or of hundreds of thousands of years? The significance of the question lies in the fact that if the difference is only a few thousand years, then in all probability the two series of moraines record merely two major advances made by the Yosemite Glacier in the course of a presumably continuous period of glacial conditions; whereas, if it is some hundreds of thousands of years, then the two series of moraines probably record two wholly distinct advances in separate stages of the ice age.

On the plains of the north-central United States, where the glacial record lies broadly spread out, there is evidence of four and possibly five extensions of the continental ice sheet that took place at widely different times, each in a separate glacial stage. Not only do the successive drift sheets differ considerably in the degree of dissection due to stream erosion and in the degree to which their constituent materials are oxidized and decomposed, but intercalated between them in some places are layers of ancient soil containing roots and stumps of trees and other plant remains that tell unmistakably of long intervals of nonglacial conditions, or "interglacial stages." In some, places, even, the character of the vegetal remains indicates for the interglacial stages a climate warmer than that of the present time, and accordingly it is to be inferred that the ice sheet each time melted away entirely, or nearly so.

The spans of time represented by these alternating glacial and interglacial stages are best gaged with reference to the length of the postglacial interval—that is, the interval that has elapsed since the last glacial stage and that has witnessed the rise of man from the stone age to his present high plane of culture. Whereas postglacial time, according to the best numerical data now at hand, comprises about 20,000 years (the ice melted away gradually, hence there was no sharply defined ending of glacial conditions), each of the glacial stages lasted probably from 50,000 to 100,000 years, and the interglacial stages were of equal or somewhat greater duration. Such, at least, are the estimates given by some of the foremost students of glacial geology.

The glacial record on the western mountain ranges, being less broadly spread out and more subject to destruction by stream action than that on the plains, has proved on the whole much more difficult to decipher. Nor has its study as yet progressed very far. On most of the ranges examined, however, evidence has been found of two distinct glacial stages. In some mountain districts three or even four glacial stages have been recognized. Whether the western mountains have suffered glaciation a less number of times than the plains, or whether the difference is only apparent, owing to the partial obliteration of the morainal record, can not yet be positively stated, but in any event it is certain that on the mountains as on the plains glaciation has been recurrent and discontinuous (except on some of the highest summits), and it is a fair presumption, in the present state of knowledge, that the successive glacial stages on the mountains corresponded to and were synchronous with those on the plains.

There is thus a priori excellent reason to suppose that the Sierra Nevada, which forms an integral part of the great Cordilleran mountain system, has been glaciated two or more times. It is conceivable, nevertheless, that the range forms an exception to the rule because of the lateness of its elevation—that it has suffered glaciation only once, in the later part of the Pleistocene epoch, because in the earlier part it was not lofty enough to bear glaciers. It is not surprising, therefore, that there should have been division of opinion on this question, some geologists taking one view, some the other. Which way, now, does the evidence in the Yosemite region point? Does it indicate a difference in age between the younger and older moraines amounting to only a few thousand years, or to hundreds of thousands of years? The moraines themselves afford no evidence of a kind that would afford a satisfactory basis for time estimates. Fortunately, however, the rock surfaces that were planed down and smoothed by the earlier and later glaciers respectively do afford some evidence of that kind, as set forth below.


The floors and walls of the canyons that were the pathways of the later glaciers appear remarkably fresh and almost unweathered. Over surprisingly large areas they retain their polish and striae and are so smooth and glassy that walking or climbing over them with hobnailed shoes is hazardous, and travel with horses or mules is impracticable. (See pl. 35, B.) In many places, it is true, the polish has flaked off and the surface of the rock is rough, but the deeper scorings and flutings are still visible. (See pl. 35, A.) Elsewhere plates of rock a quarter of an inch to perhaps a full inch in thickness have burst off or are in process of being loosened, but even there the smooth-flowing contours produced by glacial abrasion remain.

PLATE 35.—A (top), GLACIAL GROOVES ON BORDER OF LAKE IN CATHEDRAL PASS. The glacier polish has disappeared for the most part, as a result of intense frost action, but the grooves, being deep, remain in evidence. The ice came from a cirque on the right.

B (bottom), GLACIER POLISH ON FLOOR OF MASSIVE GRANITE. The surface of the rock is scaling off in places as a result of weathering, but much of the polish is still in place and is likely to endure for a long time. Photograph by G. K. Gilbert.

No one who visits the upper Yosemite region or the adjoining parts of the High Sierra can fail to be profoundly impressed by these facts. Few mountain regions, indeed, exhibit glacially worn and polished rock surfaces on a larger scale; few give the traveler a more vivid sense of the recency of the ice age. Scarcely credible does it seem to one viewing the vast expanses of gleaming granite that fully 20,000 years may have passed since the ice age came to an end; that even in the higher parts of the range, where the glaciers lingered long after they had receded from the canyons below, the rock has been exposed to the weather several thousand years.

Two circumstances explain the unusual abundance of glacier polish in the region above the Yosemite Valley—the prevalence of highly siliceous, slow-weathering types of granite, and the generally massive, sparsely jointed structure of those rocks. The superior durability of siliceous granite is strikingly demonstrated in many places where such granite is contiguous to a weaker rock, as, for instance, diorite. The granite as a rule still gleams with glacier polish, whereas the diorite has a roughened and perceptibly lower surface. Veins of hard, fine-grained aplite stand out in relief, like little narrow causeways with level, polished tops, raised half an inch or more above the rough surface of the coarse granite, or granodiorite, which they transect. Doubtless the polish itself in such places helps to accentuate the difference, for it acts in some measure as a protective coating: it promotes the quick run-off of water from the surface, thereby lessening the proportion absorbed by the rock, and it retards the growth of lichens and mosses, thereby lessening the supply of carbonic acid and vegetable acids which result from the decay of those plants and which attack the weaker minerals.

The massive structure of the granite favors both the production and the preservation of glacier polish. Where the joint fractures are spaced far apart—tens or even hundreds of feet—the glaciers can not pluck or quarry individual blocks but work wholly by abrading, and the conditions are most propitious for the development of continuous expanses of even, polished rock; and there also few avenues are available through which water may penetrate to some depth below the surface, and thus the destructive action of such percolating water by hydration, solution, or freezing is reduced to a minimum.

That the glacier polish is by no means equally distributed throughout the Yosemite region and the adjoining High Sierra readily follows from the foregoing considerations. The distribution is controlled by three independent factors—the mineral composition of the rock, the joint structure, and the length of exposure since the retreat of the ice. Thus it happens that in the Yosemite Valley itself glacier polish is on the whole rather scarce, for the chasm was evacuated by the ice soon after the climax of the last glacial stage, possibly as long as 30,000 years ago; and, besides, its walls are made up in many places of jointed rock that was quarried rather than abraded by the glacier. As might be expected, the polish that remains occurs mostly on bodies of extremely durable and massive rock. Some of it, unfortunately, is inconspicuous or hidden from view and consequently is readily overlooked. A few small patches, remarkably well preserved, are on a buttress at the eastern base of El Capitan; other patches occur at the foot of the Three Brothers, at the base of the cliffs under Union Point, on the platform above the Lower Yosemite Fall, on the sides of the Washington Column, and on the walls near Mirror Lake. Glacial grooves remain visible, though the polish has disappeared, on the buttress west of the Royal Arches, on the cliffs east of Indian Canyon, and on the wall below Union Point.

More abundant is the glacier polish on the platform above the Nevada Fall, both owing to the more recent glaciation of that platform and owing to the durability and massive structure of the Half Dome quartz monzonite, of which it is made. Indeed, the tourist as a rule catches the first glimpse of glacier polish when he arrives at the top of the Nevada Fall. The isolated patches he beholds there, however, measure but a few square yards each and are insignificant compared with the larger tracts to be seen on the floor and sides of the Little Yosemite and on the rounded backs of Liberty Cap and Mount Broderick. These tracts, in turn, seem small in comparison with the vast expanses of glacier polish that occur on the broad floor of massive granite of the upper Merced Canyon, above the Little Yosemite. The trail that leads to Merced Lake takes the traveler over this floor and affords him an excellent opportunity to view this unique area of burnished pavements and slopes. (See pl. 36.)

PLATE 36.—GLACIATED FLOOR AND SIDE OF UPPER MERCED CANYON. All the rock features shown are smoothed and polished by the ice. In few places in the world is glacier polish more abundant than here. The row of stones in the foreground serves to mark the trail across the otherwise featureless rock floor.

Almost equally remarkable for its wealth of glacier polish, but far less accessible, is Tenaya Canyon. Indeed, the prevalence of glacier polish on its steeply sloping sides adds greatly to the difficulties which this extremely rugged canyon presents to those who would traverse it. Thus far only a handful of experienced and daring climbers have had the hardihood to pass through the entire length of the chasm.52

52Le Conte, J. N., Scrambles about Yosemite: Sierra Club Bull., vol. 9, pp. 126-135, 1914.

In the upper Tenaya Basin, on the shores and in the immediate vicinity of Tenaya Lake, as well as in the shallow trough above the head of Tenaya Canyon, glacier polish is also very plentiful. Much of it may be viewed by the traveler without effort, either along the automobile road and the trails or in the adjoining areas of gentle rock swells that are easily traversed on foot. It was this abundance of glacier polish in the upper Tenaya Basin that led the Indians to refer to the outflowing stream, now called Tenaya Creek, as Py-we-ack ("river of glistening rocks").


If the glaciated rock surfaces in the area surrounded by the younger moraines seem astonishingly fresh and well preserved, quite the reverse is true of the surfaces in the area surrounded by the older moraines. These impress the observer by their manifold signs of age and prolonged exposure to the weather. They have lost not only their polish but even the smooth-flowing contours that were imparted to them by the glaciers. So roughened are they by the irregular disintegration of the rock and the eroding action of rain water that they do not differ perceptibly from ordinary unglaciated rock surfaces. Indeed, were it not for the fact that they are surrounded by ancient moraines, one would not suspect them of ever having been glacially planed and polished.

In only one locality in the Yosemite region is any of the polish that was produced by the earlier ice still in existence—namely, on the slopes above and below the Wawona Road in the vicinity of Artist Point. Its preservation there is accounted for in part by the exceptionally durable nature of the El Capitan granite, of which the slopes are composed, but mainly by the fact that the rock has long been protected by a veneer of glacial débris. The gradual removal of this now thoroughly weathered débris by rain water has recently brought the glaciated surface to light in several spots. The polish is very imperfectly preserved, in spite of the protection by the débris. As a rule, only bits of it remain on the feldspar and quartz in the granite, the hornblende and biotite being decomposed—etched out, so to speak. Indeed, the surface is so closely pitted that the remaining polish is not readily recognized as such, save under favorable light. In a few places, however, it is sufficiently continuous to permit the direction of the striae to be discerned.

Many of the older rock surfaces are diversified by irregular crags, tables, humps, and pillars that afford some indication of the depth to which these surfaces have been eroded and stripped since glaciation. In general, the depth indicated is not a matter of inches, as in the area of the younger moraines, but of feet. Some of the residual features, being composed of especially obdurate rock, stand 6, 8, or even 10 feet high and are of bizarre or monumental aspect. The crags on Glacier Point, which flank the path that leads to the famous precipice, are good average examples. Others, more fantastically shaped, occur on Yosemite Point, North Dome, Mount Watkins, the Quarter Domes, and the domes of the Starr King group.

Most of the residual crags, unfortunately, are too irregularly shaped to afford an accurate measure of the stripping done since the earlier glaciation, but a few have clean-cut forms produced under rather special circumstances and therefore have more promise than the rest. Such are the rock pedestals that support "perched" glacial boulders here and there. (See pls. 33, B, and 38, A.) These pedestals are composed of the local rock, attached to the body of the mountain, and clearly owe their preservation to the protection which the perched boulders have afforded them. They have remained standing while the surrounding unprotected rock has disintegrated and been stripped away. So eloquent is the mute testimony of these rock pedestals that, not unnaturally, they are commonly regarded as good yardsticks whereby the depth of the stripping may be measured. One author has asserted, even, that rock pedestals afford "of course an exact measure" of the stripping, and he would let it be inferred that their tops are actually remnants of the ancient ice-worn surface. But such, nevertheless, is not true as a generality. It might conceivably be true of an isolated pedestal, but in the entire Yosemite region no such example has been found. Were it generally true that pedestals afford exact measures of the stripping done since their boulders were deposited, then all perched boulders associated with a given moraine, provided they rest on rock of the same kind, should now have pedestals of about the same height, and, other things being equal, boulders associated with very old moraines, situated at high levels, should have taller pedestals than boulders associated with moraines of later date, situated at lower levels. But the facts of observation are quite otherwise.

For instance, of two boulders on Moraine Dome, deposited doubtless at nearly the same time, one has a pedestal 1 foot high, the other a pedestal 3 feet high. Another perched boulder, situated on the upper of the Quarter Domes at a level fully 500 feet above the crown of Moraine Dome, though associated with a moraine that was deposited a long time prior to the two boulders on Moraine Dome just mentioned, instead of having a correspondingly taller pedestal, actually has one only 2 feet high. Again, the perched boulder on the divide east of Mount Starr King (pl. 33, B) has a pedestal only 20 inches high, though it is probably much older than any of the others mentioned, for it is 250 feet above the highest of the older moraines in the Starr King Meadows and presumably attests a third and very early ice invasion. (See p. 73.)

The development of rock pedestals, as a little reflection will show, must be influenced by a variety of factors, and there is therefore need of caution in interpreting differences in the stature of pedestals. In the first place, the rate of growth of a pedestal depends upon the character of the rock of which it is made. One kind of rock is stripped away more rapidly than another, and even the same kind of rock is likely to be stripped away more rapidly in one place than in another, for it is bound to vary somewhat in composition, texture, or structure, and, besides, the rate of stripping is affected by local factors, such as the angle of slope, the direction of exposure, whether northerly or southerly, and the amount of débris that litters the surface. In each of the examples adduced the country rock was the same—Half Dome quartz monzonite—yet because of these different factors the rate of stripping probably was not the same.

Again it would be rash to assume that the development of a pedestal begins immediately upon the deposition of the boulder. If the boulder falls on a spot where the surface configuration is especially favorable, a pedestal may begin to form without delay; but if the boulder falls on unfavorable ground, no pedestal may be formed until perhaps thousands of years later, when the configuration of the rock surface has been changed considerably by erosion. Two boulders situated near each other and deposited at the same time may thus come to have pedestals differing appreciably in age and in height.

It must not be overlooked, further, that rock pedestals are subject to decay, in spite of the protection afforded them by the capping boulders. That protection is, after all, only partial, as the pedestals are more or less exposed to the weather on all sides except the top. Moreover, because they are largely shielded from the sun's rays, they retain whatever moisture they may absorb from rain or snow much longer than the sun-heated rock surfaces roundabout. Consequently, the rock in a pedestal is more subject to chemical decomposition than the neighboring rock and in the course of time will be converted into a weak, incoherent mass that will finally break down under the weight of the superincumbent boulder. Some of the granite pedestals in the Yosemite region are so softened by chemical decomposition that they can be scratched with a knife, or even with the finger nail. At least two of them have recently shed their boulders—one on the summit of Moraine Dome, the other on the low nameless dome on the south rim of the Little Yosemite, northeast of Helen Lake. (See pl. 29.) In each place the boulder still leans against its pedestal; hence the fact of its dethronement is evident.

The life of a pedestal also depends in considerable measure on its diameter, for the greater the diameter the more time will be required for the rock to be softened from the periphery inward to the center. A thick pedestal, in other words, has much better chances for longevity than a slender pedestal. As the diameter of a pedestal is determined in the first instance by the size of the capping boulder, it follows that in general large boulders will remain perched longer than small boulders and will develop taller pedestals. It is significant that the dethroned boulder on Moraine Dome is smaller than any of the boulders that still remain perched and that its pedestal is smaller in diameter and also lower than any of the other pedestals cited. The boulder measures only 3 feet in length, 2 feet in breadth, and 1 foot in thickness, and the pedestal measures only 2 feet in length, 1 foot in breadth, and 1 foot in height. The other dethroned boulder is really larger than some of the boulders that are still perched and has a higher pedestal (it is a slab 7 feet long, 5 feet broad, and 18 inches thick, and its pedestal is 2-1/2 feet high); but the conditions here are probably not typical, for the pedestal is not softened by chemical decomposition and does not appear to have been crushed by the weight of the slab. Perhaps the slab was unbalanced by an unevenly distributed load of snow, or by the caving away of one side of the pedestal, which happened to be weakened by a fracture.

It will be clear that, in general, pedestals can not be relied upon to afford accurate measures of the stripping effected since the earlier glaciation. Probably none of those cited indicate the total depth of stripping, not even the 5-foot pedestal figured in Plate 38, A. Certainly the low pedestals of the boulders situated among the highest and oldest moraines—notably the 2-foot pedestal of the boulder on the upper Quarter Dome and the 20-inch pedestal of the boulder on the divide east of Mount Starr King—represent no more than a fraction of the total stripping at their respective localities. Perhaps the boulders did not begin to develop pedestals until a long time after they came to rest; perhaps each of them is now developing a second pedestal, having been dethroned from the first.

Fortunately there are in the Yosemite region some residual rock features of another type that afford much more accurate measures of the stripping than the pedestals. Of these features, the most instructive stands on the very summit of Moraine Dome, in close proximity to the small dethroned boulder and within a stone's throw of the larger, still perched boulder shown in Plate 37, B, thus permitting direct comparison with their pedestals. This feature (pl. 37, A) may be likened to a stone wall partly fallen in ruins, 7 feet high at the highest point, 15 feet long, and 4 feet thick. It consists of the upper part of a dike of aplite—a sheet of fine-grained granite that invaded a vertical fissure in the coarser Half Dome quartz monzonite of which Moraine Dome is made, as shown diagrammatically in Figure 21, and it owes its prominence to the fact that the aplite, which is exceedingly resistant to weathering, has remained preserved while the surrounding Half Dome quartz monzonite has disintegrated and been stripped away. Only the thickest part of the dike thus stands out in the form of a wall, the thinner parts having tumbled down. The horizontal slabs of which the wall appears to be built are remnants of concentric shells of rock that extended formerly over the entire crown of the dome, as indicated in the diagram, and that have burst loose one after another, by the process of exfoliation, which characteristically affects all large masses of undivided granite in the Sierra Nevada and results in the production of the familiar dome forms.

FIGURE 21.—Section across parts of Moraine Dome showing the features on and near its summit that afford a measure of the stripping effected since the earlier glaciation. A is the 7-foot wall consisting of the exposed upper part of a vertical dike of aplite; B is the dethroned boulder leaning against its pedestal; C is the 8-foot wall produced by an inclined dike of aplite. The features are shown in their true proportions but closer together than they actually are. The broken line GG indicates approximately the original ice-smoothed surface of the dome, and the arrows show the direction in which the glacier moved.

PLATE 37.—A (top), WALL OF APLITE ON MORAINE DOME. The 7-foot wall is formed by a vertical dike of aplite and has remained standing because the aplite disintegrates much more slowly than the surrounding granite. The height of the wall affords a minimum measure of the depth to which the granite has been stripped away since the surface ice passed and smoothed the crown of the dome. Photograph by G. K. Gilbert.

B (bottom), LARGE ERRATIC BOULDER ON MORAINE DOME. The boulder is composed of Cathedral Peak granite, readily recognized by the big feldspar crystals that project its surface. It measures 12 by 6 by 5 feet and is perched on a pedestal 3 feet high. The pedestal consists of a remnant of a shell detached by exfoliation from the body of the dome. G. K. Gilbert standing by the boulder. Photograph by E. C. Andrews.

Now it is not conceivable that the aplite wall was in existence at the time of the earlier glaciation, for the ice then passed over Moraine Dome with a thickness of 500 feet and with sufficient power to raze all such frail projections. Indeed, when the ice withdrew, the dome doubtless emerged divested of all incoherent, disintegrating outer parts, a shining globular mass of hard, sound rock, and the aplite dike was planed off even with the surrounding granite, just as the aplite dikes in Mount Broderick to-day appear planed off as a result of the last ice flood. There can be no doubt, then, that the aplite wall has come to stand out entirely since the time of the earlier glaciation.

Though the wall obviously affords a much more accurate measure of the stripping than either pedestal it nevertheless affords no exact measure, for its top is not a remnant of the original glaciated surface. At least this is to be inferred from the fact that several slabs dislodged from the top of the wall lie at its foot. Just how much the wall has been reduced in height is difficult to estimate, but as it has lost very little in thickness and does not taper upward as a result of progressive disintegration at the sides, it probably has lost but little in height also—perhaps less than a foot.

In any event the height of the wall is to be taken only as a minimum measure of the stripping on Moraine Dome, for that part of the summit on which the wall stands is nearly level and consequently is stripped more slowly by the rain water than the steeper parts round about. It is covered in part with loose granite sand, which acts of course as a protective blanket. On the more steeply sloping parts of the dome, where the rain water runs off with some velocity, residual sand is absent and stripping proceeds evidently at a relatively rapid rate. It is not surprising, therefore, to find that two other wall-like features on the sloping south side of the crown, both consisting of masses of granite preserved under steeply inclined dikes of aplite, have greater height than the vertical aplite wall on the summit. One stands 8 feet high, the other 12 feet.

It may be concluded, then, that 7 feet is a conservative measure of the reduction in height which bare eminences such as Moraine Dome have suffered since the climax of the earlier ice flood. In some localities, doubtless, the reduction has been greater; in others less. The fantastically shaped crags on the rock platform of Glacier Point, on the summits of North Dome, the Quarter Domes, Mount Watkins, and those other heights that were overridden by the earlier ice but not by the later are to be interpreted in the light of this knowledge.

When this profound stripping of the older glaciated surfaces is contrasted with the merely incipient weathering of the younger glaciated surfaces, the difference in their age looms up impressively. Nowhere is the comparison more readily made than on Moraine Dome itself. Below the great morainal embankment that marks the highest level reached by the last ice flood the granite is still essentially unweathered and retains its polish over large areas. In some places the polish is beginning to scale off, but only to a depth of less than an inch. Above the morainal embankment, on the contrary, the granite is deeply weathered and is breaking up into slabs, slivers, and individual grains, the glacial erratics all have pedestals, and the aplite dikes stand out from 7 to 12 feet in height. Though the rate of stripping is difficult to estimate, owing to complicating factors that need not be discussed here, there seems little reason to doubt that the interval of time indicated by the thickness of rock removed is of a much higher order of magnitude than the postglacial interval—at least ten and more probably twenty times as long. If, therefore, the age of the younger rock surface is placed, conservatively, at 10,000 years, then the age of the older rock surface may readily be 100,000 or even 200,000 years. There is thus ample warrant for the conclusion that the earlier and later ice floods in the Yosemite region took place in separate stages of the glacial epoch.

The successive stages of glaciation generally recognized by American glacialists are, in order from the oldest to the youngest, the Nebraskan, Kansan, Illinoian, Iowan, and Wisconsin, named from the States in which their effects are most clearly shown. In view of the comparative freshness of the moraines and polished rock surfaces resulting from the last glaciation in the Sierra Nevada there is little doubt that it corresponds to the latest stage of the continental glaciation, and it will therefore be referred to here as the Wisconsin stage. The earlier glaciation in the Sierra Nevada is more difficult to correlate, as the glacial record on the range has not yet been connected across the intervening mountains and deserts with the glacial record on the plains. To judge by the erosion suffered by the older moraines, the depth to which their boulders are weathered, and the depth to which the domes have been stripped of disintegrated rock—allowance being made for the slower rates of weathering and disintegration in the semiarid climate of the Sierra Nevada than in the more humid climate of the central parts of the continent—the earlier glaciation of the range would seem to correspond to either the Illinoian or the Kansan stage, but more probably to the Illinoian. For the present it will be referred to as the El Portal stage, as the Yosemite Glacier during that stage terminated in the vicinity of El Portal.


In several localities in the Yosemite region glacier borne boulders occur singly, in groups or in rows, without any accompanying fine débris. Most of them lie in places where it seems likely from the character of the topography and from the courses pursued by the ancient glaciers that heavy, continuous moraines once existed. The majority of the boulders are composed of extremely durable types of rock, such as quartzite and highly siliceous granite, which weather and disintegrate more slowly than most of the rocks in the moraines of the Yosemite region. It seems entirely probable, therefore, that these boulders are the last remnants of moraines of a very early glaciation antedating the El Portal stage.

Were these boulders all situated on slopes of bare massive granite this inference as to their great age might seem open to question, for on such slopes, where the washing action of rain is particularly effective, a moraine may be reduced by the removal of the finer débris to a mere row of boulders in a relatively short time. Many moraines of the Wisconsin stage, even, are thus reduced, as may be seen on the undulating platforms of bare, smooth granite that flank the upper gorge of the Merced, above the Little Yosemite. The erratic boulders here referred to, however, lie for the most part on flat or very gently sloping ground, where the washing action of rain water is but moderately effective. The conclusion seems therefore justified that the boulders are remnants of very ancient moraines.

A number of the boulders in question lie on the broad divide east of Mount Starr King, between 200 and 400 feet above the highest moraines of the El Portal stage, which curve around the Starr King meadows. All are derived from Mount Clark or its immediate vicinity. Many of them consist of the light-colored siliceous granite of which Mount Clark itself is composed; others consist of the yellowish quartzite whose parent mass is on the northern spur of Mount Clark—the same quartzite that is represented also in the moraines back of Glacier Point. One boulder, perched on a pedestal 20 inches high (pl. 33, B) consists of a relatively dark but also siliceous granite from an adjoining part of the Clark Range. The general direction of the ice movement is thus plainly indicated by the boulders, and it is evident from their derivation and from their arrangement in an east-west belt that the Merced and Illilouette. Glaciers at one time met and coalesced on the divide, leaving there a more or less continuous body of moraine. The fact that only a few sparse boulders now remain, especially on the flatter portions of the divide, where the ground is favorable for the preservation of moraines, would seem to show that a long period has elapsed since the glaciers met on the divide, a much longer period than has elapsed since the moraines about the Starr King Meadows, which are still fairly continuous and in places massive, were laid down. It seems entirely in order, therefore, to refer the erratic boulders on the divide to a stage of glaciation earlier than the El Portal.

The highest level reached by the ice of this early stage can scarcely be determined with exactness, but two erratic boulders of Mount Clark granite, lodged in the saddle between Mount Starr King and the lesser dome to the northwest, attest in any event that the ice rose high enough to spill through the gaps in the Starr King group.

Other erratics belonging in all probability to the same early glaciation lie near the east base of Sentinel Dome, at altitudes between 7,800 and 7,900 feet. (See p. 63.) Some of them, conspicuous by their large size (pl. 38, B), are close to the trail that leads to the dome. They are strung out at intervals of 100 feet or more in an irregular line that curves from the dome southeastward to the north end of the Illilouette Ridge. Their exact positions are indicated on the map of glacial and postglacial deposits forming Plate 29.

PLATE 38.—A (top), GLACIAL BOULDER PERCHED ON 5-FOOT PEDESTAL. This perched boulder has the highest pedestal in the Yosemite region. It is situated on the mountain west of the upper Yosemite Fall. The pedestal is composed of slate of the local rock—remnants of concentric shells that formerly enveloped a large part of the summit. Photograph by F. C. Calkins.

B (bottom), ERRATIC BOULDER AT BASE OF SENTINEL DOME. A row of such boulders marks the highest level reached by the ice in the vicinity of Glacier Point. They are the sole remnants of a very ancient moraine, the rest of which has long since disappeared and they are believed to record a stage of glaciation that antedated the El Portal stage. The boulder here shown was angular when deposited by the ice and has become round by long-continued exfoliation. Photograph by F. C. Calkins.

All these boulders lie on gently sloping ground that is clearly more favorable for the preservation of a moraine than the steep slope immediately below, yet they are isolated, unconnected by finer glacial débris, whereas the steep slope below is still heavily cloaked with such material of the El Portal stage. It can hardly be doubted, therefore, that the boulders are the last vestiges of a once continuous moraine that was deposited by the ice during a stage much earlier than the El Portal. This early stage, which may correspond to either the Kansan or the Nebraskan stage of the continental glaciation, will here be referred to as the Glacier Point stage.

It happens that these boulders are composed not of distinctive rock materials derived from the High Sierra, but of a granodiorite somewhat similar in appearance to the granodiorite in the slope on which they rest, and consequently their status as ice-borne erratics might seem open to question. Close inspection, however, reveals a decided difference between the two rock types, although unquestionably both are varieties of one and the same intrusive, the Sentinel granodiorite. (See appendix.) The rock in the slope is distinguished by roughly parallel black streaks composed of minute crystals of hornblende and indicative of the flow structure of the intrusive mass; the rock of the boulders is wholly devoid of streaks, is somewhat coarser grained, and contains ragged flakes of biotite (black mica), many of which measure fully an inch in diameter.

The origin of the boulders seemed puzzling until it was discovered by Calkins that granodiorite of precisely the same variety as that in the boulders occurs in the ledges at Glacier Point and south of it along the edge of the precipice below the automobile road. These ledges are three-quarters of a mile east of the base of Sentinel Dome and from 500 to 700 feet below its level, and it would thus appear that the boulders had been carried by the ice up the slope to their present positions. This explanation may seem daring, but it is entirely in accord with the facts known about the direction and character of the ice movement. The Merced Glacier impinged with a strong westerly current against the Glacier Point promontory and, as it rose to higher and higher levels, must have carried the rock débris up with it.

The rounded forms of some of the boulders would perhaps seem to belie this explanation, for blocks torn from a ledge by a glacier are naturally angular, and in the short distance here involved angular blocks could not have been worn down to rounded boulders by abrasion. The indications are, however, that the boulders in question have become rounded in place by exfoliation, for fragments of curving scales lie about their bases, and a few such scales still cling to their surfaces. The almost complete elimination of the original angularities, as shown in Plate 38, B, would seem to imply that exfoliation has been active for a long period of time. The very roundness of the boulders, therefore, affords independent proof of their great antiquity and strengthens the supposition that they date back to an earlier stage of glaciation than the El Portal.

Boulders that presumably are indicative of this very early glaciation occur also near the summit of Indian Rock, about 1-1/2 miles north of Basket Dome and considerably above the upper limit of the more or less continuous veneer of glacial débris of the El Portal stage that covers the lower slopes. Most readily identified as true glacial erratics are those boulders which are composed of distinctive porphyritic granite from Mount Hoffmann.

At the lower border of the Yosemite region also there are dim indications of a glaciation that antedated the El Portal stage. These indications are found in the obscure form and advanced state of decay of the outermost of the concentric moraines that encircle the Big Meadow flat. Not only has the outer most moraine suffered more from erosion than the others, but the boulders in it are decomposed to the consistency of sand or loam, so that they are readily cut with a pick or shovel. Many boulders thus sliced through are to be seen in the road cuts west of Big Meadow, each outlined in cross section by a rusty ring. Similar sliced boulders occur in the road cuts near McCauley's ranch, about a mile southwest of Big Meadow. They show that the outflow of ice from the flat toward El Portal took place largely, if not wholly, during the earliest of the three glacial stages.

There remains to be considered the indirect but nevertheless highly significant evidence afforded by the capacious U form of the Yosemite Valley itself (except for the filling of sand and gravel in the basin of ancient Lake Yosemite). That U form is a typical product of glacial erosion and must have required for its elaboration from the essentially sharp-cut V form of the preglacial canyon a tremendous amount of excavation. The completeness of the transformation is patent from Figures 24-27, in which the preglacial cross proffles of the valley are superimposed on the postglacial profiles.

Now, it is evident from the mere fact that the Yosemite Glacier of the Wisconsin stage ended within the valley and, further, from the fact that the valley (disregarding the talus) is about as wide below the group of frontal moraines as it is above, that the glacier of the Wisconsin stage did not perform any significant share of the excavating. This glacier found the valley presumably almost as deep and as broad as it is to-day and did little more than reoccupy it. Only in the upper part of the valley, where it attained a thickness of nearly 1,500 feet, did the glacier effect any appreciable changes. Probably it plucked off shells from the Royal Arches, rounded off a bit more the promontory under Glacier Point and added somewhat to the depth of the basin of ancient Lake Yosemite. But it is not to be supposed that this small and relatively shallow ice stream gouged out wholly; or even largely, a rock basin whose depth exceeded 200 feet, as would appear to be indicated by cross sections of the valley. (See figs. 24 and 25.) The excavation of that basin clearly must have been accomplished by a much larger and more powerful glacier. It follows, then, that the Yosemite's capacious U form was produced very largely by the earlier ice.

So thoroughgoing, however, was the transformation, so enormous were the quantities of rock removed, that this work can scarcely be attributed to a single glaciation. Even when the great thickness indicated for the earlier ice is taken into consideration (it measured about 3,000 feet at the head of the valley and thence declined to 2,400 feet at the lower end), and when allowance is made for the special rock structures in the valley that facilitated the excavating action of the ice, it seems much more probable that the work was performed by two ice invasions, if not more than two. On this score also, then, there appears to be good warrant for supposing that at least two ice invasions took place prior to the Wisconsin.

The probability that the Sierra Nevada has been glaciated three times was first suggested by Prof. I. C. Russell, who in 1882 and 1883 studied the moraines built in Mono Valley by the glaciers that descended from the east flank of the range. The glacier that issued from Bloody Canyon, he believed, had made "three advances and three corresponding recessions," indicating "climatic oscillations of considerable duration."53 Willard D. Johnson,54 who in 1905 and 1906 investigated the glacial deposits in the Bridgeport Valley and along the West Walker River, went further and recognized, tentatively, a record of three stages of glaciation differing markedly in age. Still more significant, as affording corroboration of the conclusions here set forth in regard to the occurrence of at least three glaciations in the Yosemite region, is the finding in 1927 and 1928 by Prof. Eliot Blackwelder55 of evidence of three stages of glaciation at several points along the eastern base of the Sierra Nevada. More recently he has found what apparently proves to be evidence of a fourth stage.56

53Russell, I. C., Quaternary history of Mono Valley, California: U. S. Geol. Survey Eighth Ann. Rept., p. 340, 1889.

54Unpublished notes in the possession of the U. S. Geol. Survey.

55Blackwelder, Eliot, Evidence of a third glacial epoch in the Sierra Nevada [abstract]: Geol. Soc. America Bull., vol. 39, p. 268, 1926.

56Blackwelder, Eliot, Glacial history of the east side of the Sierra Nevada [abstract]: Geol. Soc. America Bull., vol. 40, p. 127, 1929. Professor Blackwelder has also made an attempt to identify these four stages of glaciation in the Yosemite region, but his findings could not be verified in time for incorporation in this report.

<<< Previous <<< Contents >>> Next >>>

Last Updated: 28-Nov-2006