PROCEEDINGS
of the Tomsk Technological Institute
of Emperor Nicholas II.

Vol. 9. 1908. No. 1.

I.

S. P. Peretolchin.

GLACIERS OF THE MUNKU-SARDYK RANGE.

With 10 phototype plates, 3 tables, and a map.

I–VI, 1–47.

Contents

	Page
Preface.
Historical information and literature	1
Route of travel to the glaciers of the Munku-Sardyk Range	3
Southern glacier of the main summit of the Munku-Sardyk Range	5
Northern glacier of the main summit of the Munku-Sardyk Range	16
Glacier at the source of the Bely Irkut River	28
Meteorological features	32
Conclusion	37
Table of absolute elevations	41
List of plants collected on the southern slope of the main summit of the M.-S. Range	43

The proposed study of the glaciers of the Munku-Sardyk Range, situated on the border between Russia and Mongolia, is the result of my work during time free from official duties.

These studies were begun by me in 1897 with the participation of Lieutenant Colonel of the General Staff E. M. de Genning-Mikhelis, and with some material assistance from the East Siberian Branch of the Imperial Russian Geographical Society.

In the following years: 1898, 1900, 1901, 1902, and 1903, in all—two months, I worked on these glaciers alone and during all this time received from the East Siberian Branch of the Geographical Society a subsidy of 190 rubles. All the remaining expenses for equipping the excursions were borne by me from my own, and very modest, means.

The attached map was prepared, in its essentials, on the basis of the existing map of the frontier zone of Asiatic Russia, but it has been corrected and somewhat supplemented according to the survey of E. M. de Genning-Mikhelis and according to my own notes. The plan and section of the southern glacier of M.-S. were made from the work of the same E. M. de Genning-Mikhelis; the plan of the northern glacier and its section—from my own work. All phototypes are reproduced from my negatives. The hypsometric data in the appended table were obtained with instruments that had been previously checked at the Main Physical Observatory, and subsequently in Irkutsk.

The calculations of absolute heights for 1896 and 1897 were kindly carried out by the director of the Irkutsk Observatory, A. V. Voznesensky. The heights determined in later times were calculated by me with reference to the nearest meteorological stations.

I consider it my duty to express my sincere gratitude to the Mining Department of the Tomsk Technological Institute, which gave its consent to the publication of my work, and also to Professor V. A. Obruchev, to the director of the Irkutsk Observatory A. V. Voznesensky, to the veterinarian P. E. Naletov, and to the observer of the Mondy meteorological station G. T. Sizykh, who in one way or another contributed to the success of my excursions and work.

Student of the Mining Department
of the Tomsk Technological Institute, S. Peretolchin.

Village of Tunka. 29 July 1906.

Historical Information and Literature

In proceeding to a systematic account of my investigations of the glaciers of the Munku-Sardyk Range, I shall cast a glance at the historical past of this question.

Among the learned travelers who first penetrated across the Russo-Chinese border to the main summit of the Munku-Sardyk Range, mention must be made of the well-known botanists Kirillov and Turchaninov. They visited this locality as early as 1836 for botanical purposes and did not touch upon the question of glaciers. After 1836 there is a large gap in our scientific investigations of the border belt with China. The chief reason for this may be considered the distrust and hostile attitude toward us of the Chinese and of the Mongol population adjoining the frontier.

The Imperial Russian Geographical Society¹ even then directed its attention to the investigation of the border belt and likewise mentioned the distrust of the Chinese.

Despite these difficulties, in 1857 Permyakin,² an associate member of the Geographical Society, crossed the border and descended into the valley of Lake Kosogol. Mr. Permyakin’s brief report is devoted chiefly to ethnographic information and does not touch upon glaciers at all.

In 1859, investigations to the south of the Munku-Sardyk Range, as far as the border Mongolian guard posts, were proposed to the naturalist Radde; but obstacles encountered on the part of the Chinese authorities forced him to confine himself to a survey only of the Russian limits. Nevertheless, on July 12, 1859, Radde³ made an ascent of the main summit of the range along its southern slope, from the side of Mongolia. The materials he collected on the glaciers entered Russian geographical literature and from there into the well-known work of Élisée Reclus.

In 1871, A. Chekanovsky visited the southern glacier of the main summit of the M.-S. range¹. G. N. Potanin² mentions only the summit of Munku-Sardyk and does not touch upon the question of the glaciers.

In 1887 G. Makerov³ made an excursion to the Sayan Range, but in his report he says nothing about the glaciers of the M.-S. range.

In the same year an expedition to the Sayans took place under the direction of Lieutenant Colonel of the General Staff N. P. Bobyr⁴, accompanied by the botanist Ya. P. Prein and the mining engineer L. A. Yachevsky. This expedition made an ascent to the southern glacier of the main summit of the Munku-Sardyk Range. The results of the expedition were printed in the report and entered into Ritter’s well-known work, The Geography of Asia.

All the works listed concern exclusively only the most accessible southern glacier of the main summit. The northern slope of this range, lying within Russian limits, remained from this side entirely unexplored, unless one counts certain general indications by the same Mr. Radde, made by him from a bird’s-eye height.

In 1896, visiting as a tourist the summit of Munku-Sardyk from the side of Mongolia, I twice (on June 26 and July 3) climbed to its highest point, made some possible observations of the southern glacier, and in the autumn gave a small report in the East-Siberian Section of the I.R.G.S.

In 1897 the East-Siberian Section of the Geographical Society decided to equip an expedition for the investigation of the glaciers of the Munku-Sardyk Range and Lake Kosogol, proposing to me and to Lieutenant Colonel of the General Staff E. M. de Henning-Mikhelis that we undertake this labor. The Section of the Society supplied us with instructions and issued a monetary allowance in the amount of 300 rubles. The results of this expedition were presented in two separate reports to the East-Siberian Section of the Geographical Society in the spring of 1898, after which the report of E. M. de Henning-Mikhelis⁵ was printed in the Proceedings of the Section. I at that time declined to print my own report on the grounds that many of the results of our hurried investigations proved not especially precise, were erroneous, and required thorough verification; moreover, I had the firm intention of continuing the work begun.

The Route Followed to the Glaciers

Having chosen as the central station for my final preparations the Mondy missionary camp, located in the valley of the Irkut River, near the Russo-Chinese frontier, I consider it unnecessary to describe the route from the city of Irkutsk to this point, since that road is generally known and not difficult.

The Mondy missionary camp (Table I), where a meteorological station (1349 m) has been established since 1900, is the last cultural and administrative point in the valley of the Irkut River, where one can find Russians, renew supplies of provisions and equipment, and make final preparations for further work in the remote and virgin taiga, under the open sky.

In order to proceed from the Mondy camp to the part of the southern glacier of the Munku-Sardyk Range that interested us, it is necessary to cross the Irkut River (by ferry, and at low water by ford on horseback) and continue farther along the mountain trail toward the frontier.

The road from the valley of the Irkut River begins with a rather long, steep, and stony ascent, crossing the lowered part of the Sayan; farther on, after crossing a hollow, there begins again an easy ascent onto the next spur of the same part of the Sayan. The path worsens owing to the presence of bogs; the soil here is covered with fragments of mountain rocks, partly half-hidden in the vegetation cover. Following the valley of the headwaters of the Irkut Khanga, the trail reaches the highest point of the Sayan spur at boundary marker No. 16 (1818 m).

From this point the mountain plain has a slight slope to the southwest and serves as the valley of the Khanga River, belonging to the basin of Lake Kosogol. From the boundary marker the trail descends in three directions: along one—somewhat to the southwest, by the valley of the Khanga River—it comes out to the Khangin datsan on Lake Kosogol; along another—somewhat to the right—it descends near the mouth of the Khabsyl River to the same lake; and, finally, along the third—it first goes west, descends to the Sagan-Sair River, follows its valley, crosses the stream and, gradually ascending the eastern slope of the Yangit spur, crosses it, descends by the valley of the Bain-Gol River (the old name is Uryum-Sair) to the northern end of the same lake.

Comparing the last two directions, it must be noted that in spring and during rainy times of the year the second route through Yangit is not entirely convenient because of the stony and very muddy trail; the direction toward the mouth of the Khabsyl River, although longer than the second, is considerably better in quality.

Reaching, in one way or another, the northern shore of Lake Kosogol (1638 m), from which the glacier that interests us lies at a distance of about 10 versts to the north, I consider it not superfluous to say that, as a convenient place for a good camp, in terms of fodder for horses and the best conditions for the traveler, one must recognize the valley of the rivers Zhergalanta and Khushi, where the expedition of 1897 was stationed, and where I also stopped subsequently. The remoteness of this lower station from the very point of work compelled us in 1897 to have yet another, intermediate or upper station, for which a place was chosen beneath the principal summit of the Munku-Sardyk range, at the edge of the forest, at an elevation of 2260 m.

Under these conditions, communication between the lower and upper stations is possible on horseback, without particular difficulty, along the valley of the river Khushi and the western slope of the southern spur of the principal summit of the range.

Returning back, I shall say a few words about the Munku-Sardyk mountain chain. The Munku-Sardyk range forms part of the natural boundary between China and Russia. From the northern shore of Lake Kosogol it appears as a long chain of mountains, extending up to 13 versts, with a direction from east to west, and consists of six sharply distinguished barren massifs, among which the most significant in height and mass is the first at the eastern end—the principal summit of the range. Toward Lake Kosogol the mountain chain descends in a fairly gentle foothill with soft outlines, which serves as the combined valley of several mountain streams flowing into the lake; in its western part the foothill is separated from the main chain by the small valley of the upper course of the river Zhergalanta and Lake Shera-Nor. Representing the culminating point of the uplift, the principal summit of the range seems, thereby, to serve as the center of three systems—the Tunkinsky Alps, the Sayan, and its own.

Looking attentively at the picture of the range (Plate II), it is not difficult to notice that its foothill proves to be considerably raised toward the cross-space, on the boundary between the principal summit of the chain and its second spur, being cut through at this place by the rather wild and narrow gorge of the river Khushi; along this line there seems to have occurred a fracture of the range, and a fissure was formed—the actual gorge of the river Khushi.

The genetic connection of the Munku-Sardyk chain with the Tunkinsky Alps and the Sayan is indicated in part by the similarity of the petrographic composition observed in outcrops on both sides of the valley of the Irkut River, near the Mondy missionary station, and also upstream and downstream along the valley.

Southern Glacier of the Main Summit

Translated from the Buryat and Darkhat languages, Munku-Sardyk, or Munku-Khardyk, means “Eternal Golets.” Sometimes the same summit is called “Munku Sagan,” i.e., “Eternally white.” The Uriankhai call it “Buren-Khan,” “Burion-Khash.” The Mongols call it “Munku-Zasu,” i.e., “Eternal snow.” On the basis of all these varied names, and noting that each of them is somewhat incomplete in the sense of characterization, it would be more correct to retain for the main summit of the range the name “Munku-Sagan-Khardyk” or “Munku-Sagan-Sardyk,” which in translation means “Eternally white golets.”

Dominating, by its considerable height and dimensions, the surrounding mountain country, the first golets of the range becomes visible at a very great distance from the valley of the Irkut River. The grandiose appearance of this massif is especially distinguished when viewed from the Mondy stan, despite the distance of more than 20 versts.

From the pass across the Yengit, the main summit appears as a relatively low conical mass, gradually rising from the valley of Lake Kosogol and bearing at its top an oval field of “eternal snow” (Plate III, fig. 3).

To the west, beyond the main summit, the other golets of this chain can be seen from here; to the east, the center of uplift gives off a spur, named in 1897 the “striped mountains.”

From the boundary of the forest—the site of the base station—the main summit has a folded-pyramidal appearance (Plate III, fig. 4), gives off several lateral spurs—counterforts—and is set back from the northern shore of Lake Kosogol somewhat to the north-northeast. The gentle southern slope of the summit, giving place to two stony screes, has the character of a mountain plain, descending likewise gently in two directions—to the west, into the united valley of the Khushi and Zhergalanta rivers, and to the east, into the corresponding ravine of the Bain-Gol River. Both the foothills of the range and the southern slope of the main summit, beginning almost from the shore of Lake Kosogol, are covered with the predominant representative of the forest—larch (Larix sp.)—and already at an elevation of 2260 meters lose it completely, gradually turning into a landscape with vegetation of an alpine country.

Having left the base station, I began my ascent of the main summit first in a direction somewhat toward the NE, crossed a stony scree, and continued my way in a northerly direction, along the very highly raised southern counterfort of the summit, descending by very steep slopes into narrow valleys—

ravines of the neighboring spurs. Above the ice boundary, on my route, small and bright-green turfs of tiny alpine plants were encountered; at first they were supplemented here and there by stunted forms of Betula, Rhododendron, Juniperus, Salix, Potentilla; these, as it were accidental, refuges of plant life, poorly adorned with modest flowers and scattered among the stones, occurred ever more rarely, imparting a distinctive and melancholy cast to the mountain scene.

The southern spur and the slopes of the main summit are entirely covered with fragmental material from the rocks: granite, syenite, gneiss, crystalline schists, and a small quantity of basalt and lava.

Often resting along the way, I observed the gradually dying plant life and, finally, after four hours of fatiguing ascent, was at the boundary of the clearly visible plants, and a little farther on reached the lower edge of the southern glacier. The boundary of flowering plants, as repeated barometric observations and calculations showed, lies on the southern slope of the main summit of the M.-S. Range at an altitude of 3174 m. Despite the severity of the living conditions and surroundings at this limiting height, almost at the level of permanent ice and snow, five species reach it: Papaver alpinum L., Draba sp., Alsine sp., Cerastium lithospermifolium Fisch., Sedum quadrifidum Pall. At the end of the present work is appended a list of the plants collected by me on the following excursion, beginning from the ice boundary up to the highest and limiting point of the ascent. The plants were kindly identified by the botanists B. A. Fedchenko, A. A. Elenkin, and V. F. Brotherus.

The upper part of the summit of Munku-Sardyk has the appearance of a crater-like depression, a basin or trough (Plate IV, fig. 5), now filled by the field of the southern glacier and ending in a small outlet in the direction SO, toward the narrow ravine of a tributary of the Bayn-Gol River, between the southern and southeastern counterforts of the main massif.

The basin or trough has the form of an irregular hexagon and on the south is bounded by a rampart of comparatively small rock fragments, which, rising gradually and merging, passes into the western granite ridge with clear signs of its destruction. On the northwestern side this ridge rises to the highest point; at its northern boundary it is hidden in places under saddles of ice and snow, and again becomes exposed at the eastern end.

Without stopping yet at the glacier, I moved farther on in order to determine the absolute height of the summit. I walked along the ridge of the trough,

using the accumulation of very massive and sharp-edged fragments of granite; already beyond the limits of the higher plants, these fragments are almost bare and only partly covered with certain species of mosses and lichens, whose life does not cease up to the highest point of ascent.

Under the influence of atmospheric agents, from the action of water and ice, the summit of the ridge is constantly subject to destruction, and the products of this unceasing work roll down from here into the lower-lying gorges, valleys, or onto the field of the glacier.

The highest, extreme northwestern point of the ridge of the basin, which is also the main summit of the Munku-Sardyk Range, occupies an area of about 2 square meters amid a heap of large blocks of granite. The absolute height of this point of ascent, according to my observations, is 3453 m. From this highest point there opens a splendid view to all surrounding sides of the horizon: of Lake Kosogol and its island, of the chains of neighboring and distant mountains, of rivers, lakes, hollows, and valleys... Unfortunately, this picture does not always await the tourist or traveler striving toward it. Owing to the peculiar meteorological conditions, of which there will be discussion later, it is very rarely possible to reach this proud summit under good conditions: fog, rain, snow, hail, thunderstorm phenomena, and strong gusts of wind, which nearly knock one off one’s feet, are close companions of the summit of Munku-Sardyk, and they part from it very rarely.

It must also be noted that one of the conditions poisoning the aesthetic feeling of the traveler, after the arduous six-hour trials of the ascent, is not infrequently the smoke of forest fires in our Siberian taiga—a very sad result of our Russian culture.

I turn to the description of the glacier. The field of the southern glacier is a continuous ice cover filling the above-mentioned basin of the summit. In its western part the glacier is bordered by a granite crest, which in its northern part seems to disappear, being covered along this stretch by a saddle of ice (Table VI); farther on the crest again becomes exposed, runs from west to east, and passes to the summits of the eastern spur of the central point, while the ice cover descends along the southern slope of the crest and covers the whole space down to its lower boundary—the ridge of the hollow, which has turned into a fairly considerable rampart of fragmental material. Owing to the fact that the northern and northwestern parts of the hollow are raised considerably more than its southern and southeastern edge, the ice field has the appearance of a cover concave into

in a twofold direction, i.e. from north to south and from northwest to southeast. The greatest inclinations of the field are observed on its western and northern sides, and these inclinations vary, respectively, along the two directions indicated above. The slope from south to north increases gradually from \(10^\circ\), then diminishes somewhat, afterward increases rapidly, and at the limit reaches \(40\text{–}45^\circ\), if the last determination, owing to the inaccessibility of the ridge, is made on the basis of graphical constructions. The slope in the direction from the source of the stream to the highest point of ascent is at first about \(15^\circ\), then changes, evidently following the relief of the bottom of the hollow, and, finally, at the northwestern boundary is about \(20^\circ\).

As for the size of the area occupied by the southern glacier of Munku-Sardyk, I shall, for comparison, cite the results of earlier work.

In Radde’s report, in the Bulletin of the Geographical Society for 1860, on p. 166, we read: “To reach the highest point of this summit one must cross a broad glacier extending for two versts.” The expedition of N. P. Bobyr in 1887 gives the area of the glacier as 1.5 square versts. Basing himself on the survey data of E. M. de-Henning-Michaelis in 1897 and on his own notes, I estimate this surface area at about 0.4 square kilometer. The last of the figures cited by me I consider the more probable, since it was taken from the results of a survey—the most accurate method of such measurements.

Having ascended to the glacier on 30 June 1897, I found on its surface a continuous snow cover. The glacier was quite calm and aroused in the observer a certain sad feeling by its lifelessness. The snow itself was firm on the surface and freely allowed travel on foot; in places, on the snow, there were scattered cherry-red patches of accumulations of the microscopic alpine alga Haematococcus nivalis; in various places of the field, large and small fragments of rock, fallen onto the field from the ridge surrounding the glacier, protruded from beneath the snow; the main runoff of glacial water, or the head of the right tributary of the Balyk-Gol River, in the southeastern corner of the field, was still covered with ice. Such was the picture at the end of the month of June.

In the first days of July the upper snow had already begun to melt noticeably, turning into a fine-grained, coarse mass, almost saturated with water; separate rivulets, flowing down from the icy surface, carried with them snow moistened with water, reached the southern edge of the field, and from there, in the form of a common, very rapid and muddy stream, rushed toward the exit into the narrow gorge.

In the southwestern part of the crest the glacial water has also cut a path for itself, but this runoff is insignificant and has no name.

The ice cover, laid bare of snow, had the appearance of a grayish-white surface, covered in various places with large and small stones. Since near the western, northwestern, and northern parts of the crest the slopes are rather considerable, the greatest quantity of fragmental material rolls down to the lower boundary of the field; small fragments, being held back by the irregularities of the ice and heated by the rays of the sun, sink into the ice mass and are often completely hidden within it; large stones either sink only partly or lie quite freely on the surface, thereby revealing their very recent appearance there.

Cosmic and mineral dust, falling on the surface by means of the wind and transported by atmospheric precipitation, is observed here in the form of gray and almost black spots, scattered very irregularly over the glacier and accompanied in places by the aforementioned alga Haematococcus.

Numerous little streams of water, running over the surface of the ice along the lines of the natural and greatest slopes of the field, as it were dissolve the ice beneath them and as a result form parallel furrows. Since the mineral particles and dust lying on the surface, when heated by the sun, penetrate in one way or another into the ice, the result is the peculiar pitted surface characteristic of a glacier, with glacial cups scattered over it. In their transverse dimensions the cups were up to 3 centimeters, and in depth reached 5 and more.

When the contents of these little pits were examined under the microscope, they consisted of mineral dust, filamentous fungi, spores of lichens, the alga Haematococcus, pollen of coniferous plants, and even the alga Pleurosigma.

Examining the glacial field attentively, it is not difficult to notice on it certain of its features (Table V, fig. 7).

1. In the northwestern part, on the steep slope of the field, perpendicular to the line of greatest inclination, the presence of a parallel-directed striation was found. On the northern slope of the field the same phenomenon is observed.

2. On the surface of the entire glacier, beginning almost from the lower boundary of the ice and rising fairly high upward, the presence is noticeable of a very interesting, parallel-running banding, directed from the western edge of the field toward the eastern and facing with its convexity toward the lower edge of the glacier.

1. On the steep slope of the northwestern part of the field, parallel-running crevasses are clearly visible (Plate IV, fig. 6).

2. Along the middle of the field, crevasses are observed that are not especially considerable in length, from 3 to 8 centimeters wide, filled with solid ice and intersecting at various angles.

3. The surface of the glacier does not correspond to the slopes and to the presumed form of the trough itself. In consequence of this it is raised in places and lowered in places.

Examining the ice, which is exposed in the lower part of the field, above the glacier’s water stream, in a comparatively small thickness, I found the following. The upper snow, wetted by water, has the appearance of grains the size of sago groats—this is a loose mass of upper firn; the latter, gradually increasing, passes into grains of larger size, forming, in aggregate, a denser mass of lower firn, which yields comparatively easily to mechanical efforts. This mass is semi-transparent owing to the presence of air bubbles. Beneath this lower firn follows firn ice—a denser and more transparent mass which, when struck with a hammer, breaks up into separate rounded grains, ranging in size from a hazelnut to a pea, resembling topazes of the purest water. The last stage in the transition of firn ice into typical blue-colored glacier ice I did not observe here in the exposures accessible to me. Over the entire surface of the glacier I did not find any sufficiently considerable crevasse that would allow me to investigate the ice mass in depth and to form a clear idea of its properties and thickness.

On the basis of the survey of the glacier and taking into account the excess of the crest of the trough above the lower end of the field, it is possible to imagine its ideal section along the line of greatest slope from NW to SO—A. B. (see the plan of the southern glacier of Munku-Sardyk, Plate XII). According to these data, the probable thickness of the ice should not exceed 65–70 meters.

If, however, one takes into account the irregularities of the bottom of the trough, according to the above-mentioned signs of the irregularity of the slopes of the field and the existence of a small exposure protruding near the southeastern end of the field from beneath the glacier, then the thickness of the ice under these conditions can hardly be more than 30 meters.

Corresponding to the position of the highest point of rise of the crest of the Munku-Sardyk massif, the height of the upper boundary of the ice of the southern glacier will be the same, i.e. 3453 meters.

The lower boundary of the ice lies somewhat below the limit of clearly morainic accumulations and is expressed by an elevation of 3173 meters.

Having made a preliminary description of the observed picture of the glacier, I shall try to resolve two questions:

Does the southern glacier have an independent area of nourishment?
Is this glacier moving or not?

Examining the glacier from the summit of the northwestern part of the ridge, I discovered two small platforms, of which one is inclined somewhat to the southeast, the other to the northwest. Lower down these platforms gradually pass into the corresponding ice fields of the southern and northern slopes of the summit M.-S., without presenting any distinguishing features. Thus, from all the evidence, it may be said that the southern glacier has no special area of nourishment, i.e., no separate firn basin. How, then, is the glacier formed and replenished?

Assuming that the observed and above-mentioned parallel striation of the slopes of the field is the result of the formation of ice from layers of snow falling near the summit of the ridge, we shall have the following. Since the falling layer of snow, owing to gradual melting, under certain conditions yields as a result two layers—the lower layer of cleaner, denser ice, and the upper one of white ice, with abundant content of air bubbles, mineral and organic dust falling upon the surface of the melting snow—it is clear that, as a result of the gradual accumulation of these double layers, a picture of parallel striation will appear. Being at first horizontal and adjoining with one of their edges the surface of the steep slope of the basin, the layers overlie one another and move downward under the influence of weight and pressure. The reality of this displacement of the ice strata has already been established by cracks in the northwestern part of the field.

In place of the layers of ice that have descended lower down, an accumulation of new ones takes place at the summit of the ridge, by which the continually occurring loss is replenished. Since the inner edge of the ice strata must, in its movement, be directed in accordance with the surface of the bottom of the basin, which is curvilinear, it is clear that strata initially arranged horizontally will gradually deviate from their position, follow the curve of the basin relief, and, having reached the lower points of its bottom, will rise with their outer edge.

Lower down the glacier, near the southern edge of the basin, the alternating strata, under the influence of the most intense melting and its dependence on the different density of the ice, produce the phenomenon of the above-described parallel banding, clearly visible even in the illustration (Plate V, fig. 7).

In order to verify my propositions on this interesting question, in the summer of 1898, while on the glacier, I made a vertical cut across the strike of the parallel banding. On the vertical wall of the cut I found that the observed banding is the result of the emergence at the surface of alternating layers of white and blue ice (Plate XI, Fig. 1). The thickness of the layer of white ice was equal to 14 centimeters, and that of the blue to 22 centimeters. These strata, extending into the depths of the mass, are arranged alternately, in parallel; in their strike they correspond to the observed banding and dip, at the place investigated, at an angle of about 20° toward the crest of the ridge.

The observed position of the alternating layers cannot be natural under the known conditions of the formation of the ice itself, and is the indisputable result of the progressive displacement of the ice mass from the high points of the ridge toward the lower edge of the field.

Considering this same parallel banding from the surface of the ice, one can clearly see that the alternately dark and light bands, directed from the western edge of the field toward the eastern, are not straight, but are curved in parallel and turn their convexity toward the lower boundary of the glacier. In the northeastern part of the field one observes a closing of the parallel bands, corresponding, without doubt, to the convexity of the bottom of the trough.

If to all these signs one adds the above-mentioned crevasses in the middle of the glacier—as the result of fractures of the ice during its movement over the uneven bottom of the trough—then a very clear picture of the movement of the glacial field becomes apparent: the answer to the second question posed.

The very appearance of the parallel banding from the surface is also easily explained on the basis of the proposition set out above. The ice entering into the mass of the glacier constitutes a system of strata of white and blue ice emerging in parallel at the surface. It is understandable that the intensity of melting of these strata will be different in two respects. First, the degree of melting from the surface will depend on the height of the position of the bands, i.e., the lower they are, the more quickly they will melt, depending on temperature conditions. Second, ice that is purer, denser, and blue melts more slowly than white, porous ice containing bubbles of air and dust. As a result of melting, grooves of whitish ice and ridges of blue ice are formed. Mineral and organic dust is carried off the glacier by streams of water and is retained by the recessed grooves of the field, thereby completing the picture of parallel banding, or dirty bands (Plate XI, Fig. 2). For the first reason, the difference in melting makes the banding stand out most clearly in the lower part of the field.

I turn to the question of the glacier’s moraines. Radde in 1859 and Chersky in 1871 do not recognize any moraines on the southern glacier of Munku-Sardyk. Bobyr’s expedition in 1887 says that the glacier’s moraine extends from the south-western part of the ridge and further forms the southern boundary of the glacier; at first it is a rare accumulation of sharp-edged blocks and farther on passes into a gradually widening rampart of small rock fragments.

Our investigations of 1897 and my subsequent observations again lead to negative conclusions on the basis of the following considerations. If one accepts the cited opinion of Mr. Yachevsky concerning the moraines of the southern glacier of M.-S., then in that case it is necessary to recognize one lateral moraine, situated along the south-western ridge of the mulda, and another terminal one—along the southern boundary of the field.

But Mr. Yachevsky’s moraine is situated not on ice, but on the solid base of the ridge, and therefore has no direct relation to the given glacier as a moraine. Under precisely the same conditions in relation to the glacier lies also the continuation of this moraine—a rampart of rather fine fragmental material, situated along the southern boundary of the field and gradually passing into the southern spur of the summit.

By the very condition of the formation of lateral and marginal moraines, there is required the existence of a rocky slope of the mulda or valley past which the glacier moves. In the present case this condition is not observed.

Admitting the actual existence now of a marginal, and formerly of a lateral, moraine of the southern glacier, such a supposition becomes little probable for the reason that in this case we have no clear idea of the former conditions that existed here earlier for such formations. At the present time the level of the glacier ice lies 8–9 meters below the supposed marginal moraine; consequently, in order to fulfill the condition for the existence of a former lateral moraine which later passed into a marginal one, it is necessary to admit a considerable increase in the thickness of the ice cover. Under the latter supposition, the ice mass could sooner have overflowed across this ridge of the mulda into the neighboring valley and thus have produced new ice streams. It is clear that under these assumed conditions the existence of a former lateral moraine is impossible. Moreover, in the supposed moraine there is not observed the form characteristic of such formations: the separate and very large stones in the south-western part of the ridge are distributed unevenly, while lower down they gradually turn into a rampart of ever smaller fragments. If attention is paid to the fact that, beginning from the southern edge of the glacier,

If the rampart of debris rises gradually and passes into a ridge, while farther on the ridge itself very quickly increases its slope, then the more probable supposition becomes that the phenomenon observed here has not the significance of a moraine, but is merely a consequence of the processes of destruction of the ridge under the influence of known conditions. As for the question of the supposed terminal moraine, here too one cannot agree with this view.

A terminal moraine is formed in a quite definite case only, when the retreat of the glacier takes place periodically. Each such moraine sums up the material of the lateral, medial, and ground moraines, depositing it in the form of accumulations and ramparts across the path of the glacier. In reality, in the present case there are no such moraine deposits. The very appearance of the supposed moraine does not lend itself to explanation from the assumed conditions of its formation.

It will be most probable to say that the existing rampart is nothing other than the destroyed southern ridge of that same mulda. It is very possible that, during a period of more intensive activity, the ice stream passed over this ridge and gave it its present smoothed-rounded form.

Although Mr. Yaсhevsky in his report also mentions signs of striation and polishing of the rocks which he found on the shore moraine he assumes, my own efforts to find anything of the kind were not crowned with success. Under the same conditions proved to be the phenomena of scars, polishing of the bottom, boulders, etc.—signs of movement. However, the absence of these data can by no means speak against the fact of movement of the glacier, proved on the basis of other phenomena; under known and fully clarified conditions, such signs, even if they could have existed at one time, later lost for the observer the character of their freshness, owing to the processes of weathering and destruction.

Having thus completed my investigations of the southern glacier of the Munku-Sardyk Range, I turn to the possible conclusions.

Radde and Chersky suppose that the “eternal snow” of M.-S. cannot be considered a true glacier: in it one sees neither lateral nor medial, nor even terminal moraines, nor, in general, movement of the ice mass, which is why they assign this glacier to glaciers of the 2nd category.

The expedition of 1887, although it found many signs of movement, does not give a positive answer to this question.

De-Henning-Michelis in 1897 comes to the conclusion that the southern glacier of M.-S. is a firn basin [[unclear: final words cut off at page bottom]], leaving—

…from a once large glacier that existed here. On the basis of my subsequent work, I join the opinion of G. de Hennin and Michelis and will say that the southern glacier of the Munku-Sardyk Range belongs to the type of alpine glaciers and represents a firn basin, or firn hollow, whose nourishment is maintained by firn forming on its slopes. In other words, the southern glacier of M.-S. is a firn field gradually moving toward its lower boundary, without leaving the limits of its hollow.

Looking carefully at the picture that appears when one attentively examines the southern slope of the main massif of the Munku-Sardyk Range and its southern glacier, we involuntarily observe certain faintly noticeable traces that carry us into the remote past. The entire locality from the main summit of M.-S. to Lake Kosogol is distinguished by a smoothed and rounded relief; the projecting hills have an oval form; such, in their appearance, are also the summits of the southern spur along which the ascent was made; to the west and east of the southern spur there begin two dry and stony talus slopes (Plate III, figs. 3 and 4), which at first run southward, reach the saddle of the spur, and from there one turns somewhat eastward, into the valley of the Bain-Gol River, while the other turns westward, into the connected valley of the Zhergalanta and Khusha rivers.

Since the movement of the material deposited in these talus slopes is utterly impossible without the participation of very considerable watercourses or even rivers, it remains to suppose that the indicated talus slopes are ground or basal moraines of a former glacier. In places in these talus slopes one notices continuous or interrupted parallel ridges of fragmental material, with no sorting. But why, then, on these same ground or basal moraines do we not find transverse deposits and ramparts—clear signs of the lateral, medial, and terminal moraine of a glacier? If we admit that the glaciers of the main summit of M.-S. that existed earlier crossed the crest of the above-mentioned hollow and descended from here to the southeast, south, and southwest, then it is possible that the moraine deposits belonging to the glacier, in the form indicated, could also have been absent because of the absence of the conditions necessary for this.

Below the forest boundary, in the direction of the northern shore of Lake Kosogol, angular fragmental material, similar in composition to the rocks of the summit of M.-S. and half-concealed in the vegetation cover, continuously covers the eastern and western slopes of the southern spur,

forming in places typical ridges or, more often, irregular heap-like accumulations and pits.

Having made an excursion into the valley of the Khushi River, which takes its rise between the first and second barren summits of the M.-S. chain, I will say that this valley, between the neighboring slopes, was closed off by a transverse wall—a ridge—and represented a rather large cirque, open by its outlet to the south. It is quite possible that this cirque once served as the receptacle of the firn field of the Khushi River glacier; confirmation of this is provided by the remains of moraine deposits that occur lower down, in the valley of the stream itself. Of course, all the signs of glaciation I have cited are deprived of their clarity and freshness, owing to later processes of erosion; but in any case they partly depict that remote picture which once existed here.

It is beyond dispute that the principal summit of the Munku-Sardyk chain, together with the nearby barren summits, constitutes a center of former, and once very extensive, glacial activity. With the subsequent change in the relief and climatic conditions of the country, this activity likewise weakened, and one of the witnesses of its past is the southern glacier of the Munku-Sardyk Range in the period of its gradually dying life.

The northern glacier of the main summit.

In 1859 Radde, while on the southern glacier of Munku-Sardyk, gave a brief description of the glacier on the northern slope of this summit. In his report, published in the Bulletin of the Geographical Society for 1860, on p. 167, Radde says: “Corresponding to the glacier on the southern side of Munku-Sardyk, on the steeper northern side there lies an even larger mass of ice. I estimate the surface of this second glacier at 9–10 square versts; it descends to a lake, which at that time was still covered with ice. This lake lies in the depth of a gorge and is surrounded by wild and bare peaks; it belongs to the Oka basin, forming its extreme south-eastern reservoir, and is called Jokhoi-Ekhe-Port; from it flows out to the north-west the Jokhoi brook, which joins the Oka at the Norin-Khoroi border guard post.” This is all that is known about the glacier of the northern slope of M.-S. in the literature of the past. After Radde, who gave this description from a bird’s-eye vantage, from the crest of the southern glacier, no one took an interest in the northern glacier. It is scarcely the chief reason for this that there are those difficult conditions which exist on the chain itself when ascending to the glacier.

northern slope. In 1898 I persistently overcame all these difficulties and reached the desired goal on June 21. Since the northern glacier is inaccessible from the Mongolian side because of the steepness of the slopes of the summit of M.-S., the only and entirely possible condition for reaching it will be to proceed from the Mondy camp upward along the valley of the Irkut River, i.e., from the Russian side.

The valley of the Irkut River, forming a widening near the Mondy camp, upstream begins gradually to narrow, rather quickly acquires a considerable gradient, and in places turns into a narrow gorge with steeply falling slopes of rocky and high mountains.

The trail from the Mondy camp goes upward along the valley of the Irkut River on the right side (with the current); sometimes, to go around rocky outcrops, it climbs steeply along the slopes above the river roaring below, descends again to the level of the channel, and, without leaving the limits of the valley, leads to a ford across the Bely Irkut River at the foot of the Nukhun-Daban golets.

Along its entire indicated extent the trail crosses several small streams, such as the Buluk, Bugute, Khairlyk, which in spring and in inclement seasons of the year are fairly swift and full-flowing torrents.

The Bely Irkut River—the right tributary of the main channel of the Irkut River—flows into it from the SW, has the character of a mountain stream, and carries its cold and transparent water through the picturesquely wild gorge of the foothills of the same giant Munku-Sardyk. The only place for a base station beneath the northern glacier of M.-S. might have been considered a small and level site with fine grass, situated immediately at the crossing over the Bely Irkut River, at the beginning of the ascent to the Nukhun-Daban golets. Since this site lies on the route to the Gargan guard post and to the Oka missionary camp, it is inconvenient. I found a suitable place for a station higher up the valley of the Bely Irkut River, on a terrace of the left bank of the valley, beneath the steep slope of Nukhun-Daban. The absolute altitude of this lower station was determined by me as 1586 meters. From the valley of the Irkut River the main summit of the Munku-Sardyk Range is hidden from the observer’s eyes because of the more massive foothills of this chain. The most convenient point for presenting the general picture of the northern slope of the summit of M.-S. should be considered the highest point of the Nukhun-Daban golets, or even the edge of its slope descending above the gorge of the Middle Irkut River. The latter view is depicted in the appended plate VII.

The chief difficulty in ascending to the northern glacier of M.-S. by the valley of the Bely Irkut River is the variable fords across the stream, the necessity for which is caused by the very narrow valley and the presence of rocks. Since the gradient of the Bely Irkut River gradually increases to 20°, and higher up, in the Pronkin spring, becomes still greater, it is clear with what speed and force this mountain torrent rushes after the melting of the snows in spring and after rainy days in summer. The ascent here is possible only under conditions of settled fine weather and low water. In high water the ascent will be risky.

Having crossed by ford opposite the station to the other side of the Bely Irkut River, I began my ascent.

The valley of the Bely Irkut River is formed on the right side by the foothills of M.-S., and on the left by the eastern slope of the above-mentioned bare Nukhun-Daban, and only near the mouth does it present a peaceful landscape. In this lower section of the little river, and also not rarely farther upstream, remains of the former valley floor are visible along the sides in the form of low accumulation terraces, now covered with vegetal soil and trees. The valley of the stream, within the limits of the channel, consists almost entirely of detrital material of mountain rocks in the form of sand, pebbles, and rounded boulders, gradually increasing in size upstream. At a distance of about 2 versts from the mouth of the Bely Irkut River, the Sredny Irkut flows into it from the left side, cutting through the massive Nukhun-Daban in a narrow gorge in a NW direction. Beyond this confluence the valley of the Bely Irkut River narrows within limestone cliffs so much that it forms, as it were, the gates of the stream, through which one can proceed farther only in low water. After a slight widening, the valley again passes into a rocky and very high corridor for about 200 meters, and farther on continues under conditions that are calmer up to the confluence of the Pronkin spring. From the entry of the latter stream on the right, and about 6 versts from its mouth, the Bely Irkut River turns toward the W and, crossing the same Nukhun-Daban, heads toward the foothills of the second summit of the Munku-Sardyk chain.

Pronkin spring, approaching the Bely Irkut River from the S, has here so great a fall that ascent along the valley of this stream becomes difficult. In my subsequent excursions I took a circuitous route along the very high edge of the rocky left slope of the valley, in order to avoid inconveniences and dangerous accidents.

After a steep ascent, the Pronykin Klyuch River turns somewhat to the SW and flows comparatively calmly.

Throughout the entire length of the Bely Irkut River, the vegetation cover characteristic of the valley of the Irkut River continues, and only gradually dies away as the terrain rises higher and higher. In the valley of the upper section of the Pronykin Klyuch River, and at an elevation of 2072 meters, there passes the boundary of the forest and of its last representative—the larch (Larix sp.). At this boundary, for the convenience and success of the work, I set up the upper station, where I often spent the night.

Beginning at the site of this station, the Pronykin Klyuch River flows with a slight gradient and, upon reaching a rocky ridge projecting across the valley, divides into two branches—the right and the left—which go around the aforementioned ridge; beyond the latter both channels again converge into one, reach a second stone ridge, and divide again into two streams: the right one of these flows out of an alpine lake situated in a glacial cirque, while the left rises once more into a small hollow and takes its source from the melting snows of the neighboring slopes.

A very considerable aid in ascending to the northern glacier of M.-S. is provided by those massive and extensive sheets of ice crust which remain in the lower section of the narrow and deep valley of the Bely Irkut River from winter and lie there almost until the first days of July. By using them, one can avoid frequent fords across the little river and considerably hasten the ascent itself.

After a tiring seven-hour walk, counting the time from the lower station, I stepped onto the northern ridge of the glacial cirque.

The cirque of the northern glacier of the main summit of the Munku-Sardyk Range forms an almost closed figure. On its southern side it is bounded by the steeply descending northern slope of the ridge of the main summit of M.-S.; on the west—by the ridge of the spur of the same summit, passing northward into Nukhun-Daban; on the northeast—by a lateral spur of the same main center; and, finally, on the northern side—by a transverse ridge closing it between the second and third boundaries (Plate VIII).

Using this ridge and the open view from it onto the mountain cirque and the northern glacier, I found here the only and most convenient place to lay out a small base and to make from here, with the aid of a compass and notches, at least a not entirely exact plan of the glacier. In view of the complete impossibility of directly measuring the boundaries of the cirque, in order to reduce the plan to a horizontal plane I applied barometric leveling of several principal points.

Having descended from the ridge into the depth of the cirque, to the alpine lake, I soon reached the base of the glacier’s terminal moraine, along which I climbed up to the upper horizon of the ice, and from here, along the left lateral moraine, reached the beginning of the ice river. (See the plan of the northern glacier of M.-S., Plate XII.) I now turn to a detailed description of the glacier.

Beginning at the crest of the main summit of M.-S. and occupying small parts of the boundary on the west and northeast, the firn field of the northern glacier is situated with a very considerable initial slope, up to 60°. Here it spreads out like a fan; on the crest of the summit it is in places joined with the neighboring southern glacier, and lower down, without leaving the limits of the cirque, it descends in two ice streams of no especially great length.

In its formation and in the feeding of its firn by the steep slopes of the crest, the firn field of the northern glacier differs from that of the southern glacier only in its somewhat greater size. Near the crest of the summit and lower down, within the limits of the firn field, exactly the same signs of parallel striation and bands are noticeable, the significance of which was explained for the southern glacier. Across the striation and bands were directed other, less regular ones, corresponding in part to the longitudinal cracks in the ice and to the direction of the rivulets of glacial water. Descending very steeply from the crest, the firn field gradually changes its slope, enters as it were upon a threshold, and at an elevation of 2894 meters passes into ice streams. (Plate V, fig. 8.) The right ice stream has the appearance of a river with parallel banks, with a slope varying from 25° at the threshold to 20° in the lower part; it is strewn with large and small fragments of mountain rocks, in composition the same as those of the M.-S. massif; on the surface of the stream small and parallel rivulets of glacial water are observed, and along its sides are two lateral moraines (Plate VIII).

The left stream, as the remnant of one that has retreated, possesses these signs only in part.

In places the surface of the field is colored by patches of the well-known alga Haematococcus and strewn with not especially numerous glacial cups.

On the glacier one can establish the existence of both transverse and marginal crevasses. The former are very clearly visible in that part of the glacier which corresponds to its passage over the bottom threshold (Plate V, fig. 8). Marginal crevasses were observed near the inner contour of the left moraine and were directed from the edge of the glacier toward its axis upward. Most of the crevasses are insignificant in their transverse dimensions and length; the greatest wid—

their width reached 25–30 centimeters, and they diminished with depth and could be traced only for 4–6 meters. This impossibility is explained by the fact that these cracks are not deep, owing to the not especially great thickness of the ice, and, moreover, in their lower parts they are soon filled with fragmental material. The ice in the cracks was homogeneous, transparent, and of a beautiful blue color. In the absence of significant cracks I could not directly determine the thickness of the ice. Since the descending stream of ice moves along the base of the cirque, the height of the upper point of the glacier’s end above the base of the cirque may give a very approximate thickness of the ice in this part of the stream. According to these data I found this thickness to be 36 meters.

Using the method of graphical construction, i.e., taking the relative heights of points from the leveling and the probable profile of the base of the cirque, one may say that the greatest thickness of the ice corresponds to the passage of the glacier over the bottom threshold and will be about 85 meters.

The right ice stream, up to 200 meters wide and up to 340 long, descending in the cirque in a direction almost due N, toward the alpine lake, bears two lateral moraines and one terminal moraine.

The left moraine consists of fairly large and angular pieces of rock falling onto the glacier from the rocky exposure of the dome-shaped hill of the cirque. This moraine, 8 meters wide, about 4 meters high, and 180 meters long—reckoning from the above-mentioned beginning—has a fairly regular appearance and descends to the end of the stream (Table IX, Fig. 13). The right moraine receives material from the northeastern slope of the cirque ridge, past which the glacier passes (Table IX, Fig. 12). This moraine has the same appearance as the left one, but differs from it in somewhat greater dimensions and length. Both of the indicated moraines, merging with the terminal moraine, form a fairly considerable accumulation of fragmental material, which slips down from the glacier and results in a natural slope of the terminal moraine up to 39 meters high. In addition to the formations mentioned, on the terminal moraine rise two cones of fragmental material, which appeared as the result of the weak melting of ice under conditions of a local accumulation of the greatest quantity of material.

In its upper and western part the firn field, descending from the ridge, as it were rests against and is halted in its movement by a natural saddle formed by the dome-shaped hill and the bottom of the cirque itself. Since the firn field moves toward the lower

to its edge, the detrital material will likewise be carried by it to the above-mentioned saddle and will be deposited there in the form of a moraine of mixed type, i.e., a terminal-lateral one.

This name is quite natural, because the firn field, having reached the saddle, does not remain there, but even now moves toward the large and small streams, owing to the existing slopes. The indicated moraine, 8 meters high and 320 meters long, extends from the beginning of the left lateral moraine, rises slightly above the saddle, rests on its solid foundation, and considerably exceeds the level of the ice approaching it. In view of all that has been said, the assumption arises that this terminal-lateral moraine is a moraine of the past. Evidence of this may be the other, modern moraine forming parallel to it, clearly visible in the drawing of the glacier. Radde, in his report for 1859, estimates the surface area of the northern glacier of M.-S. at 9–10 square versts. But such an estimate by G. Radde was evidently erroneous. Even if one takes into account the overly great, 40-year remoteness of this determination, even in that case it becomes impossible to suppose so rapid a reduction of the glacier to its present dimensions.

On the basis of the survey of the field of the northern glacier, the measurement of its slopes by leveling, and the straightening of the surface, it may be said that the area of the glacial cover will be 0.68 square kilometer.

In determining the absolute heights of certain points, I found that the upper point of the ice of the stream, situated above the base of the cirque, lies here at an elevation of 2776 meters. The elevation of the ice point corresponding to the base of the cirque at the tongue of the terminal moraine is 2737 meters. The height of the transition of the firn basin into the streams, or the threshold of the glacier, is 2894 meters. The highest point of ascent of the firn field corresponds to the height of the crest of the main summit of the M.-S. range, i.e., it will be 3453 meters.

During the ascent to the northern glacier, from the boundary of the forest, plant life likewise precisely reproduced the picture of gradual dying away and, in the composition of its forms, differed hardly at all from that of the southern slope of M.-S. Immediately beyond the forest boundary (2072 m), the last representative of which here is the larch (Larix sp.), the shrub forms present a somewhat different composition; namely, here are found the species: Ribes alpinum, Vaccinium mirtillus, Vaccinium uliginosum, Rhododendron, Salix, Betula, Caragana jubata, and others. The last species, or the so-called acacia—camel’s tail (in Buryat, tymen-khult), was encountered, as is known—

...in the valley of the Irkut River, here it rises to the boundary of the forest and higher. On the southern slope of the main summit of M.-S. I did not see this plant.

The upper limit of clearly phanerogamous plants is considerably lower than the same limit on the southern slope of M.-S. and was determined by me at 2872 meters.

Glacial water, penetrating from the melting surface of the ice to the base of the cirque by way of fissures, small funnels, and moraines, forms several small and turbid streams; emerging from beneath the terminal moraine of the glacier, they follow the lines of the natural slope of the cirque and flow into a small alpine lake situated below the slope of the northern boundary.

The water of the northwestern part of the glacial field rolls down the corresponding slope of the cirque and flows into the same place.

The small and almost round alpine lake (Plate X, fig. 14) is situated in the depth of the cirque, at an altitude of 2657 meters, and gives rise to the only source—the right branch of the Pronykin Spring River. The outlet for the water from the lake is formed where the northeastern and northern boundaries of the cirque meet. Here the water of the lake passes over a stony threshold and plunges down the gorge, forming small waterfalls on its way.

At the time of my excursions, i.e., most often at the end of June, the lake was covered with ice, and only near its shores showed a bluish-green ring of water. In the first days of July it becomes free of ice, and at the end of July freezes again. The depth of the lake was not measured because of the rather weak ice lying upon it. As observations show, this depth is considerable in the southern part; in the northern part the bottom descends gently into the depth and is covered with large fragments of rocks.

Radde, in his report, says that this lake belongs to the basin of the Oka River, is called Zhokhoi-ekin-nor, gives rise to the Zhokhoi stream, which flows into the Oka River. Now, however, it is entirely clear that the lake belongs to the basin of the Irkut River, and not the Oka River; it gives rise to the Pronykin Spring River and has no special name.

Repeated attempts to measure the speed of movement of the northern glacier of the main summit of M.-S. did not make it possible for me to give its exact value here. The chief reason for this was the very short visits to the glacier, which depended on weather conditions. It should be noted that the speed of movement of glaciers of the summit type under consideration should not be significant, in view of the circumstance that the glaciers themselves do not ve—

glaciers are limited in their distribution to mountain cirques, do not descend into the lower-lying valleys, and exist at a rather low mean temperature; for the greater part of the year these glaciers are almost at rest and manifest their intensified life only in summer, over the course of one or one and a half months.

If, owing to the conditions cited, the greatest velocity of the glaciers is insignificant, then my observations, made with the aid of not especially precise instruments during short intervals of time spent on the glaciers, are of no serious significance.

For the investigation of the phenomena of movement in the future, on the lateral exposure of the dome-shaped hill of the cirque, where the left moraine begins, I made a mark in paint and connected this point with a similar mark on a large stone of the moraine, now lying 10 meters from the mark on the rock. The mark on the stone is designated by the date 1906. For observations of the retreat of the glacier, an identical mark with the date was made on a large stone lying at the base of the scarp of the terminal moraine of the right stream.

Having visited the northern glacier of M.-S. in the summer of 1907, I made an observation of the position of the above-mentioned stone of the left lateral moraine and found that this stone had moved forward during the period from June 26, 1906, to July 26, 1907, i.e., almost in a year, by the very insignificant amount of 49 centimeters. Thus, the velocity of movement of the left moraine of the glacier, or its marginal velocity, is expressed as 49 cm per year. Taking the ratio of the marginal velocity of the glacier’s movement to the mean as 3 : 4, we find that the mean velocity of the stream in this case is expressed by the insignificant amount of 65 centimeters per year. If, however, one assumes that the greatest magnitude of this velocity pertains to the summer period, then in this case, to obtain its exact value, longer observations are necessary than those I could command.

According to the investigations I carried out of the northern glacier of the main summit of the Munku-Sardyk Range, all of its separate elements are clearly distinguished in it: the accumulation area, or firn basin; the ablation area—the ice streams; existing crevasses; the presence of moraines and, consequently, indisputable signs of movement.

In terms of its constituent parts, this glacier is a complete glacier. But since, in the present case, the single ice stream descends only slightly, with a rather great slope of up to 20°, and does not go beyond the limits of the cirque into the lower-lying valley, this glacier should be recognized as a glacier of the second category, or a hanging glacier.

Making an ascent to the northern glacier of M.-S. from the boundary of the forest, it is not difficult to notice those, although not entirely clear, features which are visible here as the result of a former picture of glacial phenomena. The entire landscape approaching the glacial cirque of the valley of the Pron’kin spring is smoothed, covered as if by accidentally fallen and very considerable fragments of rock, whose appearance, under certain conditions, can be explained only by transport by ice.

All these insignificant and, with time, masked signs of the past stand out much more clearly when we come closer to the glacial cirque and step onto its northern threshold.

As was mentioned above, the northern boundary of the glacial cirque is formed by a transverse enclosing ridge of granite bedrock outcrops. In its present appearance this ridge consists of two depressions, or saddles, and two elevated points of its crest. The processes of destruction that have affected this entire ridge have given it the appearance of a smoothed form, and this feature is most evident in its lowered parts. Examining the saddle of the ridge beneath the northwestern slope of the cirque, it is not difficult to see that the granite mass of the saddle, or, as I call it, the “threshold of the cirque,” has the appearance of a smoothed surface, broken only by cracks separating the massive rock; this surface is as if washed and in places strongly furrowed; here and there one encounters crescent-shaped hollows corresponding to stones that had lain there, around which water was constantly passing. The signs cited are an indisputable trace of the combined action of ice, stones, sand, and water. In the narrow part of the threshold, on the right side in the direction of the supposed glacier flow, scratches or striations of longitudinal direction are visible, which indicate the fact of former glacial polishing, even though these signs have lost their original freshness. The second saddle of the ridge, to the northern side, descends somewhat lower, and on the smoothed surface of the granite in places bears the same signs noted above. At the top of the saddle there are separate and fairly large stones that seem to have fallen there quite by chance; their appearance can be explained only by transport by ice, since nearby there are no high or overhanging cliffs from which the stones could have fallen into their present position.

Looking closely at the general picture of the glacial cirque from the saddle of its northern boundary, one can see very well that, beginning from the base of the glacier’s terminal moraine and taking into account the direction of...

of glacial water, there is observed a rather broad rampart of fragmental material, in places as if divided by transverse furrows. Since this ridge, in the composition of its material, is entirely similar to the moraine deposits of the glacier now observed and differs only in signs of being not of the first freshness, in this case it will hardly be a mistake to recognize in such a formation the former terminal moraine of the glacial stream, which had retreated toward the summits of the comb. It is very possible, of course, that the retreat of the glacier itself took place irregularly, in leaps, as evidenced by the indicated furrows. The above-mentioned alpine lake, with a greatest diameter of up to 250 meters and situated 416 meters from the glacier’s terminal moraine, was formed, in all likelihood, by natural collapses and talus from the surrounding rocky crests bordering it, those nearest to the outlet from it of the mountain stream. These collapses dammed the water and formed the lake. If one admits that an ice stream, moving along its bed, is capable of modifying its relief by the action of ploughing out, then it is very possible that the origin of the alpine lake is a consequence of these conditions in connection with those just mentioned. The slope of the southern shore and the depth of the lake are now indeed such that, during the period of the glacier’s most intensive activity, with a steeper fall of the ice stream, this ploughing out could have taken place. At the present time the shores of this lake in its northern part bear all the signs of incipient swamping, as the result of a diminution of the lake under the influence of the conditions of the activity of the northern glacier inseparably connected with it.

The presence of the significant and previously described terminal-lateral moraine on the saddle of the dome-shaped hill of the cirque, under the observed conditions of its position, and the formation of a new moraine running parallel with it, smaller in size, again indicate that the present glacier was once larger and is now diminishing. The granite, dome-shaped hill projecting in the middle of the cirque, supporting on itself the lower boundary of the glacial field, divides the glacier cirque into two unequal parts; it has an elongated-oval form, oriented with its long axis in the direction of the glacier’s movement, and bears the characteristic features of the so-called sheepback: a smoothed-oval form and the presence of traces of polishing. On the elevated and anterior part of the hill there are two stones of rather considerable size, lying completely separately and owing their appearance here to the transporting capacity of the ice.

Proceeding to the examination of the half of the cirque on the other side from the main stream, we see that the left stream of ice is considerably

contracted; lower down, in the direction of the slope of the cirque, in this half there are very clear signs of the former left lateral moraine of the glacier, later set against the slope of the cirque as a result of talus from the crest. The observed moraine, having lost the signs of its freshness, continues downward to the above-described exit from the cirque, or its gate. As for the absence of traces of morainic deposits on the right side of the former second stream, this is explained by the fact that the boundary of the stream was the hill of the cirque, which has a gentle slope to the west. In such a case the moraine could not have had a characteristic expression, since the quantity of clastic material could not have been considerable, and the fragments themselves, if any were brought from the firn field, rolled lower down, forming the gentle slope now observed.

All the signs given here of the former limits of glaciation of the main summit of M.-S. on its northern side are not especially prominent and clearly expressed, but, nevertheless, they do exist and give grounds for speaking of the glacier’s past.

It is beyond dispute that the glacier of the northern slope had considerably larger dimensions during the period of its most intensive activity; then, evidently, the ice mass covered an extensive surface near the crest of the summit, higher than the present one, and passed over the dome-shaped hill of the cirque, representing a single continuous stream. According to the most probable data, this stream descended into the lower-lying basin of the lake that formed later, and from there spilled over the northern boundary in its lowest parts.

Since erosion and denudation do not proceed by very abrupt leaps, it is understandable that the very change in the glacier’s existing dimensions occurred gradually; in this condition lies the reason why it is now impossible to reconstruct clear signs of that picture of the glacier which would relate directly to the above-mentioned moment of its existence. The process of the glacier’s change from its presumed and greatest dimensions to its present appearance proceeded gradually, and this latter condition can be observed more clearly thanks to the testimony of the remaining signs. With the lowering of the height of the cirque crest, with the change in climatic conditions, the ice mass began to decrease, so that the hill of the cirque, at first hidden beneath the glacier, began to expose its smoothed summit; under these conditions the continuous glacier must have divided into two streams, flowing around the hill on both sides, merging together in the depth of the cirque and passing farther over the northern boundary into the lower-lying valley. The retreat of the glacier that had begun, of course, did not have any especial—

for both reasons, stopped and continued under the same conditions. Since the cirque depression corresponding to the presumed left stream is relatively higher than that of the right stream, with the glacier’s further retreat its left stream began to diminish more rapidly, then separated completely and has now almost disappeared; the right stream, decreasing in its dimensions and thickness, continued to retreat from the lake toward the summit, leaving behind the observed terminal-moraine ridge. Consequently, according to the data for the northern glacier of M.-S., one must again establish the fact of its retreat.

Glacier at the source of the White Irkut River, on the second summit of the Munku-Sardyk Range.

A glacier hitherto unexplored by anyone, at the source of the White Irkut River, is situated on the northern slope of the spur of the second summit of the Munku-Sardyk Range. To reach it, one must follow the course of the White Irkut River, beginning from its confluence with the Pronykin Spring. The small valley of the White Irkut River cuts across the Kukhun-Daban golets somewhat in a SW direction, with a fairly considerable, though not so sharply expressed, fall as is seen in the Pronykin Spring. For the greater part of its length within the foothills of the M.-S. Range, this valley is characterized by rather smoothed forms and clear traces of erosion. The vegetation of the valley, consisting of the same usual representatives that characterize the valleys of the Irkut and Pronykin Spring rivers, accompanies the river upstream and passes into alpine forms, leaving, at an altitude of 2077 meters, the last specimens of typical larch.

At this boundary, for the convenience of excursions, I also established the upper station, where I usually spent the not especially warm night before the final setting out for the glacier.

Beginning from the forest boundary, the valley of the river rises slowly, and the very bottom of the valley is formed in places by turf with grassy vegetation. Farther on, the river turns rather sharply to the E, ascends with a steep gradient to the corner of the second bend somewhat to the S; here it enters between two steep slopes of a narrow gorge, blocked by enormous blocks of rock; it becomes precipitous, turns again to the SE, and approaches the base of the glacier’s terminal moraine.

For the ascent from my lower station to the forest boundary in the valley of the White Irkut River, about three hours must be spent, and from the latter point to the glacier about 3 hours.

Having ascended the rather steep slope of the terminal moraine, I step onto the upper surface of the ice cover and proceed to the description (Plate X, Fig. 15).

The cirque of the Bely Irkut River glacier lies on the northern slope of the foothills of the second summit of the Munku-Sardyk Range and is formed on the south, west, and east by a rocky ridge, while to the north it opens into the narrow gorge of the river. The glacier itself begins on the crest of the cirque and descends from there in two streams. Lower down, entering the base of the cirque, the firn masses continue their forward movement, owing to the slope of the cirque floor, and form a gently sloping, dome-shaped surface; the latter has a slight fall from S to N, toward the gorge, and bears all the traces of its formation in the manner of the southern glacier of M.-S. The direction of the median line of the field, across its greatest dimension, coincides with the NS line.

The firn streams near the crest of the cirque show clear traces of parallel striation, which lower down, in the gently descending part of the glacier, assumes the appearance of parallel banding, curved in an arc owing to the conditions of movement of this entire mass. In some places within the system of bands there is observed a convergence corresponding, probably, to convexities in the floor of the cirque.

The presence of glacier mills, numerous water runnels, the coloration of the field by a certain alga, and a considerable quantity of detrital material near the lower boundary of the field all give a general picture very similar to the southern glacier of M.-S. In the western part of the field, marginal crevasses are observed, the explanation of which lies in the existing movement.

The surface area of the glacier field of the Bely Irkut River, by comparison with the glaciers studied and without any particular error, may be determined at about 0.3 square kilometer.

Along the lower margin of the icy field lies a terminal moraine of rather considerable size; beyond it, parallel to it and farther from the glacier, are found clear signs and remnants of an old moraine.

The terminal moraine of the glacier, descending from its front edge toward the base of the cirque, on the western and eastern sides of the field turns into an accumulation of detrital material, in character and composition resembling glacial lateral moraines. Such a lateral moraine is observed in the western part of the field and begins as a typical ridge from the firn slope of the crest. The features cited indicate that the firn mass is moving progressively forward without leaving the limits of the basin shaped by the glacier, which indeed exist—

it, if one takes as its boundaries the crest of the cirque on one side and the terminal moraine on the other.

Determination of the elevation of the upper point of the lower edge of the glacier field gave a figure of 2800 meters; assuming that the ice mass at its lower end lies on the floor of the cirque, I found the elevation of this point of the ice to be 2753 meters. On the basis of these figures it may be thought that the probable thickness of the ice at the end of the stream will be about 47 meters. Because of its inaccessibility, the elevation of the crest of the cirque was not determined. There is no doubt, however, that this elevation is less than the elevation of the principal summit of the Munku-Sardyk Range.

The rocks composing the crest of the glacial cirque, and supplying material also for the moraine formations, are the same as on the principal summit of the Munku-Sardyk Range. Receiving nourishment from the firn of the steep slopes of the cirque, the glacier field is formed analogously to the southern glacier, and therefore represents a firn basin, with a complete absence of an area of runoff. But the presence of terminal moraines and the slight development of lateral ones indicate that this glacier is more likely a transitional form from a full glacier to a firn basin. Turning to the past, it should be said that the narrow valley of the Bely Irkut River lying below the glacier, with slopes abundantly covered with clastic material, bears all the traces of former glaciation, later concealed and masked by the results of weathering, erosion, and rockfalls. Thus, for example, on approaching the glacier itself from below, on the left side there is visible a former lateral moraine, leaning against the slope of the valley and strewn above with fragmental material. The existence on the glacier described of two terminal moraines, the presence of lateral moraines passing into marginal ones—all this indicates that this glacier is shrinking or retreating toward the summit.

The second source of the Bely Irkut River, from the point of its confluence with the stream taking its rise in the glacier described, flows through a shallow valley, with a soft, smoothed relief and traces of bogging around the channel. At first the valley runs somewhat in a NW direction, forming an insignificant angle with the direction of the M.-S. chain, and, gradually bending toward N, merges with the mountain plateau of the Nukhun-Daban bald summit. Thus, at the source of the second headstream of the Bely Irkut River there are no glaciers, and the stream receives nourishment from the atmospheric precipitation that falls in one form or another.

Wishing to clarify the question of the sources of the Sredny Irkut stream, I undertook an excursion into its valley.

As was mentioned above, the Middle Irkut River flows into the channel of the White Irkut River from the left side and cuts through the Nukhun-Daban plateau in a NW direction by a rather narrow and deep gorge. At the entrance from the valley of the White Irkut River this gorge is not very wide, and in places reaches, at the stream, a width of from 4 to 6 meters. Following the gorge along the very bed of the river, crossing by ford from one side to the other to bypass sheer cliffs, I reached a beautiful waterfall up to 12 meters high. Climbing higher, I also became convinced of the existence of a second one, which, bursting out of a cleft no wider than 0.5–0.6 meter, concealed behind it lateral splashes of an invisible and, perhaps, very great fall.

Below, beneath the second waterfall, there proved to be a small body of water, into which the water fell with a roar, forming large masses of foam. Because of the complete impossibility of following the gorge of the little river farther, I undertook another excursion into the valley of the Middle Irkut River, going around through the Nukhun-Daban golets. Having made a very fatiguing crossing from the place of my lower station along a mountain path across the Nukhun-Daban golets, only 3–4 versts from the mouth of the little river, having passed the impassable and sheer cliffs of the gorge, I descended to its bed already in the widened part of the valley and then proceeded without hindrance upstream. The Middle Irkut River, in size, is almost the same as the White Irkut River and differs only in the brown color of its water, clearly indicating the influence of a bog and of decomposing vegetation. Above the gorge described, the valley continues with slight slopes, bears traces of processes of erosion and glacial activity; in places, at the level of the little river, the valley is covered with a mossy carpet on a stony subsoil, and farther on it passes gradually, tapering to an end, onto the Nukhun-Daban plateau. The summit and all the tributaries of this little river have no glaciers.

While at the highest point of the ridge of the main summit of the M.-S. range, I saw a fourth glacier, or rather on its third golets. This glacier is an ice field situated close to the summit ridge, on the steep slope of a cirque, and facing east. With its lower edge the glacier descends into the valley with the neighboring second golets and gives drainage, in all probability, into the system of the Oka River. In appearance the glacier is also a firn basin. The area of the surface of this glacier, estimated by eye, will be about 0.3 square kilometer.

As for the investigation of glacial formations on the remaining goletses of the Munku-Sardyk Range, this question lies outside the scope of my present work. Since the main summit of M.-S., with

with an absolute height of 3453 meters, is considered the highest point of the Sayan uplift in their eastern part; on this basis it may be assumed that the glaciers of the Munku-Sardyk Range studied up to the present represent the most considerable in size and the most typical forms for the Sayans. The remaining bare summits of the chain are undoubtedly lower than the main peak. The very position of the main peak and its significance as a watershed center indicate this latter circumstance.

Thus, to the south this center of uplift gives rise to a small group of mountain streams in the basin of Lake Kosogol; to the east, on the Russian side, it sends several right tributaries of the Irkut River; to the northeast, from the center itself and from the adjacent—second—summit, it sends the Bely Irkut River of the same system. The spur of M.-S.—the Nukhun-Daban bare summit—and its plateau give rise to the rivers Srednii Irkut, Susyr, Ainak, and others, which likewise enter the basin of the Irkut River; only, beginning with the second bare summit of the M.-S. chain, the fall of the streams taking their origin on the northern slope of the range is directed westward, into the system of the Oka River. Since the absolute height of the main summit of M.-S. does not at all by itself determine the exclusive occurrence of glaciers only at this height, it is clear that the other bare summits of the M.-S. chain are not devoid of glaciation, as future work will indicate.

Meteorological Features

In carrying out, incidentally with my investigations on the glaciers, meteorological observations, I of course did not have full opportunity to conduct them continuously and with the accuracy of meteorological stations. Nevertheless, from the collected results of these incidental observations there is some possibility of passing to brief and general conclusions concerning the state of the weather and its dependence on the conditions of the given locality. The Munku-Sardyk Range and Lake Kosogol, being in immediate proximity to one another, with a difference in absolute heights of 1815 meters, exert mutually a very considerable influence on the character of the meteorological factors peculiar to them. The massive Munku-Sardyk chain, by virtue of its dimensions and height (the height of the main summit of the M.-S. Range is 3453 meters), possesses an enormous cooling surface. On the other hand, Lake Kosogol (1638 m), situated farther south and almost on the same meridian as the main summit of M.-S., has a very considerable evaporation area of about 3000 square kilometers. It is natural to expect that in summer the air, rich in vapors, rising from the lake surface heated by the rays of the sun, tends to ascend along the slopes

of the mountains surrounding the lake, forming ascending warm and moist currents of air—the wind from the lake to the shore, or the lake breeze. The maximum of the ascending current must be directed along the slopes of the highest points of the valley, as a result of which the principal direction of the lake breeze must coincide with the axis of the upland valley of the lake and, consequently, with the southern slope of the aforementioned mountain chain. In reality this is what is usually observed here after sunrise, and it continues, gradually increasing, until 3–4 o’clock in the afternoon, when the barometric minimum sets in, not infrequently accompanied by strong gusts of wind, a thunderstorm, and rain. Quite another thing occurs when the sun begins to sink toward the west and disappears beyond the horizon. In this case the influence of the summits of the Munku-Sardyk chain proves predominant, whence the cooled air, in the form of a descending current, produces the reverse phenomenon of the shore breeze—from the summits of the mountain chain toward the lake. This breeze usually begins after 4 o’clock in the afternoon, continues all night, and is again replaced by the lake breeze with the appearance of the sun’s rays on the horizon of the lake. Owing to such conditions, it may be said in advance that the meteorological data for the upper station in 1897 and 1898, which was situated at the timberline below the southern glacier of M.-S. (2260 m), were not especially favorable. Thus, for example, at night, on June 27, 1897, the temperature was only 6.5° C.; on June 29, 4.5° C.; and on June 30 it fell even to \((-)\)1.5° C. The mean temperature of the day for the entire time spent at the station, i.e. from June 26 to July 15, old style, did not exceed 11.0° C. Adding to all this the frequent occurrence of fogs and rains, and sometimes the falling of snow on the high mountain summits, it becomes clear that the weather at the end of June 1897 was rather cold and changeable, and only from the first days of July was a change in these conditions noticeable and the advent of comparatively warm days; at the same time thunderstorms began to appear, arising chiefly in the southern and south-western part of the horizon, beyond the southern extremity of Lake Kosogol.

Owing to the indicated conditions, the mean temperature at the end of June and almost until the middle of July is not especially high here; therefore the lake breeze mentioned above, bearing with it a considerable quantity of water vapor, on encountering the enormous cooling surface in the mass of the Munku-Sardyk chain with its glaciers, reaches there the state of saturation and is often resolved in the form of one or another kind of atmospheric precipitation. If the fall of the latter occurs in the lower-lying gorges and valleys, which often happens because of their natural protection by forest from heating—

...by the rays of the sun, which had earlier warmed the high and bare summits, then such precipitation later, when heated by the sun, is again converted into vapor, rises with the ascending currents of air, and follows the same path to the summits of the mountain chain. Since the gorges and valleys surrounding the main summit of M.-S. approach this center radially from all sides, the very movement of the heated masses of air and vapor—whether this be lake mist or such a current from the places lying lower around the summit—will be varied in direction. For these reasons, during the day on the summit of M.-S. excessively intense and opposing currents of warm and cold flows arise, under the influence of which the clouds move and rotate around the summit as around a center.

While making an excursion to the summit of M.-S. from the southern slope, I had occasion twice to observe hail falling there.

The first time I observed this phenomenon was at 10 o’clock in the morning on July 1, 1897. The hailstones had a conical form, with a spherical base, up to 6 mm in diameter, about 5 mm in height, and consisted of thin, needle-like ice crystals. During the second observation, at 4 o’clock p.m. on July 4, hail fell during a thunderstorm; the largest hailstones reached 1.5 cm in diameter; they likewise had a conical form, with a convex base and a small head at the top. Accompanied by peals of thunder, the hail fell for about 1/4 hour, being gradually replaced by hailstones of smaller size, later turning into groats, and finally into rapidly melting snow.

Having had no opportunity to make observations of the temperature course at the highest point of the ascent to the crest of the summit of M.-S., I cite here incidental figures that were obtained along with other observations.

Main summit of the Munku-Sardyk Range.

July 5, 1896 (old style)	July 3, 1897	July 4, 1897
4 h. p.m. temp. 6.0° C.	7 h. a.m. temp. 1.5° C. 8 h. a.m.—0.0° C.	3 h. p.m. temp. 9.0° C. 3 h. 30 m.—8.0° C.

July 7, 1897	June 23, 1898	July 22, 1900
11 h. a.m.—4.0° C. 2 h. p.m. 6.0° C.	12 h.—8.0° C.	11 h. a.m.—4.0° C. 2 h. p.m.—3.0° C.

Judging by the evening reading of the thermometer for July 3, 1897, it is probable to suppose that the night temperature on the summit of the crest of M.-S. is almost always below 0°.

Turning to the northern slope of the main summit of the Munku-Sardyk Range, it remains only to say that in this case the lake breeze of the southern slope is replaced here by the usual ascending currents of warm air along the valleys adjoining the summit and, chiefly, of course, along the valley of the White Irkut River. Since the mass of air, having reached the summit, comes on its northern slope into contact with a much more considerable surface of ice and undergoes the greatest cooling, then, descending into the lower-lying valleys, this return current brings with it a strong lowering of temperature. If to this are added the geographical conditions of the position of the northern slope of the summit, the multitude of winter snows that remain on it almost throughout the entire summer, and the movement of the descending currents chiefly along the valley of the White Irkut River, then it may be stated with certainty that the meteorological factors of the northern slope will be considerably different. In fact, this is so.

A short stay at the forest boundary, beneath the northern glacier, convinced me well that the intensity of the current of cold air and its low temperature make themselves very clearly felt during an overnight stay even in a tent. On a very clear night, with evening and morning there are observed heavy fogs, penetrating with their dampness. Lower down the valley, at the height of the base station, these conditions are, of course, not so severe, but nevertheless, on days of bad weather, the lowering of temperature is fairly considerable. Thus, for example, on June 24, 1906, the mean temperature of the day at the lower station fell from 14° to 3° C.

The Munku-Sardyk Range, with the glaciers situated upon it, undoubtedly exerts a very great influence on the climatic conditions within very extensive limits of the region adjacent to it. Since the establishment of a highly desirable meteorological station at this interesting point and in its vicinity is hardly ever likely to be carried out, then, for the purpose of obtaining data at least on the annual minimum of temperature, as early as 1897 I installed a minimum thermometer at the highest point of the ridge of the main summit of the M.-S. The Mongols, having learned from our guides about the placement of an enigmatic object for them on the sacred summit of Munku-Sardyk, soon after our departure from Mongolia threw down the entire installation.

In 1898, visiting for the first time the glacier of the northern slope of the M.-S., I decided to take advantage of a convenient place on this Russian glacier and again place a thermometer there. In 1899 my attempts in this direction were in vain: bad weather and floods of the mountain streams did not allow me to reach the glacier. Thus,

since the highest point of the ridge of the main summit of M.-S. is inaccessible from the side of the northern glacier because of the steepness of the slopes, I had to abandon the placement of the thermometer on the summit itself and confine myself to choosing the most suitable place lower down. Having ascended to the northern glacier in 1900, I climbed the dome-shaped hill of the cirque, and here found a suitable place for setting up the thermometer. For this I made use of one rather large stone, 1.5 meters high, which lay quite separately on the level surface of the top of the hill; beneath the northern side of the stone I piled up a pyramidal heap of stones, in which I placed the instrument. The spirit minimum thermometer (made by F. O. Müller in St. Petersburg) is suspended on wire supports inside a cylindrical iron box 15 cm in diameter and 50 cm long, provided with a small oblong door for inspecting the instrument.

The said box, with the thermometer, its bulb turned to the west, is laid horizontally in the upper part of the heap of stones, along the line from O to W, at a height of 0.8 meter above the ground, and is masked by stones placed on top; the box is arranged so that, after removing the upper stones from it, the latter can be carefully opened and, without taking it out of the heap, a reading of the thermometer can be made. To reset the indicator, the box must be taken out of its place, inclined toward the column of alcohol so that the indicator rolls to its end, and then put back in its former place, closed with the door, and covered on top with the same stones. The thermometer was emplaced on 19 June 1900, at an absolute elevation of 2860 meters, somewhat below the boundary of visible plants. (Table XIII, fig. 3). A certain departure from some scientific-technical conditions for installing the thermometer is positively necessary in the present case, since here, too, situations are possible in which an instrument set up conspicuously may arouse the idle curiosity of some Buryat hunter and result in its inevitable damage or breakage.

Using this thermometer, I found that the minimum temperature from 19 to 22 June 1900 was $-3.0^\circ$ C. The minimum on 26 June 1901, during inclement days, was $(-) 6.5^\circ$ C.; on 28 June 1906 it was $(-) 3.5^\circ$ C. Having begun observations of the annual temperature minimum at the indicated point in 1901, and taking thermometer readings every year, I present these data in the table.

For the winter of $19\frac{00}{01}$, the minimum . . . $(-) 36.0^\circ$ C.

» $19\frac{01}{02}$, . . . . . $(-) 35.5^\circ$ C.

For the winter of \(19^{02}_{03}\), the minimum . . . \(({-})\,33.5^\circ\) C.

” \(19^{03}_{04}\) . . . . . . \(({-})\,35.5^\circ\) C.

” \(19^{04}_{05}\) . . . . . . \(({-})\,32.4^\circ\) C.

” \(19^{05}_{06}\) . . . . . . \(({-})\,35.0^\circ\) C.

” \(19^{06}_{07}\) . . . . . . \(({-})\,34.2^\circ\) C. (*)

The mountainous landscape adjacent to the Munku-Sardyk Range, the region of Eastern Siberia, and a mean annual temperature below \(0^\circ\), evidently provide all the conditions for the formation of glaciers. But since at the same time the amount of atmospheric precipitation falling is not especially great, as is also attested by the climate of Eastern Siberia, the glaciers themselves, as is known, have in this case a very limited distribution on the Munku-Sardyk and Tannu-Ola ranges. Having considered in general outline the meteorological features of the locality under study, I draw attention to a very important factor—namely, the release into the air of a considerable quantity of vapor from the surface of Lake Kosogol, almost 3000 square kilometers in area. It was mentioned above that precipitation is observed very often on the summits of the M.-S. chain; the same phenomenon has also been found on other summits of the surrounding and high mountains, as, for example, on the Bain-Ola mountains of the western shore of Lake Kosogol.

Thus, this vast lake does not remain without influence on the existing glaciers of the M.-S. Range, and sends to them, for almost five months of the year, air that is heated and saturated with vapor, so that it may be cooled at the height of the summits and yield precipitation for replenishing the budget of the glaciers. The presence of glaciers in the Munku-Sardyk group is closely connected with Lake Kosogol, and is hardly not due chiefly to its evaporating waters.

CONCLUSION.

In concluding my description of the small glaciers of the Munku-Sardyk Range, I shall try to summarize the results obtained from the investigations I have carried out.

The highest point of uplift of the M.-S. chain is the first summit of the range from the east, with a height of 3453 meters; the remaining bald peaks of this chain, according to all observational data, are little inferior in height to the main

*) The observation was made during a journey in the summer of 1907.

peaks, although they do not emerge from the limits of the snow line. The insignificant surface of the glaciation of the range at present comprises four glaciers.

Southern glacier of M.-S.	3173 m	0.4 sq. km.
Northern glacier of M.-S.	2776 m	0.68 sq. km.
Glacier of the Bely Irkut River	2800 m	0.3 sq. km.
Glacier of the III golets of M.-S.	2800 m (?)	0.3 sq. km.
Therefore, the entire surface of the surveyed glaciers occupies an area of		1.68 sq. km.

According to the observations and calculations I made by the Hugi method, the snow line on the southern slope of the main summit of the M.-S. range lies at an altitude of 3173 meters; on the northern slope the same boundary descends somewhat lower and is at an altitude of 2936 meters. The descending end of the southern glacier is found at an altitude of 3173 meters, while the ends of the glaciers of the northern slope have an average elevation of 2758 meters. For the southern slope of the Altai, according to Heim’s table, we have the corresponding altitude for the Belukha glaciers at 2200 meters, and for the northern slope at 2600 meters. Consequently, the difference in the altitude of these boundaries, for Munku-Sardyk—385 meters and for the Altai—400 meters, is almost the same.

The boundary of evident phanerogamous plants on the southern glacier corresponds to an altitude of 3174 meters, and for the northern slope it is equal to 2872 meters.

Finally, the forest boundary on the southern slope of M.-S. rises to 2260 meters, while on the northern it is 2074 meters.

The question of traces of the former glaciation of the locality under study is now answered in the affirmative, since the signs indicated above on the southern slope of the main summit of M.-S., in the form of oval hills, moraine formations, and even partly a moraine landscape, are at present observed considerably below the existing glacier, in the direction toward Lake Kosogol. On the northern slope, likewise, traces of former glaciers are found, although they are not so sharply expressed, owing to somewhat special orographic conditions. The investigations I carried out lead to the conclusion that the former dimensions and number of the glaciers now known were once much greater. Direct observation of the present also indicates that in some glaciers all the signs of their continuing retreat are clearly noticeable. In what, then, should one seek the cause of the phenomena observed? Since the amount of atmospheric precipitation, independently of other elemen-

...the level, is one of the chief conditions for the existence and very development of glaciers, then the connection that exists between glaciers and the nearby water basin becomes clear. Consequently, the first cause of the phenomenon of retreat must be recognized as the existing moisture and its change. The second cause will be those important consequences which are the results of the processes of erosion and denudation. Under the influence of these two causes there occurs a strong change in the configuration of the prominent summits, a change in the form of the mountain cirques and valleys that formerly existed, and, together with this, in the appearance and the very thickness of the existing glaciers.

Rivers, with the lowering of the height of mountain summits, must yield a smaller quantity of water and thereby contribute to the shallowing of the basin fed by them, i.e. to the lowering of its level.

Such a connection of these phenomena and their mutual dependence may be seen in the example of the relations between Lake Kosogol and the glaciers of the Munku-Sardyk Range.

The existence of ancient and more recent shore terraces on Lake Kosogol, signs of the swamping of its shores, and the probability that in the past there existed the island of Dolon-Ola, now turned into a peninsula, all speak in favor of the fact that Lake Kosogol is lowering its level or becoming shallow¹.

The northern tributaries of Lake Kosogol, including the largest of them, the Khorë River, take their rise from the bald mountains of Munku-Sardyk and its immediate continuation to the west; the remaining, lesser-watered tributaries flow into it from the east, from the west, and partly from the south. According to observations, it appears that many of the lake’s tributaries long ago ceased their existence and function partly only during floods.

With the lowering of the height of mountain ranges and summits there occurs an inevitable change in the physico-geographical and climatic conditions of the locality, consisting chiefly in an increase of the mean temperature and in a decrease in the amount of atmospheric precipitation falling, or in a different distribution of it. These latter, and very important, conditions serve as the principal reason why the existing glaciers of Munku-Sardyk are gradually reducing their dimensions, and the time will come when they will completely cease their life.

During my excursions in 1897 and later, I found that the composition of the rocks of the Bailyk-Gol River includes boulders of basalt and

lava; during the ascent to the southern glacier of M.-S. I came across small pieces of basalt. Having this evidence, one may say with certainty that the present-day sharp summit of Yangit, which is a table mountain, as well as the main summit of the Munku-Sardyk Range itself, were formerly covered by eruptive rocks.

Above Yangit and closer to the summits of M.-S. there is a second typical table mountain, covered, apparently, by a surviving cover of basaltic magma.

Table mountains—these mute witnesses to the picture of a distant past—can assure us that at one time the surface of the mountainous country under study was not as it is now. The erupted rocks covered the summits of the range to such an extent that their elevations in former times were much greater than those observed today. It is natural to suppose that glacial activity, too, was initially developed to its fullest possible extent, traces of which we now find on the slopes of the range, in mountain cirques, and in river valleys formed by erosional processes at high points of the relief.

Table of Absolute Elevations

No.	Location	Meters	Feet	Number of observations
1.	Pass over the Elovsky spur near the village of Guzhirskoe	880	2887	2
2.	Village of Tunka	706	2316	10 + 1 g.
3.	Village of Shimki	760	2493	4 + 1 g.
4.	Ford across the Bolshoi Zangisan River	869	2851	3
5.	Crossing over the Irkut River below the Goltsy of Khara-Daban	1087	3566	2
6.	Pass over Khara-Daban	1542	5060	2
7.	Mondy missionary camp and meteorological station	1349	4426	15 + 10 g.
8.	Pass into Mongolia. Boundary marker No. 16	1818	5965	1 + 1 g.
9.	Lake Kosogol	1638	5374	60 + 1 g.
10.	Timberline on the southern slope of the main summit of the Munku-Sardyk Range	2260	7415	15 + 2 g.
11.	Limit of conspicuously flowering plants in the same place	3174	10414	7 + 1 g.
12.	Lower limit of the ice of the southern glacier of the main summit of the M.-S. Range	3173	10410	9 + 2 g.
13.	Highest point of ascent of the main summit of the M.-S. Range	3453	11329	15 + 2 g.
14.	Confluence of the Chorny and Bely Irkut Rivers	1553	5095	1 + 1 g.
15.	Lower station on the Bely Irkut River	1586	5203	28 + 3 g.
16.	Ice limit on the northern slope of the main summit of the M.-S. Range	2072	6798	5 + 2 g.
17.	Lake in the cirque of the northern glacier	2657	8717	4
18.	Lower limit of the ice at the base of the cirque of the northern glacier	2737	8980	2 + 2 g.
19.	Upper point of the ice at the end of the stream, above the base of the cirque	2776	9106	2 + 3 g.
20.	Limit of conspicuously flowering plants on the northern slope of M.-S.	2872	9423	2 + 1 g.

g — observation by hypsothermometry.

No.		Meters	Feet	Number of observations
21.	Threshold of the glacier, or beginning of the ice streams on the northern glacier of M.-S.	2894	9495	1
22.	Site where the minimum thermometer was set up on the northern glacier of the main summit of the M.-S. range	2860	9383	2
23.	Boundary of the forest on the northern slope of the second summit of the M.-S. range	2077	6815	3 + 1 g.
24.	Lower boundary of the ice at the base of the cirque of the glacier of the Bely Irkut River	2753	9032	1
25.	Upper point of the ice at the end of the glacier stream of the Bely Irkut River	2800	9187	1 + 1 g.
26.	Summit of the Nukhun-Daban bald mountain	2019	6624	2 + 1 g.
27.	Confluence of the two headwaters of the Zhekhoi River, a tributary of the Oka River	1958	6424	8
28.	Trail to Lake Ilchir. Summit of the pass into the Tymelik River valley	2328	7638	1
29.	Ford across the Tymelik River, trail	1984	6509	1
30.	Lake Ilchir	1965	6447	4 + 1 g.

LIST OF PLANTS,

collected on the southern slope of the main summit of the Munku-Sardyk Range.

(As determined by B. A. Fedchenko).

1. Ranunculaceae.

1. Anemone narcissiflora L.
2. Oxygraphis glacialis Rgl.
3. Isopyrum grandiflorum Fisch.

2. Papaveraceae.

1. Papaver alpinum L. v. nudicaule Fisch. et Mey.

3. Cruciferae.

1. Cardamine bellidifolia L. var. petiolaris Dc.
2. Draba sp.

4. Sileneae.

1. Dianthus superbus L.
2. Silene tenuis Willd. var. lenissea Rohrb. lus. caule glabro.
3. „ repens Patr.
4. Melandryum apetallum Fenzel.

5. Alsinaceae.

1. Alsine verna L. v. alpestris Fenzel.
2. Arenaria capillaris Poir. v. formosa (Fisch) lus. alpina Rgl.
3. Cerastium lithospermifolium Fisch.

6. Papilionaceae.

1. Oxytropis uralensis (Pall.) D. C.
2. Hedysarum obscurum L. var. connatum B. F. nova var.

7. Rosaceae.

1. Dryas octopetala L.
2. Potentilla nivea L.
3. » gelida C. A. Meyer.
4. » biflora Willd.
5. Sanguisorba officinalis L.

8. Onagraceae.

1. Epilobium angustifolium L.

9. Portulaceae.

1. Claytonia Ioanneana Roem et Sch.

10. Crassulaceae.

1. Sedum quadrifidum Pall.

11. Saxifragaceae.

1. Saxifraga bronchialis L.
2. » flagellaris Willd. v. stenosepala.
3. » hirculus L.
4. » melaleuca Fisch.
5. » cernu L.
6. Chrysosplenium alternifolium L.

12. Umbelliferae.

1. Bupleurum falcatum L. var.
2. Pachypleurum simplex Led.

13. Valerianaceae.

1. Patrinia sibirica Iacq.

14. Compositae.

1. Aster flaccidus Bge.
2. Pyrethrum lanuginosum (Schulz. Bip. et Herd.) m. Tanacetum lanuginosum Schulz., Bip. et Herd.
3. Artemisia borealis Pall.
4. Senecio frigidus Less.
5. Saussurea Tilessii Led.

1. Scorzonera radiata Fisch.
2. Crepis polytricha Turcz.

15. Campanulaceae.

1. Campanula simplex Stev. v. sileniflora Fisch.
2. „ pilosa Pall.

16. Ericaceae.

1. Osmothamnus fragrans Dc. v. pallidum Herd.
2. Rhododendron chrysanthum Pall,
3. „ lapponicum L. v. albitlorum Max.

17. Primulaceae.

1. Primula farinosa L.

18. Gentianaceae.

1. Gentiana humilis Stev.
2. „ altaica Pall.
3. „ algida Pall. v. sibirica Kusn.

19. Boragineae.

1. Myosotis silvatica Hoffm.

20. Scrophularineae.

1. Pedicularis omoena Ad.
2. „ versicolor Wahlb.

21. Labiatae.

1. Thymus serpyllum L.
2. Dracocephalum altaiense Laxm.

22. Polygoneae.

1. Oxyria reniformis Hooc. f.
2. Polygonum Bistorta L.
3. „ „ L. v. angustifolium.
4. „ viviparum L.
5. „ „ L. v. angustifolium Mesin.
6. „ divaricatum L. var. angustissimum Mesin.

23. Empetreae.

1. Empetrum nigrum L.

24. Salicineae.

1. Salix myrsinites L.
2. „ berberifolia Pall.

25. Betulineae.

1. Betula fruticosa Pall.

26. Liliaceae.

1. Lloydia serotina Rchb.
2. Allium Schoenoprasum L.

27. Cyperaceae

1. Carex ustulata Wahlb.

28. Gramineae.

1. Festuca ovina L. v. violacea Gaud.
2. Leucopoa sibirica Gris
3. Avena subspicata Clairv.

29. Polypodiaceae.

1. Woodsia ilvensis.

Musci.

(As identified by Mr. Brotherus).

30. Splachnaceae.

1. Tetraplodon urceolatus Br. Eur.

31. Grimmiaceae.

1. Rhacomitrium uliginosum.
2. Grimmia incurva Schw.

32. Encalyptaceae.

1. Encalypta rhabdocarpa.

33. Polytrichaceae.

1. Polytrichum alpinum L.?

34. Hypnaceae.

1. Amblystegium fluitans.
2. ” subtrifarium Broth. n. sp.

Lichens.

(As identified by A. A. Elenkin).

35. Sphaerophoraceae.

1. Sphaerophorus sp.

36. Lecideaceae.

1. Rhizocarpon geographicum L.

37. Lecanoraceae.

1. Haematomma ventosum (L.) Mass.
2. Gasparrinia sp.

38. Gyrophoraceae.

1. Umbilicaria sp.

39. Parmeliaceae.

1. Alectoria ochroleuca (Ehrh.).
2. Cetraria juniperina (L.) Ach.
3. ” nivalis (L.) Ach.
4. ” lacunosa Ach.

40. Cladoniaceae.

1. Cladonia sp.

Plate XI

1.

2.

Plate XII

Southern glacier of M.-S.

Eternal ice.

Stony talus.

Section along line A.B.

Northern glacier of M.-S.

Eternal ice.

Stony talus.

Section along line A.B.

Scale: 200 sazhen in one English inch.
200. 100. 0. 200. 400. 600 s.

Table XIII.

1. View of the installation of the minimum thermometer.

Map of the Location and Disposition of the Glaciers of the Munku-Sardyk Range

Scale: 1:210,000

1906

Conventional Signs

— Ridge crest
— Lateral moraine
⋯ Route of the excursionists
— Boundaries of the survey
◎ Eastern point
▲ Station of excursionist observations
〰 Stone scree

Visible Map Inscriptions

Russia

Mongolia

Munku-Sardyk Range

Lake Khubsugul

Irkutsk

Tunka

Mondy

Lake Ilchir

Lake Khara-Nur

Lake Balykty-Nur

Lake Ekhe-Gol

River Irkut

River Bely Irkut

River Muguvek

River Bugovek

River Buryk

River Oka

River Sentsa

River Tissa

River Khankhar

River Khubuty

River Shumak

River Zun-Khandagai

River Dund-Khandagai

River Barun-Khandagai

River [[unclear: river name]]

Mount Munku-Sardyk

51°45′

100°30′

100°45′

101°00′

1. My sketch of Lake Kosogol. Proceedings of the Society of Naturalists at Kazan University. Vol. XXXVII, issue 6, 1903. ↩↩↩

2. Potanin. Sketches of Northwestern Mongolia. 1881–1883. ↩↩

3. Makerov. Journey to the Sayans. Report for 1887. ↩↩

4. L. A. Yachevsky, Preliminary report on the geological part of the Sayan expedition. Proceedings of the East-Siberian Section of the I.R.G.S. for 1888, vol. XIX. ↩

5. Report of the expedition of 1897. Proceedings of the East-Siberian Section of the I.R.G.S. for 1898, vol. XXIX, No. 3. ↩