CONTENTS. (continued 3)

ZUNGSAM RIVER. August, 1847.

A little way above Shialkar and Chango, two very considerable rivers unite to form the Piti river. The larger of these, descending from the north-west, is known by the same name. The other, which flows from the north-east, may be called the Parang river, by which name it is known in the upper part of its course; lower down, it seems to be usually called Zungsam. The direct road from Hangarang to the Indus lies up this river, which unfortunately flows for several days' journey through districts which are included within the Chinese frontier. It was our wish to proceed by the most expeditious route, and at the same time that nearest the line of boundary, to Hanle. It was, therefore, our object to effect, if possible, a passage up the Zungsam river, though, as we knew that Captain Gerard and M. Jacquemont had both been stopped upon the frontier, we had no reason to anticipate any more favourable result.

We therefore took, on leaving Chango, a north-easterly direction, proceeding, on the 27th of August, to a village on the left bank of the Parang or Zungsam river, called Changar, the same place which, by Gerard and Jacquemont, is named Changrezing. Leaving the cultivated lands of Chango, and crossing the stream which skirts the plain, we immediately commenced a steep zigzag ascent over a barren shingly road, to the heights which overhang the village to the north-east. After a very fatiguing climb of not less than 1300 feet, we attained the summit of the ridge, and advanced along it for some distance without much change of level, but still gradually ascending among low-topped gravelly hills. A very steep ascent followed to the summit of the pass, which was called Changrang La[8], and could not be much under 13,000 feet. The whole ascent was extremely barren, the arid slopes producing a minimum of vegetation. A fleshy Cruciferous plant, with a strong pungent taste not unlike horse-radish (Christolea of Decaisne in Jacquemont), a fine Nepeta (N. floccosa, Benth.), and a little Stipa, were the only novelties; and these, with the Ephedra, a little Lactuca, an aromatic species of Chenopodium (C. Botrys), the Tibetan Euphorbia, and a shrubby white-flowered spinous Astragalus, were almost all the plants observed.

CHANGAR. August, 1847.

From the summit of the pass, the road descended abruptly into a deep ravine, which originated in a snowy mountain to the south. At the bottom of this ravine, between rocky precipitous banks, ran a considerable torrent, which was crossed by a very frail wooden bridge. Immediately after crossing, the road began to ascend rapidly, rising to an elevation only a few hundred feet lower than the pass from which we had descended, after which, half a mile of nearly level road brought us to our camp at Changar, a small village on a stony hill, of which only one house seemed habitable. A few fields of barley, not yet ripe, separated our tents from the village; these were irrigated by a small streamlet, whose source was a spring on the rocky hill-side a few hundred yards off, shaded by a few rose-bushes and a small clump of juniper-trees.

We remained one day stationary at Changar, to complete some arrangements which were required previous to our leaving the district of Hangarang; and on the 29th we proceeded to ascend the valley of the Zungsam river, intending, if no obstacles were offered, to follow its course and the regular road to Hanle; but in case of obstruction, which there was every reason to apprehend, to adopt the plan which had been already followed both by Gerard and Jacquemont, of crossing the river, encamping on its north bank, and proceeding in a westerly direction along the course of the Piti river to the Parang pass, in which direction we could effect a passage to Hanle without the necessity of entering on the territories under Chinese control.

BLACK CURRANT. August, 1847.

Our road, for about three miles, was undulating, with rather a tendency to descend, but without any abrupt change of level. It lay along the gently sloping side of the ridge, and crossed a good many little ravines. To the right was the crest of the ridge; to the left, the valley of the Zungsam river, which was nearly 2000 feet below, the slope being very precipitous. In many of the ravines, where there was a stream of water, there was a dense jungle of shrubs, which contrasted strongly with the barrenness of the hills. A willow, rose, Lonicera, a shrubby Astragalus, an Artemisia, a Potentilla of large size, and a black currant, closely resembling that of our gardens, were the principal shrubs; and the herbaceous vegetation was the same as in similar places since entering the Tibetan region. The currant, which occurred here for the first time, was quite a new species: its ripe fruit was quite black, and had the size and flavour of the common black currant, with, however, a considerable degree of acidity.

About three miles from Changar, an abrupt descent led from the platform on which we had been travelling, to the level of the banks of the river, more than 1000 feet lower. The path by which we descended was steep, rocky, and difficult. The rock was still clay-slate, with granite veins. The granite in general very much exceeded in quantity the rock into which it had been injected, as was well seen on several precipitous cliffs along the course of the stream, in which the stratification of the slaty rock and the ramifications of the granite could be examined in detail. The banks of the river were adorned with a species of Myricaria, a small tree, with very delicate graceful foliage and beautiful rose-coloured flowers.

CHINESE FRONTIER. August, 1847.

On reaching the small streamlet which forms the frontier of the Chinese dominions, we found, as indeed we expected, that there was no intention of permitting us to proceed by the direct road to Hanle; and all arguments to induce a compliance with our wish proving ineffectual, we agreed to take the route up the Piti river by Dankar, and were then permitted to proceed about a mile, to the village of Kyuri, where we encamped for the day. I have now no doubt that if we had resolutely advanced, no serious opposition to our progress would have been made; but our instructions were so precise that we should not have been justified in using the smallest degree of force, or incurring any risk of a collision.

To reach Kyuri[9], we crossed the Zungsam river by a very remarkable natural bridge, composed of an enormous block of granite, which has in some way been placed across the stream, at a spot where it is much contracted in width, flowing in a deep rocky fissure from fifteen to twenty-five feet in width, evidently of great depth. At this point a considerable number of boulders of large size are piled on both banks, of which the one that spans the channel is eighty-five feet in length and probably not less than forty in width and twenty in depth; it is placed obliquely across the stream, its left or southern extremity being lower, and inclined at so considerable an angle that the passage is one of some difficulty for horses and loaded cattle, though for men, with ordinary care, it is quite easy. The higher end is so much elevated above the surface on which it rests, that a rudely constructed stair of stones is necessary to enable travellers to descend.

KYURI. August, 1847.

Our encamping ground at Kyuri was on a gently sloping barren plain, seven or eight hundred feet above the valley of the Zungsam, and was reached by a short steep ascent from the bridge. There was no cultivation; but about a mile to the east, a long sloping tract of alluvium interposed between the mountains and the river was covered with green fields, though it had only two houses and not a single tree. Alluvium abounded in every direction, forming steep sloping banks, often much worn away by running water, and occasionally from two to three hundred feet in thickness. The plain on which we were encamped was also of recent origin; it consisted of a fine clay, curiously worn into cliffs and narrow ridges. A few layers of fine sand were included in the clay, and by a careful search I found three or four small fresh-water shells in the clay, belonging to at least two species—one a Lymnæa, the other a Planorbis. The shells were, however, very scarce, and all found near one spot, nor did any other portion of the deposit seem fossiliferous. The clay was in front of and below our encampment, and was covered by coarse alluvial conglomerate.

THE ROAD REJOINS THE PITI VALLEY. August, 1847.

The road up the valley of the Parang river being tabooed to us by the jealousy of the Chinese Government, it became necessary to make a very considerable détour, no practicable road being known in the mountains north of Piti, between that which we were thus prevented from following, and the Parang pass, to reach which we had to make five or six marches up the Piti river before turning to the north. On leaving Kyuri, on the 30th of August, we ascended gently on a bare gravelly hill for several hundred feet, and then proceeded for two miles to the westward, along the steep side of the mountain. The road was rocky and very barren, the caper and an Astragalus being almost the only plants seen. We then descended rapidly, so as to reach the bank of the Piti river, at the place where it makes its great bend and assumes a southerly direction. Here it is joined by the Giu (Gumdo of Jacquemont), a considerable stream, which has its source in the lofty and inaccessible range to the north. This torrent had excavated a deep channel in the alluvial beds, which were composed of alternations of coarse incoherent conglomerate and fine clay. In this ravine, which sheltered them from the bleak winds of the more exposed slopes, I found a luxuriant growth of shrubs; of which the commonest forms were the rose, ash, Colutea, Rhamnus, Myricaria, Capparis, Ephedra, and Artemisiæ. As soon as the stream was passed, a steep ascent commenced, but the luxuriant vegetation at once disappeared, and the road was as barren, dusty, and stony as usual. We ascended only to descend again, and encamped on a small level spot forty or fifty feet above the Piti river, destitute of cultivation or inhabitants, but known to the people of the district by the name of Huling.

On our next day's journey, the country at first presented the same general character. The mountains along the road were lofty and rugged, and sloped steeply to the river. A mass of alluvial deposit generally rested on their bases, and the road lay at no great distance above the river, rising a few hundred feet to pass over the spurs, and again descending on their western sides. The steep slopes were in several places covered with an incrustation of hard angular breccia, with a calcareous matrix, the origin of which I conceive to have been the same as that of the breccia noticed in the vicinity of Sungnam. Across the river there was a considerable tract of level ground, covered with cultivation surrounding a small village, with a few poplar and willow trees; but the left bank, on which we travelled, was entirely barren. Nearly opposite this village, the bank of the river becoming steep and precipitous, we ascended about a thousand feet, and continued at that elevation till the difficulty was passed, after which we returned to the river-side. At this elevation two or three springs broke out on the steep face of the hill, and, trickling down among the rocks below, promoted the growth of a few willows and rose-bushes, and a small thicket of Hippophaë. Round the springs the ground was covered with a slight saline exudation.

LARI. August, 1847.

The village of Lari, at which we encamped, is elevated 11,200 feet. It occupies a large extent of alluvial surface, sloping at a very small angle from the base of the mountains to the river, at a place where a stream issued from among the mountains. The cultivated lands are extensive, but very bare of trees when compared with the villages in Kunawar, or even in Hangarang. One apricot-tree only could be seen in the village lands, but there were still a few willows and poplars. The flora of the cultivated tracts had not altered. The little Iris, first seen at Chango, was very common, and the gentians, Potentillæ, Astragali, and other small plants, were the same as had been common since crossing the Hangarang pass; the season, however, was so far advanced, that much of the luxuriant vegetation had withered away. The crops of wheat and barley were quite ripe, and had been partly cut; but a few fields of millet were still green.

MOUNTAINS OF PITI. August, 1847.

In the neighbourhood of Lari, the Piti valley is considerably more open than lower down. It had, indeed, been gradually expanding since we joined it at Lio. The mountains now recede considerably from the river, a long sloping surface of alluvium being interposed, which is at one time largely developed on the north side of the river, in which case the southern spur generally projects. A little further on, the northern mountains send down a projecting spur, and an open tract is seen to the south. The mountains behind the alluvial platforms rise very abruptly, and present towards the plain, steep, almost perpendicular slopes, which, from the peculiar nature of the rock, a very fragile slate, are covered by a steeply-sloping mass of débris almost to the top. This talus, indeed, on some of the cliffs behind Lari, seems to rise to the very summit of the ridges.

It is not easy to convey an idea in words of the mode in which these mountains are arranged, unless it is recollected that it is an universal rule that all mountains are ramifications of an axis, giving off branches on both sides, and that each branch is again divided in a similar manner, till the ultimate divisions are arrived at. All mountainous districts are in this respect similar to one another, and differ principally in the proportion borne by the altitude to the superficial extent of the ranges of which they are composed. An examination of the map will show that the axis of the range which lies north of the Piti valley, passes through the Parang pass, and in fact occupies the midway between the Piti and Parang rivers, terminating in the great bend of the latter, to the east of its junction with the Piti. The whole of this range is of great altitude, and it seems to rise in elevation to the eastward, no passage being known further east than the Parang pass. The primary branches of this chain, descending towards the Piti valley, are separated by considerable tributaries which discharge themselves into that river. In general, these lateral streams have, in the lower part of their course, very rugged rocky channels, but they rise rapidly, and, at a distance of a few miles from the main river, their ravines expand into open valleys, three or four thousand feet above its level. The ramifications of the primary branches are, as might be expected, in their upper part concealed among the mountains, but those near their termination abut upon the main valley, in a series of ridges separated by little streamlets. We have, therefore, as we ascend the Piti river, not a wall of mountain, parallel to its course, but a succession of ridges, more or less perpendicular to it, all descending from a great elevation, and rapidly diminishing in height. The result is necessarily a great degree of irregularity, the width of the alluvial belt varying much, while the direction of the ridges, and of the cliffs by which they are bounded, is constantly changing.

ALLUVIAL PLATFORMS. September, 1847.

Leaving Lari on the 1st of September, we continued our journey up the Piti valley. The road lay partly on the platforms of alluvial conglomerate, and partly over the steep shingly talus which rested on the hills where they were not separated by alluvium from the river. One alluvial plain, about two miles from Lari, was well cultivated with the usual crops, the barley being quite ripe, the wheat very nearly so, the oil-seed and buckwheat out of flower, and the millet, of which there were only a few fields, still green. The platforms of alluvium have, in general, an irregularly triangular form, the base resting on the river, the apex at the termination of a mountain ravine, down which a stream runs. This stream, instead of bisecting the platform, usually runs in a hollow channel on one side or other between the mountains and the alluvium, and is, where practicable, carried off in small artificial conduits for the purposes of irrigation. The platforms always slope gently from their apex to the river, and they are generally cut off in a cliff at the lower end. These cliffs always show marks of stratification, sensibly parallel to the river, and the pebbles which the alluvium contains, are (and have been for the last two days) usually rounded.

I ought not to omit to mention, that I use the word alluvium merely as a convenient mode of expression, without meaning to convey an idea of the mode in which these beds originate. No equally suitable word suggests itself, and the phenomena occur so frequently, that it is necessary to have some short expression by which to describe them. The origin of these alluvia is certainly very puzzling. At first sight, in any particular spot, the most natural suggestion is, that they have been deposited under water, and probably therefore in a lake. Their occurrence day after day, notwithstanding the greatest changes of altitude, their enormous thickness in many places, and the peculiar position in which they occur, soon dispel this idea, and throw the observer into a maze of doubt and difficulty, at last leading him to the conclusion, that no one cause is sufficient to explain the highly variable phenomena which he observes, and that a lengthened series of patient observations will be necessary before the subject can be understood. These observations have yet to be supplied, but I believe I shall best serve future observers, by detailing as fully as possible the points which attracted my attention, without attempting for the present to speculate upon the causes of the phenomena. The suggestions which I have to offer to the reader, will be best understood when I have detailed all the facts upon which they are founded.

It is especially necessary to distinguish between three forms of alluvium, all of which have already occurred in Piti. These are, first, the fine clay; secondly, the platforms, such as I have described in the last paragraph; and thirdly, the enormous masses, which are without any definite limits, and do not seem referable to any present valley system.

POK. September, 1847.

We encamped at Pok, a large village nearly nine miles from Lari. Here we found again an extensive alluvial platform, covered with much cultivation; and on the mountain ravine above the village there was a considerable grove of young juniper-trees. A week or two before, I should have considered them as scattered trees; now they had quite the appearance of a forest, so bare had the country been since crossing Hangarang.

West of Pok, our journey of the 2nd of September was over the alluvial platform, which continued for two miles beyond the cultivation of the village, gradually contracting in width by the encroachment of successive spurs, which at last advanced close to the river. The road now ascended by a short steep path on the mountain-side, to a higher level. At the base of this ascent there were a great many angular masses of limestone, evidently transported from the valleys behind. These fragments were very numerous, and many of them of great size. They continued abundant during a great part of the day, but no limestone was seen in situ. I have not preserved any record of the exact position of these angular fragments with regard to the valleys behind, but I have little doubt that they will be found to be of glacial origin, such being certainly the case in many other similar instances. The limestone was very compact, of a blue or grey colour, and many of the fragments were almost full of coralline remains. I collected many fossiliferous specimens, which were afterwards despatched from Hanle to Simla by a messenger, on whom we thought we could rely, but they never reached their destination[10].

DANKAR. September, 1847.

We encamped at Dankar, after travelling ten miles. This place is the principal village of the Piti valley, and is 13,000 feet above the level of the sea. The valley of the Piti is here very wide, and divided into numerous channels, which are separated by low gravelly islands, the whole width of the river being not less than half a mile. Here the alluvium is very highly developed, lying in patches on the face of the steep hills. The village of Dankar, though 1000 feet above the river, occupies both sides of a steep ridge entirely composed of alluvium. Nor is this its utmost limit; for several hundred feet above the houses, similar alluvial masses occur. These beds are not, however, continuous from these great elevations, down to the level of the river: they rest, on the contrary, on the ancient rocks, which are here very steep, and the clay may be seen in isolated projecting masses, capping the most prominent ridges[11].

RANGRIG. September, 1847.

The village of Dankar is built on arid barren soil, but the cultivated lands stretch from about the level of the village almost to the river, on a very steep slope. Thickets of Hippophaë were scattered among the cultivation, where the ground was swampy; and notwithstanding the great altitude, the exposure being favourable, the crops seemed good, and the wild plants were more luxuriant than usual. One of the new species observed was a pretty gentian (G. Moorcroftiana, Wall.), interesting as having been one of the few plants sent from the Tibetan country by the unfortunate traveller whose name it bears. It is also a common species in the valley of Dras, in which, perhaps, Mr. Moorcroft's specimens were collected, unless, indeed, they were obtained in Piti by Mr. Trebeck, during his journey to that valley from Ladak.

Leaving Dankar on the morning of the 3rd of September, we ascended the heights behind the village to the side of the main ridge behind, along which we proceeded without change of level. The mountain was almost precipitous, and extremely barren, but commanded a fine view of the open flat plain of the Piti river, descending from the north-west; and of the course of the Pin, a large tributary which descends from the south-west, at the source of which there is a pass, by which it is possible to descend upon the Sutlej at Wangtu. The mountain range interposed between the Sutlej and Piti valleys was, from the elevation at which we now stood, seen to great advantage. These mountains are, indeed, in the terse words of Jacquemont, "d'une affreuse stérilité;" yet, in their varied outline, massive forms, and snow-sprinkled summits, there is no doubt a degree of grandeur, which produces a powerful impression.

At about a mile and a half from Dankar, during which we had, with the ridge, gradually approached the river, the road began to descend, and we at last reached the bank of the river, close to which, and sometimes even on its gravelly bed, we continued for several miles. Where the banks were lowest, and the gravel was moist, there were thickets of low shrubs, Hippophaë, Myricaria, Ribes, and willow; elsewhere, the gravel was barren and unproductive. We encamped at Lara, a village nine miles from Dankar, at which there were only two poplar trees, and a very small extent of arable ground. The wheat was ripe and very luxuriant, the ears being large and well filled.

On the 4th of September, we continued our progress up the Piti valley, which had quite the same aspect as on the day before, encamping on the left bank of the river, opposite to the village of Rangrig, on a desert spot among limestone rocks, at an elevation of 12,300 feet. Here we had attained our furthest limit in a north-westerly direction, our road now turning to the right, and ascending a considerable valley towards the Parang pass, in a direction which promised much novelty and interest, as it had only been traversed by one traveller, the unfortunate Trebeck, who, in the year 1822, travelled from Le to Dankar by this route. The further course of the Piti river, which, as we learn from Moorcroft's travels, was visited in 1822 by Captain Mercer, was afterwards surveyed by Captain Broome. It communicates with Lahul, which is the upper part of the valley of the Chandrabhaga or Chenab river, by the Kulzum pass, a depression in that great branch of the trans-Sutlej Himalaya, by which the waters of the Sutlej and its tributaries on the east, are separated from those of the Chenab and Beas.

During our journey through the district of Piti, the weather had been almost uniformly dry and serene, though we were now in the very height of the Indian rainy season. The only exception occurred while we were encamped at Changar, on the lower part of the Parang river, about the 29th of August, when the sky was for two days very cloudy, and on one night it rained gently for nearly half an hour. The clouds were, however, high, and never dense, and the unsettled state of the atmosphere was of very short continuance. While it lasted, it was accompanied by violent wind, very irregular in direction.

SALINE INCRUSTATIONS. September, 1847.

In every part of Piti we found the margins of springs, and the grassy turf which grew on low swampy spots along the river, covered with a saline incrustation, in the form of a dry efflorescence, which encrusted the blades of grass. It appeared to be confined to the vicinity of water, the barren rocky tracts being destitute of it. This saline matter, as elsewhere in Tibet, consists of sesquicarbonate of soda, and, as a consequence of the abundance of that alkali, soda-producing plants were common, especially Chenopodiaceæ, among which the common Salsola Kali was very abundant.

PITI. September, 1847.

The district of Piti, which was formerly almost independent, but paid tribute to, or exchanged presents with, all the Tibetan countries in its neighbourhood, namely, with Garu, Ladak, and Lahul, as well as with Kunawar, followed in 1846 the fortunes of Lahul in being transferred to British rule. It is a very thinly populated valley, the villages being small and distant, and the arable tracts of no great extent. The mountains on its southern border, by which it is separated from Kunawar, are so very elevated that they entirely intercept all access of humidity from the districts to the northward of them, and render the climate entirely rainless. The houses are in consequence very generally built of unburnt bricks, made of the fine lacustrine clay so common in the valleys, and their flat roofs are thickly covered with a layer of the same material.

CHAPTER V.

Leave Valley of Piti river—Kibar—Cultivation above 14,000 feet—Vegetation of mountains—Rocky gorge—Encampment at 17,000 feet—Parang Pass—Snow-bed and glacier—First plants at 16,500 feet—Parang valley—Gorge leading to Chumoreri Lake—Kiang, or wild horse—Chumurti—Remarkable grassy plain—Lanak Pass—Granite boulders—Plants above 18,000 feet—Undulating hilly country—Hanle plain—Vegetation—Monastery of Hanle.

Our last occupation in the valley of the Piti river was to make the necessary arrangements for the transport of our baggage through the deserts which were to be traversed before we should again arrive at inhabited tracts. The principal part of our effects were carried by men, but our party was so large that it was not easy to provide porters for the necessary amount of food during a journey of a week in an uninhabited country. A motley group of ponies, asses, and yaks therefore formed part of the train which accompanied us into the desert country between Piti and the Indus.

Three miles north-west of our encamping ground opposite Rangrig, we left the Piti river on the morning of the 5th of September, turning up the valley of a considerable stream which here joined the main river. The platform of alluvium on which we had been travelling continued for about half a mile up the lateral valley, and was covered with large boulders of angular fragments. The rock was limestone, the same as had occurred everywhere since leaving Lara. A little village called Ki, and a large monastery, situated on a curious, seemingly isolated, conical hill above the village, were passed on the right hand. Soon after, the ascent became rapid on a steep ridge to the east of the stream, and the Piti valley was completely shut out from view as we got in among the mountains. The ridge by which we ascended was barren and stony, and produced little vegetation. A curious broad-leaved Allium was the only novelty. We continued to ascend along the stream till we reached the village of Kibar, at which we encamped, at an elevation of 13,800 feet, in a narrow valley surrounded on all sides by lofty mountains.

KIBAR. September, 1847.

Kibar is rather a pleasing-looking village, remarkable for its houses being all built of stone, instead of the mud or unburnt brick so commonly used in the valley of Piti. It is situated on the summit of a limestone rock, on the right bank of the stream. Our tents were on a patch of green-sward on the opposite bank, separated from the village by a deep ravine. Crossing this on the morning of the 6th, we ascended the slope of the hill above the village, among cultivation which rose on the hill-side fully 300 feet higher. Except one field of oil-seed, the crops were all barley, which was ripe, and partly cut: it was apparently very poor, being thin and deficient in ear. After leaving the cultivation, we continued to ascend on the ridge, till we attained an elevation of nearly 15,000 feet, at which height the road wound round the sides of hills, without any considerable change of level, for two or three miles. It was still early morning, and the air was very frosty. Every little rill was covered with a thick coating of ice, and some small swamps which we passed were crisp with frost.

VEGETATION. September, 1847.

Notwithstanding the considerable elevation, I noticed but little in the vegetation different from that common in Piti. The forms were by no means so alpine as on the passes between Kunawar and Hangarang, though the elevation was greater than on any of these. It was probably owing to the aridity of the climate that the flora, at elevations of 15,000 feet, instead of being composed of delicate alpine plants, was much the same as it had been 4000 feet lower. The rose, the common Rhamnus of Piti, a little shrubby Potentilla, a spinous Astragalus, and several Artemisiæ, were the common shrubs, and two species of rhubarb grew abundantly on the dry hills above Kibar. The Dama, which shuns the level country, the Allium first observed the day before, and Lamium rhomboideum of the Hangarang pass, were almost the only striking plants observed; all the others were those of the ordinary flora of the dry hills and gravelly plains of the Piti valley. It is necessary, of course, in comparing this vegetation with that of the passes, to recollect that we were here in a valley, on slopes surrounded on all sides by lofty ridges, not on the summit of a range overlooking everything around, or only surpassed a very little by the continuation of the same ridge; so that the temperature of the summer months must be considerably higher than on the more exposed though less elevated passes.

ROCKY GORGE. September, 1847.

Further on, the road descended rapidly to the stream, which flowed in a rocky gorge, through which we held our course for three miles. A few willows, and stunted shrubs of Myricaria, occurred on the descent, and the willow was found occasionally on the banks of the stream in the gorge, which was enclosed by high and steep limestone rocks on both sides. These gradually contracted as we advanced, but again expanded at the point where we encamped, which was close to the bank of the stream. The ravine being now more open, we could see the hills to better advantage, and were struck with astonishment at the desolation by which we were surrounded. We were, in truth, in a wilderness of rocks, which to the south closed together, so as to shut in the ravine by which we had ascended. High walls of cliffs rose on either hand to an elevation of at least 1500 feet, displaying a natural section of a multitude of strata, which seemed to be repeated again and again in a succession of beds of limestone and slate. The elevation of our encampment was 14,800 feet.

On the 7th of September, the wish of our guides and porters according with our own, we did not cross the Parang pass, which was still five miles distant, and nearly 4000 feet above us, but contented ourselves by ascending to the highest water, perhaps 1500 feet below the summit. We ascended on a steep shingly ridge to the right of the stream where we had passed the night. Tufts of Lamium rhomboideum grew among the loose shingle, but no other plant seemed to vegetate in such an ungenial soil. When we had passed from the shingle, which was confined to the base of the ascent, the ridge was dry and gravelly, with tufts of Dama and of a species of nettle. Above 16,500 feet, the spur was rocky and uneven, and some alpine vegetation was observed, for which I conjecture that the melting of the snow had probably supplied moisture, as lower down the sterility had been complete. About fifteen species were collected, two Potentillæ, Biebersteinia odora, a Lychnis, a little tufted saxifrage, and species of Nepeta, Artemisia, Gnaphalium, Saussurea, Allardia, Polygonum, Rheum, Blitum, one grass, and a fern. Three or four lichens grew on the stones, and I obtained one specimen of a moss without fructification. The Allardia, a pretty little rose-coloured flower, with an agreeable smell, was the only new species; all the others were already familiar to me. They grew in the crevices of the rocks, in extremely small quantity, struggling as it were for existence against the unfavourable circumstances to which they were exposed.

ASCENT TOWARDS THE PARANG PASS. September, 1847.

A stony ravine, elevated about 17,000 feet, was the place selected for our encampment. A small stream, supplied by a patch of snow a little way above, trickled down under the angular gravel. The ascent had been extremely fatiguing, because almost without intermission, and we were glad of rest on reaching that elevation. During the day, however, I ascended a ridge of rugged rocks, which rose above our tents to a height of more than 500 feet, being desirous of ascertaining to what elevation I should find vegetation. An Alsine was common among the gravel, with two small plants which were not in a determinable state; and on the rocks, to the highest level to which I succeeded in ascending (probably 17,600 feet), the little Allardia continued to occur occasionally. The ridge afforded a good view of the mountains round. The range to the north, which we had still to cross, lay in a semicircle behind; to the east was the continuation of the ridge by which we ascended; and a deep hollow lay to the west. Rugged rock everywhere met the view. The slates which alternated with the limestone were so very brittle that they everywhere formed piles of angular fragments, which filled all the hollows, and formed a sloping talus against every precipice. The view was one not to be forgotten, its desolation far surpassing any conception of waste and utter barrenness which I could have formed.

During the whole day I was never free from a dull headache, evidently caused by the great elevation. Rest relieved it, but the least exertion brought it back again. It continued all evening, as long as I was awake, and still remained in the morning of the 8th, when I rose soon after daybreak to prepare for the journey. A few paces took us beyond the shingly ravine in which we had been encamped, and the remainder of the ascent was throughout over loose angular fragments, the débris of the cliffs on the right. Under the latter we passed, winding round the side of the semicircular bay, till we got to about its centre, when the ascent became excessively steep and toilsome. The exertion of raising the body was very fatiguing, and the last few hundred yards were only accomplished after many pauses. A few large patches of snow lay in hollows along the road; but up to the very crest of the pass there was no trace of perpetual snow, nor even any continuous snow-bed.

THE PARANG PASS. September, 1847.

The summit of the Parang pass is a narrow ridge, covered with large blocks of stone. To the north lay a large field of snow, sloping downwards at a very gentle angle. In this direction the view was limited within two miles by steep rugged mountains, which closed in on both sides. To the right and left also, the pass was overlooked by ridges close at hand. The only direction in which a distant view was obtained was south, where the mountains beyond the Piti river were beautifully seen: from the great elevation at which we stood, their summits were everywhere in view; their elevation was surprisingly uniform, and the whole range was capped with snow. The mountains close at hand presented much the same appearance as I had seen from the rocks above our encampment the day before.

GLACIER. September, 1847.

I reached the summit of the pass, which has an elevation of 18,500 feet, at a quarter before eight in the morning. At that time the temperature was 28°; and a cold southerly wind blew with considerable violence, making us seek the shelter of the blocks which lay around. A small red lichen, (Lecanora miniata,) on the fragments of rock, was the only vegetable production I observed. After an hour's rest, we commenced the descent over the snow-bed, proceeding towards a gap which was visible in the mountains. The snow was hard frozen, and crisp under the feet. Descending steadily without any fatigue, we were soon evidently on a snow-covered glacier. A few fissures were passed, but mostly not above a few inches wide, and none that we could not with ease step over, the widest not exceeding two feet. At a distance of about a mile and a half from the crest, the mountains, which on both sides surrounded the snow-bed in the form of a circle, had so much approached to one another, that they formed a narrow valley, down which the snowy mass continued in the form of a rugged glacier. We now left the surface of the ice, and proceeded along the stony side of the ravine, with the glacier on our left hand, and steep limestone rocks on our right. Blocks of limestone strewed our path as we descended, and numerous small fragments of the same rock covered the edge of the glacier.

About three miles from the summit of the pass the glacier terminated abruptly in a bluff precipice, the height of which was more than 100 feet. Little rills of water were, at the time we passed (9½ A.M.), trickling from every part of the surface, and a small streamlet ran along the edge of the glacier under an arch of ice. The structure was here very evident: broad white bands, and narrower ones of a dirty colour, from the earthy matter which they had absorbed, ran parallel to the slope of the ravine, the arches or loops (so well explained by Professor Forbes in his delightful work on the glaciers of the Alps) being drawn out to a great length.

At the termination of the glacier, we descended from the steep mountain-side, along which we had hitherto travelled, to the flat plain, the continuation of the surface on which the glacier rested. On this descent the first vegetation appeared at an elevation of about 16,500 feet. Two small grasses, Biebersteinia odora, a Lychnis, and a little villous Astragalus, were the plants observed: they grew in the crevices of the rock, and scarcely rose above the ground. None of the species were different from those collected in the mountains of Piti.

THE PARANG VALLEY. ITS VEGETATION. September, 1847.

When we had reached the middle of the valley, so as to be exactly in face of the glacier, we found that a large stream issued from a vaulted cavity at its termination. For some hundred feet the stream ran among large masses of ice, as if the glacier had very recently extended further, and had melted away irregularly, leaving these masses standing. Leaving the glacier, we still followed the valley, which was confined on both sides by steep cliffs. We kept close to the stream, walking over its gravelly bed, and I collected a few more plants as I descended; none, however, new to me. A little Nepeta, four species of Potentilla, a Gnaphalium, several grasses and Carices, and a very small fern, were the species. About three miles from the end of the glacier we found our tents pitched on a small plain, connected with a lateral ravine, and covered with tufts of Dama, and a little species of Alsine in flat tufts, which was quite new to me. The elevation of our encampment was 16,000 feet.

We followed the course of the valley into which we had thus descended, for three days, without meeting with any inhabitants, and through so uniform a country, that it is unnecessary to detail each day's journey. Rugged and rocky mountains, of moderate elevation, principally limestone, bounded the view on both sides. In front we seldom saw more than a few miles; and behind, the view was in general equally limited, though occasionally we could see, up a lateral valley, the peak of a snowy mountain. The valley was almost invariably wide and level, once or twice only interrupted by projecting ridges of low rocks advancing to its centre. Low platforms of alluvium, like those of Piti, occupied the wider parts, their upper angles resting (as in Piti) on the opening of lateral ravines, while their bases were cut into cliffs by the stream.

During these three days we descended from 16,000 to about 14,800 feet. The surrounding mountains were quite barren and desolate. The gravelly plains were covered with tufts of Dama and of the curious tufted Alsine, which formed dense flattened hassock-like masses, of considerable size. The soil was very saline, and as we descended it gradually became more so. In the earlier part of the descent, the alpine forms were the same as those to the south of the Parang pass, and the plants were few in number and much scattered. Lower down, however, more novelty was met with. A little willow was the first shrubby plant, and was followed by Ephedra, Myricaria, and Hippophaë, all much stunted. Still lower there were large patches of green-sward along the stream, generally swampy, and always covered with a saline incrustation. Artemisiæ, Astragali, Gentianæ, and Potentillæ, were the commonest forms, with a number of saline plants, chiefly Chenopodiaceæ, which abounded on the lowest spots.

On the 11th, the last of these three days, the vegetation had quite lost its alpine character, notwithstanding that the elevation was still 15,000 feet. No Biebersteinia was seen, and the little species of Potentilla, Alsine, Saxifraga, Cruciferæ, and Parnassia, were no longer met with. The large Hyoscyamus of Piti (Belenia of Decaisne) had made its appearance, with tall Artemisiæ, a Clematis, a rank-growing Corydalis, Cicer Soongaricum, and other plants in no way alpine. I was much surprised to observe so complete a change in so moderate a descent, and very much interested to find that the alpine flora had so completely disappeared. I regret that I am as yet unable to give my results in more perfect form, the necessary comparison and determination of the species collected still remaining to be done.

CHUMORERI. September, 1847.

During our descent we had gradually taken a more easterly course, and on the 11th our direction was nearly due east. On this day we passed the gorge in the mountains, up which the road turns to the Chumoreri lake, by which Mr. Trebeck had travelled to and from Ladakh. This would have been our most direct route to Le, but we were desirous of visiting the more eastern districts, so as to reach the Indus as soon as possible. The mountains in this gorge suddenly lowered; a wide gravelly plain sloped gently up to a low ridge, which did not appear to rise higher than two or three hundred feet above the level of the Parang river. Beyond this ridge, on the assurance of our guides, confirmed by Major Cunningham, who had on a former occasion travelled along the Chumoreri lake as far as its southern extremity, lies the lake, without any more considerable elevation separating it from the Parang river.

It is much to be regretted that the late period of the season, and the other important objects which we had to accomplish, should have prevented us from crossing this narrow neck of land. It would probably have thrown much light upon the question of the origin and nature of the salt lakes, which are, as is well known, scattered over Tibet, Central Asia, and Siberia. The Chumoreri lake has certainly no outlet, but from the nature of the surrounding mountains, everywhere steep and lofty, there can be no doubt that at one period its waters were discharged at its south end by the narrow valley which we saw from the south side of the Parang river[12]. An accurate determination of the height of the separating ridge above the present surface of the lake, a careful examination of the configuration of the surface at its southern end, and an analysis of the water, which is described as sufficiently brackish to be unpleasant though not absolutely undrinkable, would certainly enable conclusions to be drawn as to the nature of the cause which has lowered the level of the waters of the lake, and so put an end to its discharge.

WILD HORSE. September, 1847.

In the plain which sloped gently upwards from the Parang river towards the Chumoreri lake, we saw for the first time a Kiang, or wild horse, but at too great a distance to enable his shape and appearance to be distinctly made out; and the river, which was interposed between us, prevented our approaching nearer. We afterwards frequently saw these animals, but from their extreme wariness, and the open nature of the country, we were never fortunate enough, notwithstanding repeated trials, to get within gunshot distance of them. They appear to abound at elevations between 14,000 and 16,000 feet, on the open undulating tracts on the summits of the mountain ranges, and to avoid valleys and rocky districts, where they would be liable to surprise.

THE PARANG VALLEY. September, 1847.

To the eastward of the former outlet of the lake, the valley of the Parang river was more contracted than it had been in any previous part of its course. Rocky hills, projecting from the southern mountains, advanced so close to the river, that no passage was practicable along their base, and the road several times ascended several hundred feet to cross these ridges. This obstruction was, however, but temporary, lasting only for a few miles, beyond which the valley expanded into a very wide plain, extending for five or six miles in an easterly direction, by about half that distance from north to south. The borders of this wide expanse were very low platforms, almost horizontal, and not more than from six to ten feet above the river. The middle portion was a plain of gravel, scarcely higher than the level of the stream, and evidently occasionally submerged. Here the river bends rapidly round towards the south-east. The district at which we had now arrived is called Chumurti, and about eight or ten miles to the east of our encampment on the 11th of September, is a village or assemblage of tents called Chumur, from which we obtained a supply of porters, to relieve the party who had accompanied us from Piti. Here also, in accordance with the instructions we had received on leaving Simla, Captain Strachey left us, with the intention of following the course of the Parang river, as far as he conveniently could, and then turning to the left across one of the passes of the great trans-Sutlej chain to the Indus. Major Cunningham and myself, on the other hand, proposed to proceed by the direct, and equally unknown, route to Hanle, and thence to visit the Indus, and proceed to Le.

THE PARANG RIVER. September, 1847.

The Parang river, whose source is in the mountains immediately north of the Parang pass, has, as we have seen, at first a northerly direction, but gradually bends more and more to the eastward and southward, and finally has a nearly south-west course, where it joins the Piti river, nearly opposite Shialkar. Its source, as well as its confluence with the Piti river, are within the British territory; but the most important, because the most populous, part of its course lies within the Chinese border. The boundary of the Chinese district runs nearly from north-east to south-west, passing a little to the west of Rodok, and crossing the Indus at the village of Chibra, where Mr. Trebeck was stopped in his attempt to penetrate up the Indus; thence a little south of Haule, and across the course of the Parang river. It then bends more towards the south, and again crosses the Parang at the point where we were stopped in the end of August, whence its direction is nearly due south as far as Nilang, on the Jahnavi branch of the Ganges.

The Parang river being a tributary of the Sutlej, by crossing the great chain at the Parang pass we had not reached the Indus valley, but had descended into a lateral valley still connected with the drainage of the Sutlej. The great line of watershed between the Indus and Sutlej lay still before us. This chain, which is the prolongation of Kailas, must be called the trans-Sutlej Himalaya, unless the name Himalaya be restricted to the chain south of the Sutlej, in which case the mountains of Lahul, Kishtawar, and Kashmir, would lose their claim to that appellation.

ASCENT TOWARDS LANAK PASS. September, 1847.

Towards this chain, which we were to cross by the Lanak pass, we commenced our journey on the morning of the 12th of September. Our road lay across the Parang river, which flowed in several channels among the wide expanse of gravel which here formed its bed. The morning was bitterly cold, and the water almost icy, to the great discomfort of our porters. The largest stream was perhaps twenty-five feet wide and two and a half deep, with a moderately rapid current. After crossing the river we took a northerly direction, leaving the valley or plain of the Parang river, and ascending an open, almost level valley, bounded by low hills. The mountains on the left, which were interposed between our route and the Chumoreri lake, were the most rugged in sight. In the centre of the plain was the channel of a stream, very shingly, but without water, along which, or on alluvial banks only a few feet higher, we gradually advanced. The level of our camp on the Parang river had been 14,800 feet; and from this we were now gradually but imperceptibly rising. The hills on either hand were rounded and low, but increased in height as we receded from the Parang river. The soil was very barren, and showed many indications of salt. Scattered plants of Salsola were common, with Christolea, a pretty Cruciferous plant, with purple flowers and fleshy wedge-shaped leaves, tasting strongly of horse-radish, which has been described by Decaisne from specimens collected by Jacquemont in Piti. A little white Alyssum, which I had not previously met with, was also very common.

Four miles from the Parang river we reached a flat grassy plain of considerable extent, with deep black soil, in which meandered a very slowly running stream, perhaps twelve feet wide, which seemed to have an outlet by an open valley on our right, and to join the Parang some miles to the east of where we left it. A great part of this plain was swampy, the turf rising in little knolls, but round the edges and in all the higher parts it was covered with a thick incrustation of white efflorescent salt. To the north and east, low gently-sloping hills as barren as ever rose from the edge of the green plain; and in the north-east corner, close to the foot of the hills, a large fountain, discharging copiously clear tasteless cold water, was evidently the source of the stream which flowed over the plain. The grassy turf produced a considerable number of plants, not a few of which were new to me. An Umbellifera, an Aster with large purple flowers, a Saussurea, and two species of Pedicularis, one with white, the other with yellow flowers, were very common, as were also a species of Triglochin, a white Juncus, several Carices, and three or four very beautiful grasses. In the shallow water of the pools scattered over the plain, a species of alga was common, floating without attachment. It was a broad foliaceous green plant, and has been determined by the Rev. M. J. Berkeley to be a species of Nostoc, closely allied to, if not identical with N. commune, a species which occurs in all parts of the globe.

After crossing this plain, and stopping to rest by the fountain, we began to ascend the long slopes of the hills, partly on a level ridge, partly along the wide sloping valleys by which the low hills were separated. Both hills and plain were frightfully arid, the aspect of the country being of an uniform grey colour; and coarse gravel, with scattered stones of larger size, everywhere covered the surface. The ascent was very inconsiderable till towards the end of the day's journey. The distance travelled was about ten miles, and we encamped at about 15,800 feet, on the left bank of a small stream which descended from the north, the borders of which were swampy and covered with green turf, in which the common plants of the country occurred, such as little gentians, Ranunculi, Parnassia, several Polygona and Potentillæ, Carices, and grasses. On the west bank of the stream was a low ridge of clay-slate rocks, while on the right and in the valley was a heap of granite boulders; no doubt an ancient moraine, for the fragments were piled on one another to a great height, and rose far above the stream as well as the ordinary level of the plain.

LANAK PASS. September, 1847.

On the 13th of September we crossed the Lanak pass, which lay before us at a distance of about five miles. From our encampment the mountains appeared easy of access and rounded in outline, and we commenced the ascent by a nearly level walk across the gravelly plain. After a mile and a half we rejoined the stream, and kept along it for a little way. Its banks were green with a narrow belt of turf; and the bed was often rocky, the rock being still clay-slate, notwithstanding the granite boulders everywhere scattered about. The edges of the stream were frozen, spiculæ of thin ice adhering to the herbage. The vegetation was quite alpine, the elevation being certainly above 16,000 feet. A Delphinium, which seemed the same as the D. Brunonianum of the Hangarang pass, a little yellow saxifrage, and a white-flowered species of the same genus, which I believe to be the Scottish alpine S. cernua, an entire-leaved yellow Ranunculus, a Pedicularis with purple flowers, and some grasses, were the most remarkable plants observed.

After a mile, we left the ravine and ascended to the open gently-sloping ground on its left, still rising sensibly as we advanced. The surface was, as usual, dry and gravelly, and Oxytropis chiliophylla and a little Stipa were almost the only plants. We continued nearly parallel to the ravine, and crossed it again a little further on. It was now dry, and its steep stony banks were covered with bushes of Dama. Still gradually ascending, we crossed the same ravine a third time, where its bed was upwards of 17,000 feet. There was again no water visible, but the ground was still moist, the streamlet probably, as is very general in these arid regions, trickling under the surface among the loose gravel. The little alpine nettle, which I had first found on the northern spurs of Porgyul, near Changar, and again on the southern face of the Parang pass, was here common, as were two species of Alsine, which formed dense tufts. A little saxifrage and the Delphinium were also still observed, but all the other plants had disappeared.

Leaving the ravine for the last time, we continued the ascent, which became steeper as we advanced. A rounded ridge lay to our right hand, and we rose nearer and nearer to its crest. Fragments of granite, piled on one another in increasing numbers, covered the steep slopes. Rock in situ was only to be seen in one place; it was still clay-slate, containing a good deal of mica. The top of the pass was nearly level for several hundred yards, and covered with boulders, principally of granite, but a few of quartz and of a trappean rock, quite black and homogeneous. The outline of the mountains was generally rounded, and they rose gradually in both directions above the pass, which had an elevation of 18,100 feet. The view, both towards the direction in which we had come and that in which we were proceeding, was rather extensive, but from the prevailing uniformity of outline and colour it was more striking than beautiful. There were no trees or villages, no variation of surface greater than an occasional grey rock, but everywhere the same dreary sterile uniformity. Nothing could be seen of Lake Chumoreri, which lies at least fifteen miles westward, and is surrounded by mountains, everywhere (except in the direction of the former outlet) higher than that on which we stood.

The occurrence of great accumulations of boulders, of a rock different from that which occurs in situ on the very summit of the pass, was quite conformable to what I had observed on some of the passes between Kunawar and Hangarang. It was not, however, on this account the less puzzling, nor was it till I crossed the Sassar pass, in August, 1848, that I could at all conceive in what way it was to be explained. On this pass, as I shall afterwards relate in detail, a glacier occupies the crest of the pass, descending from higher mountains to the north, and presenting a bluff termination in two directions.

On the summit of the pass I collected specimens of three phenogamous plants, probably nourished by a recently melted patch of snow; for though there was none on the pass itself, nor on the descent on either side, a steep mountain, half a mile to the right, in a due northern exposure, was still covered with snow to at least five hundred feet below the level of the pass. The small quantity of snow seen in the distant view was very remarkable, and the more so as there was no indication of diminished elevation; ridge rising beyond ridge, and peak behind peak, to the utmost limits of view. The three plants which were observed were a little Arenaria or Stellaria, and two Cruciferous plants, one of which only was in fruit. A red lichen, the same as that seen on the Parang pass, covered the stones.

The descent from the Lanak pass was at first gentle, but very soon became steep, to the bottom of a valley in which a small stream of water was running, derived, I suppose, from some small snow-beds in a lateral ravine out of sight, for it almost immediately disappeared under the gravel. Soon after leaving the crest of the pass, we came upon clay-slate rock finely laminated, and dipping south-south-west at a high angle. The valley by which we descended gradually contracted into a rocky ravine, at last very narrow, with high precipitous walls, and full of large boulders. We encamped for the night at its junction with a large stream descending in a rocky dell from the west. Around our camp, on both sides of the stream, there was an outbreak of greenstone, which had upheaved the clay-slate rocks.

On the 14th of September we proceeded along the stream close to which we had encamped the day before. High mountains, whose summits could not be seen from the bottom of the narrow ravine, rose on both sides. The rock on both banks was clay-slate, much altered by heat, often very hard, and with numerous quartz veins; no more greenstone was observed. The stream, copious when we started, gradually disappeared as the ravine widened, and water soon lay only in pools along the gravelly bed. Boulders of granite were abundant all along. After three miles the ravine opened into a wide gravelly plain, skirted by rounded hills of considerable elevation, to which the alluvial platforms sloped very gently on both sides. Christolea, a little shrubby Artemisia, and a small Stipa, were the plants which grew among the gravel.

UNDULATING COUNTRY. September, 1847.

After about a mile and a half, the direction of the plain trending to the south more than was suited to our purpose, we turned to the left, to cross the ridge which ran parallel to it on the north-east. A long gravelly plain, sloping almost imperceptibly upwards, led us to the summit of the ridge, which was not more than two or three hundred feet above the plain we had left. From this pass, for such it was, though an insignificant one, an open valley, skirted on both sides by low rounded hills, ran to the north-east for nearly five miles. The appearance of the country was very remarkable. The hills were all very gentle in slope, and quite rounded in outline, so that the surface was almost undulating. It required reflection on the fact that we were traversing a tract in which the bottoms of the valleys were from 15,000 to 15,500 feet above the level of the sea, to make us aware of the very mountainous nature of the country we were passing through, which was, if any part of Tibet (which I have seen) may be so called, the Table Land north of the Himalaya. The height of the mountains, too, was in fact greater than we had at first been inclined to believe, the gentleness of the slopes making us think the ridges nearer than they really were, and therefore leading to a false estimate of their height. In general they were from 1000 to 2000 feet in height, and their summits therefore from 16,000 to 17,000 feet above the level of the sea.

OPEN VALLEYS. September, 1847.

The open valley along which we now proceeded was remarkable in another point of view. It was quite waterless, and seemed hemmed in on both sides by hills, so that its drainage must take place in the direction of its long axis; at least, no lateral depression could be perceived on either side. About a mile from its eastern end, this plain was lower than in any other part. We had been descending along it from west to east, and we could see that beyond that point it rose gently to the eastward. The surface of the lowest part was covered with a hard shining white clay, without any of the fine gravel which abounded elsewhere. A few tufts of an Eurotia were the only plant which it produced. It was evident that the winter snows which fall on this isolated spot, when melted in summer, finding no exit, form a small lake, till they completely disappear by evaporation.

HANLE PLAIN, ITS VEGETATION. September, 1847.

After crossing this low clayey tract, we ascended gently for nearly a mile in an easterly direction, when the valley terminated very abruptly and unexpectedly in a precipitous descent of four or five hundred feet, the clay-slate rocks emerging suddenly from beneath the gravel at the very edge of the precipice. The road descended in a narrow gorge, which had apparently been worn by aqueous action in the almost perpendicular cliff. On emerging from this gorge, we found ourselves on the border of a very extensive perfectly level tract, seemingly surrounded by hills, and approaching in shape to a circle, though its outline, from projecting ranges of hills, was very irregular. The margins of this plain were dry and gravelly; the centre, as seen from a distance, was green, but in many places encrusted with a saline efflorescence.

Skirting this plain, which lay on our right, while ranges of hills, separated by wide gravelly valleys, occupied the left, we reached Hanle, a Buddhist monastery inhabited by about twenty lamas, built on the summit of a steep hill which rises abruptly out of the plain. We encamped in a ravine at the foot of the hill on which the monastery is built, in which the tents of the wandering population are erected when they bring their flocks into this neighbourhood.

The plain of Hanle, which is not, I think, less than six or eight miles in diameter, resembles very much that curious flat tract which we passed on the 12th of September, on the south side of the Lanak pass; it is, however, much larger in dimensions. Several streams, very tortuous and sluggish, wind over its surface. These were frequently three feet or more in depth, and contained multitudes of small fish, usually about six inches in length, but growing to eight or ten inches at least. They were a species of carp. We tried to eat them, but, though sweet and well-tasted, the bones were so numerous and troublesome that we relinquished the attempt. We were much interested at the occurrence of fish at an elevation of 14,300 feet, a height at which, à priori, it would scarcely have been expected that they would have existed.

The surface of the plain was very saline, and, where not swampy, covered with coarse grasses and Cyperaceæ. It was very uneven, hummocks or knolls being scattered over the surface, which made walking very difficult. These, I presume, were caused by the gradual growth of plants, which, in process of time, formed heaps in spots not covered by water during the melting of the snow in spring. In some parts there were extensive patches of Dama. A species of Elymus and a Blysmus were very abundant. The ground in the vicinity of the streams was swampy, and the coarse grasses of the drier parts were replaced by little Potentillæ, Glaux maritima, Taraxacum, Aster, and a number of Chenopodiaceous plants. In the running waters a Potamogeton and Ranunculus aquatilis were plentiful. The streams, which must, I believe, as in the case of the plain of the 12th, principally derive their supply from springs which break out on the edge of the flat country, all converge to a point at the north-east end of the plain, and, uniting into one, continue their course down an open valley in a northerly direction towards the Indus.

As no section of the bed of this remarkable plain is anywhere to be seen, it is not possible to form an estimate of the depth of its boggy soil, or of the nature of the subjacent deposit. It can scarcely be doubted that it has at one time been a lake, which has been gradually silted up; but it is not easy to conjecture the length of time which has elapsed since it became dry land, in the absence of any knowledge of the nature and contents of the deposits which occur beneath the surface. As an outlet for the waters of the plain exists to the northward, we may infer that the waters of the lake were always fresh.

We remained two days at Hanle, to effect a change of porters, a matter which cannot be accomplished in a hurry in an almost uninhabited country, without unnecessary hardships on individuals. There is no settled population except the monks or lamas; a few stone huts without roofs, which were scattered about the foot of the rock, having no tenants. To the east of the monastery, on the border of the plain, watered by an artificial channel brought with considerable labour from the river, we observed two or three small fields. The grain, which was barley, had been cut and carried away, so that harvest at Hanle was over. The view from the top of the monastery was extensive, as we overlooked the whole plain to the south, and the valley of the Hanle river on the east. The mountains were highest to the east, where a very lofty, steep, and irregular range, with a good deal of snow in some places, separated Hanle from the Indus. To the south and west, the mountains, though high, were rounded.

The rock on which the monastery is built is wholly igneous, but varies from a coarse-grained granite, rapidly decaying, to a dark-coloured greenstone, with large crystals scattered through it. Close to the foot of the hill, the clay-slate was in a few places visible, considerably altered by igneous action, as was to be expected from its proximity to the greenstone.

CHAPTER VI.

Descend Hanle river—Unsettled weather—Encamp on banks of Indus—Upper course of Indus—Pugha ravine—Forest of Myricaria-trees—Borax plain—Hot springs—Borax lakes of Eastern Tibet—Sulphur mine—Pulokanka Pass—Salt lake—Lacustrine clays with shells—Ancient water-mark—Rupchu—Tunglung Pass—Fall of snow—Alluvial conglomerate—Giah—Narrow ravine—Miru—Upshi—Indus valley—Marsilang—Richly cultivated plain of Chashut—Bridge over Indus—Le—Buddhist edifices.

On the 17th of September we left Hanle, en route to Le. Our road lay down the left bank of the river by which the waters of the lake-plain are discharged into the Indus. The valley through which it flowed was open and level, and its slope imperceptible. On the left lay a low range of hills, an irregular mass increasing much in width, as well as in height, as we proceeded northwards, the Hanle extremity being the termination where it slopes into the plain. On the right, a very lofty range, some of the peaks of which were certainly not less than 21,000 feet in elevation, ran parallel to our course, separating the open valley of the Hanle river from the Indus.

HANLE RIVER. September, 1847.

The width of the valley varied from one to three miles. The stream was very winding, crossing from side to side, and often pressing the road close to the spurs of the range on the left. The range on this side was principally clay-slate, with occasional outbreaks of trap, which had in many places converted the stratified rock into a hard red or green jasper. From the immediate proximity of the igneous rock the stratified masses were very much contorted, and no regular dip was observable.

Saline efflorescence occurred everywhere in great quantity in the vicinity of the stream; as a consequence, Chenopodiaceous plants were more than usually abundant, and I collected at least three species of that family which I had not previously observed. The banks of the stream were everywhere bordered by a belt of green herbage, more or less broad, in which the usual species of Ranunculus, Gentiana, Pedicularis, Juncus, Cyperaceæ, and grasses were common. Glaux maritima also occurred abundantly. Two other European plants were found in the swamps along the course of the river, which were very interesting as a proof of the extremely European nature of the flora: these were Hippuris vulgaris and Limosella lacustris. Towards the end of the day's journey, Caragana versicolor (Dama) became very common, covering a large extent of surface, and growing to a much greater size than I had ever before seen, with an upright stem nearly six feet in height. I could scarcely persuade myself that the species was the same as the little depressed shrubs which grew on the passes further south. Two species of Myricaria, both of which I had seen in Piti, also reappeared during the day, so that we were evidently approaching a lower level and more genial climate.

Banks of alluvial conglomerate occurred on the sides of the valley, in the spaces between the projecting spurs of the range on the left hand, on the latter part of the day. The beds were distinctly stratified and very sandy, more or less full of rounded stones, and often passing into pure sand, which was interstratified with the coarser beds. The day was very cloudy and threatening, and a few drops of rain fell for the first time since the 29th of August, the weather during the whole of that interval having been brilliant and quite dry. We encamped eleven miles from Hanle, on a gravelly plain close to the river.

Dining the night the weather did not improve, but continued very cloudy, and on the morning of the 18th the mountains on the right side of the valley were covered with snow, down to within 1500 feet of the plain. The wind blew strongly from the northward, and the day, which was still very cloudy, was bitterly cold, and, to our feelings, extremely uncomfortable. We continued to follow the course of the Hanle river, passing over long gravel flats, which alternated with turfy saline meadows. Several low spurs from the mountains on the left, which projected far into the plain, making the river bend much to the right, were crossed as we proceeded. About ten miles from our morning's camp, we left the course of the river, which turned to the right and entered a rocky mountain gorge, while our road kept its northerly direction. An open valley led us to the crest of a low ridge of trap and slate, from which a very long stony monotonous valley descended to an extensive plain covered with fine mud and saline exudation, on which the only vegetation was a few tufts of Suæda and coarse grass. Crossing this plain, on which the dry clay was in many places deeply cracked and fissured, as if it had till within a short time been under water, or at least swampy, we encamped, at an elevation of 13,800 feet, on the banks of the Indus, here a muddy torpid stream, without any apparent current, about four feet deep and twenty or twenty-five feet wide. There was, however, another channel, separated from that on which we were encamped by a small island.

RIVER INDUS. September, 1847.

So sluggish was the stream at the point where we joined it, that we were for a long time uncertain in which direction the current was flowing; and though we were prepared to find the Indus at the end of our day's journey, the river on whose banks we were encamped was so much less than our anticipations, that we were very unwilling to be convinced that we had really arrived at the great river, to which we had so long looked forward as one of the most interesting objects of our journey. The island in the centre of the channel was a bank of very fine sand or mud, on which large flocks of wild-fowl were resting; it was very little elevated above the surface of the water, which must frequently, I should think, rise sufficiently to cover it. The bank on which we were encamped, though rather higher, was not more than four feet above the water; it was quite vertical, and composed of fine clay, without any intermixture of stones or gravel.

UPPER COURSE OF THE INDUS. September, 1847.

The course of the river Indus, from its source to Le, has hitherto been less known than any other part in Tibet; but as Captain Strachey, a month or two after our visit, descended along it from the Chinese frontier, as far as Le, the unknown portion is now very much reduced. It rises in the mountains north of the lakes of Mansarawer and Rawan Rhad, and runs in general towards the north-east. Moorcroft has described its appearance at Garu or Gartop, where it is a very insignificant stream; but the intervening country is so little known, except by native report, that we can scarcely be said to have an exact knowledge of the upper part of its course. There is in some maps an eastern branch laid down, but of that we have no definite information. From the arid and snowless nature of the country through which it must flow, it is probably a very small stream, but its length may be considerable.

Immediately above the open plain in which we joined the Indus, it would appear to have a very rocky and rugged channel. Such, at least, was the description given to us by our guides of the lower course of the Hanle river, which we left only a few miles before it joined the Indus; and as the mountains to the south-west appeared to close in very abruptly within a very short distance of our encampment, we could not doubt that the open and level plain which we found in this portion of the river's course was of limited extent, and quite an exceptional feature in the character of the country through which the Indus flows. From the great elevation and abrupt slope of the range which runs parallel to the Hanle river on the east, there can be no doubt that the spurs which it sends down on its north-east slope, towards the Indus, must be bold and rocky; and though the hills on the left bank of the Hanle river are much less elevated, yet they rise as they advance to the eastward. The descent of this river too, though very gentle in the upper part of its course, while its valley is broad, is probably very abrupt in the last few miles, where its channel is rocky and its ravine narrow. The elevation of its junction with the Indus is, I believe, about 13,800 feet above the level of the sea.

INDUS VALLEY. September, 1847.

On the 19th of September our road lay in a westerly direction down the Indus. The weather was still extremely unsettled, the sky being cloudy and a violent north or north-west wind continuing to blow in frequent gusts. No rain, however, fell. The plain gradually narrowed as we advanced, and the mountains on the left approached by degrees close to the river. Low grassy plains, covered with a saline incrustation, quite dry, and without any brushwood or tall herbaceous vegetation, skirted the river, the course of which we followed very closely. Indeed, notwithstanding the considerable diminution of altitude, the aspect of the valley of the Indus was more dreary and barren than we had for some days been accustomed to. The rocky spurs were quite bare; and even on the level tracts no vegetation was seen, excepting on the very lowest banks, which were moistened by the river. This utter sterility was no doubt due to the absence of lateral rivulets, the hills which rose on our left hand being stony and steep, and not rising to a sufficient elevation to be covered with perpetual snow, or to accumulate and retain snow-beds in their ravines till a late period of the year.

The rock on the left-hand mountains during the day was quite different from any that had hitherto occurred, being a conglomerate, with rounded stones of various sizes, many of them granite. The matrix was of a very dark colour, and generally extremely hard; more rarely it was a coarse sand, crumbling to pieces. This conglomerate was everywhere stratified, the beds dipping to the south-west, at an angle of about forty-five degrees. During the day the river varied much in width, being seldom less than twenty-five yards, and sometimes as much as eighty. The stream was generally very gentle, not exceeding two miles an hour, except in a few rapids, and the river was in most places fordable. We encamped on the left bank, in a place where it was shallow and wide.

On the 20th of September we continued at first to follow the left bank of the Indus, which gradually assumed a more northerly direction. The mountains on both sides approached much more closely to the river than they had done the day before, and those on the right continued extremely lofty. The river now flowed more rapidly, and was often wider and more shallow; one rapid was not less than 150 yards in width. Banks of alluvial clayey conglomerate were usually interposed between the mountains and the river, forming cliffs which attained not unfrequently an elevation of fifty feet. These were separated by projecting spurs, over which the road passed wherever they advanced so close to the centre of the valley as to prevent a passage along the level plain. Some small streamlets were crossed during the day, and in consequence the vegetation was at times more varied, and at the same time more luxuriant, than it had been the day before. A few bushes of Myricaria were seen on the bank of the river; and in the lateral ravines the ordinary shrubs and herbaceous vegetation were common. The only new plant was a species of Labiatæ, a coarse-growing under-shrub, probably a species of Ballota.

PUGHA RAVINE. September, 1847.

The hard conglomerate of the day before did not again occur, various forms of clay-slate being the prevailing rock. The steep slopes were, however, very frequently covered with a talus of angular fragments, which obscured the structure of the lower portions of the mountains, at the same time that it revealed the nature of the higher strata, which would otherwise have been inaccessible. Red and green jaspery rocks, very hard and brittle, were abundant, with various forms of greenstone, at times closely resembling syenite. These were evidently the same rocks as had been met with in the neighbourhood of Hanle, and along the river for some way below that town. Their recurrence here, therefore, tended to confirm what had for some time appeared to me to be the prevailing strike of these formations, namely, from S.S.E. to N.N.W.

After following the course of the Indus for about eight miles, we turned abruptly to the left, ascending a narrow gorge, in which a considerable stream flowed from the south-west. The slope was, from the first, considerable, and the course of the ravine very winding. Steep rocky cliffs rose precipitously on both sides, and generally approached so close to one another that their tops could not be seen. The channel of the stream was at first stony and quite bare, but after a mile bushes of the Myricaria became common, fringing the stream, but nowhere growing at any distance from it. These gradually increased in size and abundance, and at our camping place, three miles from the commencement of the ravine, they were generally small trees, many of them not less than fifteen feet in height, with stout erect trunks five or six inches in diameter.

The morning of the 21st of September was bright and clear, and intensely frosty, the unsettled weather which had continued since our leaving Hanle having quite disappeared. Our road still lay up the gorge, which had quite the same appearance as on the previous day. High precipices, or very steep banks, hemmed in the stream on both sides. Small trees of Myricaria still continued abundant in the immediate vicinity of the water; elsewhere, all was as desolate as ever. Some of these trees were not less than a foot in diameter; the trunk was generally very short, often branching within a foot of the base. At intervals there was a good deal of alluvium, partly in the shape of coarse conglomerate, partly a fine micaceous sand, filling up the recesses at the bends of the ravine. After three miles, the ravine suddenly expanded into a narrow plain, the surface of which was irregularly undulating, and completely encrusted with salt. As this plain was interesting in consequence of the production of borax, we encamped on the bank of the little stream about a mile from the end of the gorge, and remained stationary the next day in order to examine the nature of the locality in which the borax is found.

HOT SPRINGS. September, 1847.

As the day's journey was a very short one, we arrived at the salt plain by eight o'clock A.M. The air was still quite frosty. While our tents were being pitched on a dry bank a little way above the stream, we proceeded to its bank, and were not a little surprised to find the water quite tepid, notwithstanding the extreme cold of the air. On procuring a thermometer, it was found to have a temperature of 69°. Advancing up the stream, we found that numerous hot springs rose on its banks, and sometimes under the water. The hottest of these had a temperature of 174°. From these springs gas was copiously evolved, smelling strongly of sulphur; and in their immediate neighbourhood the water of the little river had a faintly sulphurous taste, though elsewhere it was quite pure and good. The stream, which was perhaps twenty feet wide, was usually rather deep. Dense masses of aquatic weeds, chiefly species of Zannichellia and Potamogeton, grew in the water, and along the margins their dead stems, mixed with mud, formed immense banks, scarcely strong enough to bear the weight of a man, and yet seemingly quite solid. A small crustaceous animal was common among the weeds, but though I searched with care I could find no shells. The stream was full of fish, which swarmed among the weeds, and darted backwards and forwards in the tepid water in immense shoals. They were generally about six inches in length, and appeared to my inexperienced eye to belong to two or three species, all different from those which had been seen at Hanle. In the hottest water of the hot springs I collected three species of Conferva.

MYRICARIA TREES. September, 1847.

The existence of the tree Myricaria in the gorges between Pugha and the Indus, which had appeared to us at the time very remarkable, was fully explained by the occurrence of the hot springs, and the consequent high temperature of the water of the stream, and was peculiarly interesting as an illustration of the influence of temperature upon vegetation. It may fairly be considered, I think, as a proof, that arboreous vegetation does not cease at great elevations in consequence of the rarefaction of the air, but only on account of the diminution of temperature which usually accompanies increased elevation. The trees of Myricaria, it must be observed, came abruptly to an end with the ravine, none occurring on the open plain. We cannot suppose that the trifling increased elevation caused their disappearance; it seems probable that the narrow walls of the gorge, by concentrating the heat, prevented its escape, and that, therefore, the temperature was more elevated than in the open plain, where the action of winds and free radiation combined to lower it. The occurrence of fish in the water of Pugha, at an elevation of nearly 15,500 feet above the level of the sea, is also very remarkable, and still more strikingly demonstrative of the same fact, inasmuch as it would certainly not have been very surprising that air at that elevation should, from its rarity, be insufficient for the support of life in animals breathing by gills.

At the gorge, where the narrow ravine expands into the lake plain of Pugha, the rock is clay-slate, but the hills which skirt the open plain are micaceous schist, varying much in appearance, often with large crystals of garnet, and crumbling rapidly to decay. On the surface of the plain lay many scattered boulders of a peculiar kind of granite, evidently transported from a considerable distance along the stream; and in all the central parts of the plain, a very remarkable conglomerate in horizontal strata, consisting of angular fragments of the surrounding rocks, cemented together by calcareous matter, was observed.

BORAX PLAIN. September, 1847.

The whole of the plain is covered, to the depth of several feet at least, with white salt, principally borax, which is obtained in a tolerably pure state by digging, the superficial layer, which contains a little mixture of other saline matters, being rejected. There is at present little export of borax from Pugha, the demand for the salt in Upper India being very limited, and the export to Europe almost at an end.

BORAX LAKES OF TIBET. September, 1847.

It has long been known that borax is produced naturally in different parts of Tibet, and the salt imported thence into India was at one time the principal source of supply of the European market. I am not aware that any of the places in which the borax is met with had previously been visited by any European traveller, but the nature of the localities in which it occurs has been the subject of frequent inquiry, and several more or less detailed accounts have been made public. These differ considerably from one another, and no description that I have met with accords with that of the Pugha valley. Mr. Saunders[13] describes (from hearsay) the borax lake north of Jigatzi as twenty miles in circumference, and says that the borax is dug from its margins, the deeper and more central parts producing common salt. From the account of Mr. Blane[14], who describes, from the information of the natives, the borax district north of Lucknow, and, therefore, in the more western part of the course of the Sanpu, it would appear that the lake there contains boracic acid, and that the borax is artificially prepared by saturating the sesquicarbonate of soda, which is so universally produced on the surface of Tibet, with the acid. At least, the statement that the production of borax is dependent on the amount of soda, leads to this conclusion. The whole description, however, (as is, indeed, to be expected in a native account of a chemical process,) is very obscure, and not to be depended upon. Mr. Saunders does not notice any hot springs in the neighbourhood of the borax; but in the more western district described by Mr. Blane, hot springs seem to accompany the borax lake as at Pugha.

It is not impossible that the three districts in which the occurrence of borax has been noticed, which are only a very small portion of those which exist, may represent three stages of one and the same phenomenon. The boracic acid lake may, by the gradual influx of soda, be gradually converted into borax, which, from its great insolubility, will be deposited as it is formed. On the drainage or drying-up of such a lake, a borax plain, similar to that of Pugha, would be left behind[15].

From Pugha, two roads towards Le were open to us. We might either return to the Indus, and follow the valley of that river throughout, or proceed by a more direct route across the mountains to join the road from Lake Chumoreri to Le, by which Mr. Trebeck had travelled on his way to Piti. As we knew that the Indus route would be surveyed by Captain Strachey, who was desirous of following the course of the river as far as practicable, we preferred the more mountainous road, and, therefore, on leaving our encampment at Pugha, on the morning of the 23rd of September, we continued to ascend the valley of the little stream, on the banks of which we had been encamped. For the first two miles the plain was nearly level, and similar in character to what has just been described, hot springs being observed at intervals.

SULPHUR MINE. September, 1847.

Two miles from our encampment, we stopped and examined the spot whence sulphur is obtained, at the base of the mountain slope on the north side of the valley. Ascending a few feet over a loose talus of shingle, which skirted the bottom of the hill, we found two narrow caverns in the slaty rock, apparently natural, or only a little widened by art, roughly circular, and less than three feet in diameter at the mouth. One of these caverns continued a long way inwards, nearly horizontally, but it contracted considerably in diameter, and was so dark that we could not penetrate far. The rock was principally gypsum, interstratified with very friable mica-slate. Sometimes the gypsum was amorphous and powdery, at other times in needles two or three inches long, perpendicular to the strata of slate. The sulphur was in small quantities, scattered among the gypsum, and was more abundant in the lower beds. It was frequently in very perfect crystals, not, however, of any great size.

The air which issued from these funnel-shaped apertures was very sensibly warm, and had a strongly sulphurous odour. Unfortunately, we had not anticipated the necessity for observing the temperature, which was not by any means oppressive, and was only remarkable in contrast with the extreme cold of the external air.

In the neighbourhood of the sulphur-pits, the hot springs along the course of the stream were very numerous, evolving much gas. A little higher they ceased altogether, and the upper part of the plain was without any springs, as was evident from the quantity of ice by which it was covered. For more than a mile it was a dead level, and very swampy; but afterwards the valley became gently sloping and gravelly, the little stream being often hidden under the pebbles. Large boulders of the same granite which we had observed the day before, were scattered over the surface. The vegetation in this valley was extremely scanty, a few scattered tufts of Dama, and some shrubby Artemisiæ, were occasionally seen, but the herbaceous vegetation had been almost entirely destroyed by the intense morning frosts, which had for some time been of daily occurrence. On the latter part of the day's journey the rock on the mountain-side changed from mica-slate to gneiss, of which very lofty scarped cliffs rose abruptly on the right hand. We encamped on a level spot, after ten miles of almost imperceptible ascent.

Next morning we continued to ascend the valley, which was now very rugged, from masses of boulders, which were heaped one on another to a very great thickness. The stream had cut for itself a narrow channel, nearly a hundred feet in depth, the walls of which were entirely composed of huge incoherent masses of rock, all more or less angular. A walk of three miles brought us to the crest of the pass, which was nearly level and grassy for about a mile; its elevation was about 16,500 feet. The pass (Pulokanka La) is a very deep depression in the axis of the chain, which runs parallel to the left bank of the Indus, separating the waters tributary to that river from those which join the Zanskar river, some of the feeders of the latter springing from the valleys on the western slopes of these mountains. The hills right and left of the pass rise very boldly into rugged masses, contrasting strongly with the level plain which constitutes the pass, in which the watershed is scarcely perceptible.