On the 2nd of July I crossed the Zanskar river to the village of Yulchung (13,700 feet). At Nira, besides the usual crops of barley, there was a good deal of buckwheat, which was just above ground. The fields were bordered, as usual, by a rank vegetation. A Nepeta, very like N. Sibthorpiana, was quite new to me, and a tall erect Wahlenbergia, with very large pale greenish-blue flowers, and coarse, somewhat fetid leaves, was very abundant, just coming into flower; the rest of the plants observed were the same as in the upper part of Zanskar. The stream which ran by the village had in some places spread out into a marshy meadow, in which a large pink-flowered Cardamine or Dentaria occurred plentifully, with Orchis latifolia? a white Juncus, and many common plants.

Below the village the descent was bare and stony, and extremely abrupt the whole way down to the river; the Tibetan rose was in full flower on the road-side. The river did not come into sight till it was close at hand, the bottom of the ravine through which it flowed being narrow and rocky. A common wooden bridge, without side-rails, forty or fifty feet above the surface of the water, was thrown over at the narrowest part, where the stream was hemmed in by high rocky walls, and was, I think, not more than forty feet broad. The current was rapid, and the water much discoloured. The course of the river at the bridge was easterly, but below, after a slight bend to the south of east, the valley seemed to take a more northerly direction, and above the bridge it came from the south-west. The banks of the river did not seem to be at all practicable, and I was informed that it was only when the river was frozen that travellers could proceed down it to Le. Accounts differed much as to the length of time required for the journey, and I could not discover that any of my party had ever travelled it, so that I presume the route is not very much frequented.

Immediately after crossing the river, a long, steep, utterly barren ascent commenced over stones and shingle. A deep ravine, with a small stream at the bottom, lay to the right of the road, beyond which were the lofty rugged precipices which had been so conspicuous from the heights the day before. At about 13,000 feet I gained the summit of a projecting ridge, which rose, a little to the right, into a rocky peak, and then sank abruptly down to the ravine. The road then dipped into a hollow filled with large boulders and fragments of rock, perhaps of glacial origin, and rose again more gradually to a second ridge, in the hollow beyond which lay the village at which I had determined to encamp, its lowest houses overhanging the deep ravine on the right. The elevation not being materially different from that of Nira, the plants of the cultivated grounds were the same. Potentilla anserina was very plentiful, and remarkably luxuriant.

The rocks during the ascent were chiefly a very hard but very brittle quartz or schist, alternating with loose crumbly slates, and a little limestone. I diverged a little from the direct road, to visit an iron mine, and to see the process of smelting. The ore was yellow ochre, occurring in a breccia-looking conglomerate situated on the flanks of a steep narrow ravine. There were two smelting furnaces, built of stone, of a conical shape, three feet in height, and about six inches in diameter at the top. The fuel employed was charcoal, and no flux was mixed with the ore.

On the 3rd of July, I crossed Singhi La, the pass which I had seen so distinctly on the 1st. The ascent commenced at once from the village of Yulchung, over dry rounded hills, at the same time receding considerably from the deep ravine on the right. No rock in situ was visible on the earlier part of the ascent, the hills being entirely covered with coarse gravel and small stones, among which a spinous Astragalus and a species of Polygonum were the predominant plants; a glabrous Artemisia, a little Euphorbia, and the prickly Statice, were also frequent. After about 1000 feet of ascent, plants of the alpine zone began to appear. Afterwards the ascent was more gentle, over similar ground, till I attained an elevation of about 15,000 feet; at which height the road was for some distance nearly level, winding round a deep bay or hollow in the mountains, with high hills rising on the left hand, and the deep ravine still on the right. Several small streams were crossed, and many alpine plants seen, all familiar to me, except a species of rhubarb, which grew among the shingle in considerable quantity, and which is probably an undescribed species.

After completing the circuit of the deep bay, the ascent recommenced, but was not at all rapid, till within a few hundred yards of the top, when a short steep pull occurred. On the latter part of the ascent, from the loose, stony nature of the soil, vegetation was very scanty; and at the top, which was rounded, there was absolutely none. The elevation was 16,500 feet. Several large patches of snow occurred on the south side when close to the top, but not continuously. The view was extensive to the south, embracing a considerable portion of the great snowy range north of the Chenab, which, from the great elevation of the spot on which I stood, as well as of the intermediate ranges, and from the much smaller quantity of snow on its northern face, looked much less imposing than it does when viewed from the Indian side. Right and left were huge rocky peaks, and in front the view was obstructed by mountains close at hand, except to the north-west, in which direction a long gently-sloping valley was visible, running between two steep ridges, along which, I was informed, the next day's journey lay. From the top of the pass I attempted to form an estimate of the height of the neighbouring ranges, taking the quantity of snow as a guide, and it appeared to me that they were in general between 19 and 20,000 feet, a few isolated peaks only exceeding that altitude. Such guesses, however, are necessarily extremely vague.

Quartzy rock, slate, and limestone, alternated during the ascent; and near the summit of the pass the limestone evidently contained organic remains, perhaps coralline, though the traces were not sufficiently distinct to enable me to decide the point. The fossils were not observed in situ, but the angular fragments in which they occurred did not appear to have been transported from any distance.

On the north side of the pass a snow-bed commenced at the very crest, down which the descent was very steep for a few hundred yards. The snow was very soft, and was rapidly melting, but it possibly covered a permanent mass of ice, as it terminated abruptly, and the valley at its base was wide and but little inclined, with only a few patches of snow. The ground near the snow was swampy, owing to the rapid thaw. Here a little sweet-scented Primula was abundant, with one or two more alpine plants. The road followed the course of a wide arid valley, descending very gently. Two species of rhubarb were common, and a dwarf willow fringed the margins of the stream.

As I advanced, the valley gradually narrowed, and on the right high precipitous rocks ere long overhung the stream, so that I crossed to the left bank, and, instead of keeping on the bottom of the valley, proceeded horizontally along the hill-sides. A little further on, the stream, which had hitherto had a north-west course, turned suddenly to the north, and entered an extremely narrow rocky ravine, which to all appearance was quite impassable. Here the road turned abruptly to the left, and ascended to cross a low ridge. On attaining the summit an open valley was seen 1000 feet below, which at its lower extremity contracted into a fissure precisely similar to that just described; and as the two ravines were only separated by a narrow rocky ridge, which rose to the north into a high cliff, there can be no doubt that the two streams joined a mile or two below. Descending gradually into the valley, I encamped at the village of Phutaksha, at an elevation of about 14,300 feet.

Notwithstanding its great elevation, the valley of Phutaksha was partially cultivated. The fields formed a narrow belt parallel to the stream, along which they extended almost up to 15,000 feet, but the crops were scanty. The wild plants of the borders of the cultivated land were the same as those common in Zanskar, and grew with great luxuriance along the margins of the irrigation streamlets. Alluvial boulder clay was common in the valley; and I saw also a great deal of the fine cream-coloured clay, which I have elsewhere noticed as being probably of lacustrine origin. The occurrence of this clay at an elevation of upwards of 14,000 feet is rather uncommon, and here, as well as elsewhere, appears to be accompanied by such a conformation of the mountains as to render the former existence of a small lake probable. Below Phutaksha, as I have already observed, the ravine of the little stream is exceedingly narrow and rocky, and as likely as any other part of Tibet to have been blocked up by alluvial deposits so as to form a lake.

On the 4th of July my road lay up the valley. The banks of the little stream were lined with most beautiful green turf, producing all the characteristic plants already mentioned. I took the right-hand branch of two which here united, and, on looking up the other, observed that the snow-line on the northern slope of the mountains, at its head, was very considerably above the level at which I stood; its height, where lowest, seemed to be about 16,000 feet. In one small side-ravine there was an incipient glacier. After leaving the cultivated lands the valley became extremely stony and barren, fragments of a brittle limestone rock being everywhere scattered about. The vegetation changing to that of the alpine zone, several new species of Astragalus and Phaca were collected. Following the streamlet almost to its source, the road afterwards ascended to the top of a steep ridge, elevated probably a little more than 16,000 feet; this ridge was rocky, or covered with shingle of a dark slate, which had succeeded to the limestone. The yellow Thermopsis was almost the only plant which grew on the summit, from which I had a fine view of the pass crossed the day before, and of the range of mountains I had left; but to the north there was no distant view, the valley bending abruptly to the right.

From the top of the pass I descended rapidly along a deep valley, generally at some height above the stream, to the village of Hanupata, elevated 13,100 feet. This valley was throughout barren and stony, and became very narrow in the lower part. Dama was very plentiful, but otherwise there was little novelty in the vegetation, except along the bank of the stream in its upper part, where I made a rich collection of small alpine species. A large-flowered Aster, a white Pyrethrum, and a little Pedicularis, were the new species obtained. In the lower part of the valley willow and Lonicera as usual appeared; and when close to Hanupata, I met with a shrubby species of Labiatæ (perhaps a Ballota) which is an extremely common plant in the valley of the Indus from 7000 to 14,000 feet, but seems never to occur far from that river.

On the 5th of July I proceeded down the same valley to Wandla, a distance of about eleven miles. The fields of Hanupata occupied only a narrow strip along the bank of the stream, the sides of the valley being steep and rocky. The crops were much further advanced than any I had hitherto seen; the barley in particular was very luxuriant, and one field was already in ear. Along the margins of the field there was the same rank herbage as usually occurs in similar situations. Lucerne and melilot, both seemingly the common European species, were very plentiful. Poplars and willows were cultivated; and I observed some large juniper-trees. Beyond the cultivation the valley became very narrow. The bed and banks of the stream were gravelly, and on the latter grew a dense thicket of Myricaria, Hippophaë, willow, and rose. After two or three miles there was not left space even for these, the mountains coming so close together that in many places there was not room to pass between them and the water. The current was too rapid for fording, so that it repeatedly became necessary to ascend to a considerable height in order to effect a passage. One of these ascents was not much less than 1000 feet perpendicular, up a narrow lateral ravine, and then over a very steep bank of loose shingle, descending again with great abruptness to the water's edge. The road also crossed the stream several times.

In one place I observed a very remarkable natural tunnel, where the stream flowed below a solid mass of conglomerate rock, which formed an arch obliquely across it. The conglomerate was exceedingly hard, and rested on both sides on very soft friable slate, by the excavation of which, by the action of the stream, the tunnel appeared to have been formed. The original channel of the stream was still visible six or eight feet higher than its present level a little to the right. The ravine continued narrow and rocky for nearly seven miles, but during the last two of these the road lay high upon the mountain-side, and was tolerably level and good. Near the end the valley became wider, and several small patches of cultivation appeared, with a few apricot-trees; and a double yellow rose was planted near some of the houses. The last mile of the day's journey was entirely through very rich and luxuriant cultivation, which was further advanced than any I had yet seen.

The elevation of Wandla is only 11,000 feet, and the heat of the sun was very oppressive. On the latter part of the march, many plants of the Indus valley which were familiar to me from my journey of the year before, but which I had not seen during my present visit to Tibet, made their appearance. Echinops and Nepeta floccosa, Mulgedium Tataricum, a large and handsome yellow Corydalis, Capparis, and numerous Chenopodiaceæ were abundant. The leaves of Tussilago Farfara were common along the water-courses; in the corn-fields a little viscid Cerastium (Lepyrodiclis) was only too plentiful. By far the most conspicuous plant was the rose (R. Webbiana), which, in the rich and well-watered soil of the cultivated plain, grew most luxuriantly, forming dense almost spherical bushes, many of which were at least fifteen feet high, as much in diameter, and bushy down to the ground. They were now in full bloom, and the foliage was almost entirely concealed by the profusion of bright red flowers.

I was obliged to remain a day at Wandla, owing to the serious illness of one of my servants, who, though a native of a mountainous country, had suffered much more on the high passes than any of the inhabitants of the plains of India, and was now so much exhausted as to be unable to move. On the 7th, however, I proceeded towards the Indus, not a little glad to be at last within a day's journey of that river, as I was considerably later than I had originally calculated, not having made allowance for the very rugged nature of the country between Zanskar and Le.

The valley of Wandla, I was informed, contracted again into a rocky ravine a very little way below the village. This ravine was not quite impracticable, but the stream had to be forded very frequently; and as it was at least four feet deep, I was recommended to follow another route, a little more circuitous, but free of difficulty. For the first mile I proceeded up an open valley, which joined at a right angle from the west that which I had descended on the 5th. I then turned to the right up a very sterile ravine, with much saline efflorescence; in a few places a small streamlet trickled among the stones, but for the first part the channel was quite dry, the water filtering underneath the gravel. The sides of the ravine were bare and shingly and without vegetation, except at the entrance, where a Corydalis, thistle, and one or two other plants occurred sparingly. On the most stony parts Güldenstädtia cuneata, Benth., was common, and here and there in the gravelly channel was a bush of Myricaria (not M. elegans, but a smaller and much less handsome species). After a gentle ascent of about two miles, I gained the head of the ravine, and crossing a stony ridge not high enough for alpine plants, descended another valley on its north side, which, though at first if possible more barren than the ascent, soon became somewhat green with willow-bushes and the ordinary plants. After descending perhaps a thousand feet, I reached an extensive tract of cultivation, just above which, in another ravine, lay the village and monastery of Lamayuru, of which a circumstantial account has been given by Moorcroft[20]. At this place, I joined the road from Kashmir by Dras to Ladak, which has been repeatedly traversed by European travellers, and is particularly described in Moorcroft's Travels.

Below this village the valley contracted, and was for some distance full of immense masses of lacustrine clay; lower down it became a narrow rocky ravine. The road descended with great rapidity till I reached the Wandla stream, which I had left in the morning; it was afterwards less steep, following the banks of that river through a winding rocky valley to its junction with the Indus, which was not seen till close at hand. The valley of the Indus, where I entered it, was very barren, with bare rugged mountains on both sides. A stony platform of alluvial conglomerate usually intervened between the mountains and the river, over which my road lay for about three miles up the river, to a good wooden bridge, defended on the north side by a small, very indifferent fort. By this bridge I crossed to the right side of the river, and a mile further on reached the village of Kalatze (or Kalsi, as it is commonly pronounced), at which I encamped.

In the lower part of the Wandla ravine, the clay-slate rock became much indurated, and alternated with a very hard conglomerate, the matrix of which had a semi-fused appearance, while the pebbles which it contained were all rounded. This rock is very similar to, and probably identical with, that of the Giah ravine north of the Tunglung pass, and of the upper Indus. A modern conglomerate, with an indurated sandy and calcareous matrix, in horizontal beds, rested unconformably upon the more ancient rock, but afforded no indications by which I could form an opinion of its exact age.

The elevation of my tent at Kalatze I made to be 10,400 feet; but I was encamped at the highest part of the village, and the bed of the river was not much above 10,000 feet. The cultivated lands, which are very extensive, lie on the top of a thick platform of alluvium, through which the river has excavated a deep broad channel. The lands of the village slope gradually from the base of the mountain to the edge of the cliff overhanging the river, and the fields are made into level terraces by walls of stones from three to six feet in height. Numerous streams of water are conducted through the fields for irrigation, upon which cultivation in Tibet entirely depends. The crops had an appearance of great luxuriance: they consisted of wheat and barley (both in full ear, the latter even beginning to turn yellow), buckwheat, peas, and oil-seed (Brassica Napus). Fruit-trees were abundant, chiefly apricots; but there was no deficiency of apples, pears, walnuts, and mulberries. Along the water-courses and on the edges of the fields grew plenty of wild plants, many the same as occur everywhere in Tibet, but, from the diminished elevation, numerous novelties were observed. A Clematis, with dingy brownish-orange flowers, straggled over bushes; a shrubby Ballota and a Perowskia covered the walls; Iris, Capsella, Veronica biloba and agrestis, Lamium amplexicaule, Mentha, Potentillæ, Plantago Asiatica, Thalictrum, and numerous other plants grew along the water-courses; while in the fields among the corn the weeds were much the same as are common in Europe and in the plains of India in the cold season; Vaccaria, Silene conoidea, Stellaria media, Malva rotundifolia, and Convolvulus arvensis being plentiful.

On the 8th of July, I marched to Nurla[21], about eight miles up the valley of the Indus. After leaving Kalatze, the whole day's journey was quite barren, the road usually lying on the top of an alluvial platform. Just beyond Kalatze, a large stream had cut a deep ravine through the platform, showing it to be composed of large incoherent water-worn stones, mixed with gravel and clay. The mountains on both sides were steep, rocky, and bare. The vegetation on these platforms was scanty: Boragineæ and Chenopodiaceæ were the two prominent orders; Nepeta floccosa, a little Hyoscyamus, Güldenstädtia, a large and handsome Corydalis, a Matthiola, and several Astragali, Cruciferæ, and Artemisiæ, were also prevalent. Of grasses, Stipa was the most common, but several sub-tropical forms were observed, which were interesting and somewhat unexpected. A species of Cymbopogon, and an Andropogon allied to A. Ischæmum, grew among rocks close to the river. In similar places I met with two species of Vincetoxicum, one a twiner, and the other erect; Tribulus, too, was common on the most barren spots. At Nurla, the cultivated lands are very extensive: the crops and fruit-trees as at Kalatze; some of the barley was nearly ripe. The common bean seemed a good deal cultivated, usually intermixed with wheat; Lathyrus sativus was also a common crop[22].

Behind the village of Kalatze, rounded hills of moderate elevation were capped with incoherent beds of sand and boulders of considerable thickness, horizontally stratified; similar beds, sometimes indurated into a soft sandstone rock, occurred at intervals throughout the day. Boulders of granite were abundant in the alluvium and on the surface of the platforms, derived, I believe, from the axis of the chain separating the Indus from the Shayuk. These transported masses of granite were not observed anywhere between lower Zanskar and the Indus; it may therefore, I think, be inferred that the superficial alluvium (which, where the two occur together, generally covers the lacustrine clays) has been deposited since the present river system was in full operation, and is not, as I at one time conjectured, analogous to the drifts of Europe. The ancient rocks between Kalatze and Nurla were alternations of friable slate with indurated conglomerate and grey sandstone.

Between Nurla and Saspola, to which place (eleven miles) I marched on the 9th, the valley of the Indus was narrower than before, as well as more rocky. The rock was chiefly grey sandstone. The road frequently ascended to some height in places where the banks of the river were too rugged to permit a passage. On the 10th of June I proceeded to Nimo, ten miles further. At Saspola the road leaves the banks of the Indus, to ascend a barren valley, among hills of loose conglomerate. At first, the banks of the little stream were green and turfy; but after about a mile I entered a dry stony ravine, along the bed of which the road gradually ascended. The rocks were clay-slate, conglomerate, and sandstone, and all the hills were capped with modern alluvial clay conglomerate. Granite boulders occurred abundantly, and marks of the action of water were seen on the rocks far above the reach of the present streams. At the summit, which must have been nearly 1000 feet above the Indus, I emerged suddenly upon a wide and open gravelly plain. To the right, a number of low hills concealed the course of the Indus; to the left, the mountain range had receded to some distance, and could be seen to be here and there tipped with snow. The road lay for several miles over this barren plain, which was entirely alluvial, descending afterwards very abruptly into a deep flat-bottomed hollow, excavated out of the soft conglomerate by a considerable stream. In this hollow, quite concealed till close at hand, was the village of Bazgo, with a long narrow strip of cultivation along the margin of its stream. Following the course of this valley till near the Indus, I then ascended its left bank, and emerged upon another extensive alluvial platform, high above the river, but parallel to it. At the east end of this platform was the village of Nimo, the termination of my day's journey.

From this place my journey of the 12th brought me to Le, about twelve miles. About a mile above Nimo the Indus is joined by the Zanskar river. The valley where the two rivers unite is very rocky and precipitous, and bends a long way to the south. The road to Le does not follow the river, but ascends among gravelly ravines behind the village, and emerges on a wide open plain, which, as on the previous march, is interposed between the northern range of mountains and the present channel of the Indus. The height of this plain above the river was at least 1000 feet; it was lowest in the centre, sloping up not only towards the mountains to the north, but to a range of round-topped hills of moderate elevation, which overhung the valley of the Indus, sinking on their south face very abruptly down to the river. The higher mountains were chiefly granite, with a few interposed beds of slate dipping at a high angle. The granite exhibited the usual tendency of that rock to decay in spheres, or rather in irregular-shaped masses with rounded angles.

In proceeding along this plain, the road at first rose almost imperceptibly, but after two miles I reached the highest part of it, from which it sloped down towards the east. From this point the course of the Indus in front of Le, and to the south-east for many miles, was finely seen. The river runs through a wide valley, but the range of mountains to the north sends down many rugged spurs, which, in the shape of low rocky hills, advance close to the river. On the south or left bank, on the contrary, a wide, open, gently-sloping plain extends to a considerable distance. From the highest level of the plain a long gradual descent brought me to the Indus, to which it was necessary to descend in order to get round one of the spurs just referred to. It is here a tranquil but somewhat rapid stream, divided into several branches by gravelly islands, generally swampy, and covered with low Hippophaë scrub. The size of the river was very much less than it had been below the junction of the river of Zanskar, the latter appearing to contribute considerably more than half the amount of water. At the point of the low spur lay the village of Pitak, on an isolated hill, surrounded by extensive deposits of cream-coloured lacustrine clay. From this village there is a gradual ascent of about four miles to the town of Le, which is built on a low hill at the upper corner of a wide open valley.

The bed of the Indus at Pitak, below Le, has an elevation of about 10,500 feet above the level of the sea, but the town is at least 1300 feet higher. Its sheltered situation, in a hollow surrounded by hills, and facing the south, compensates to a certain extent for this increase of elevation; still the crops are very much inferior to those on the banks of the Indus. There are but few trees, the apricot being the only fruit-tree cultivated, and it does not seem to thrive. Water is plentiful in the valley, and is conveyed through the cultivated lands in deeply-cut canals or trenches, faced with walls of stone. Natural meadows of tall grasses, intermixed with luxuriant lucerne and melilot, are common along the banks of the river, especially above the town.

The vegetation in the vicinity of Le scarcely differed from that of the Indus at Kalatze. The most abundant families of plants were Chenopodiaceæ, Labiatæ, and Artemisiæ, which covered the barren and stony tracts; the Boragineæ, so abundant throughout Tibet in early spring, had already quite dried up and disappeared. In the meadows tall species of Thalictrum, Silene, and Heracleum, were coming into flower, and in swamps Veronica Beccabunga and Anagallis, Limosella, and a yellow Pedicularis, were the most abundant plants.

At Le I had the pleasure of meeting Captain Strachey, who had spent the winter there, and had returned shortly before my arrival, from an exploring journey to the eastward. After a week's stay I set out for Nubra on the 19th of July, crossing the lofty chain separating the two rivers by the pass directly north of Le, which, during the summer months, presents no difficulty, and is therefore preferred as being the most direct. The pass is distinctly visible from the town of Le, to which it appears very close, though the distance is at least ten or twelve miles. I did not attempt to cross it the first day, but encamped as far up on the southern face as I conveniently could, so as to reach the top early in the morning. At starting, the road lay for about three miles through an open valley, partly cultivated, and with a good deal of swampy ground. Higher up, the valley contracted into a barren ravine, with a narrow strip of green along the margin of the stream. About half-way, the road left the bottom of the valley, and for the remainder of the march I proceeded along the bare side of the mountain, ascending very rapidly. There was a striking change in the vegetation as the height increased. On the lower slopes Cicer and Statice were abundant, with several Astragali; on the latter part of the ascent many alpine plants were observed, belonging to the genera Corydalis, Elsholtzia, Potentilla, and Draba. A very small violet was extremely plentiful in the crevices of the rocks, and among stones, after I had reached 15,000 feet. I encamped at about 15,700 feet, on a level piece of ground, a few hundred feet above the bottom of the valley.

On the 20th I crossed the pass, starting about sunrise. The morning was intensely frosty, and the stones and vegetation near the water were encrusted with ice. The path lay close to the stream, ascending somewhat rapidly among the green turf which grew along its margin, in which I found many little alpine plants, among which, a large-flowered Aster and a small poppy with still unexpanded flowers were the most conspicuous. The last part of the ascent was extremely steep, among immense angular granite boulders, with here and there a little snow in the crevices. Here a most elegant sweet-scented species of Primula was common, so firmly fixed in the frozen mud, that I could with difficulty procure a specimen. Except in very small patches, there was no snow till within two hundred yards of the top of the ascent, for which distance it was continuous, but very soft, and evidently melting rapidly. The crest of the pass was a narrow ridge of large spheres of granite, seemingly quite detached from one another, but which had probably been formed on the spot they now occupied by the peculiar decay characteristic of that rock.

The continuation of the ridge on both sides was for some distance very little more elevated than the pass itself, the height of which was 17,700 feet. To the south, the view was very extensive, embracing a great extent of snowy mountains, with numerous lofty peaks, as well as a part of the Indus valley, and the town of Le, immediately below; to the north it was much more limited, as hills close at hand completely excluded all distant view, except directly in front, where one snowy peak could be seen a long way off, evidently beyond the Shayuk.

On the north side of the pass snow commenced at the very top, and continued for at least 1200 feet of perpendicular height. The descent for this distance was extremely steep, over a snow-bed, which appeared to cover an incipient glacier. About 1200 feet below the top I came to a small oval-shaped lake, completely frozen over; a little higher up I had passed a small bare piece of rock projecting through the snow, and perhaps thirty feet long, on which the beautiful blue-flowered Nepeta multibracteata, Benth., had already put forth its flowers. Beyond the frozen lake the descent became at once much more gentle, and was partially free of snow. The path lay over a vast accumulation of angular stones, which appeared to have fallen from the rocks above. Many parts of the valley were swampy, evidently from recently melted snow, and in such places the Primula, noticed on the ascent, occurred in great abundance, its scapes rising to the height of six to eight inches, and bearing large globes of deep rose-coloured flowers. Among the loose stones Nepeta multibracteata was common. About three miles from the top I passed the end of an exceedingly well-marked moraine, which must have been deposited by a glacier at a time when, from increased cold, these masses of ice stretched down much further than they do at present. The remainder of the descent was again more abrupt, but very bare, stony, and uninteresting. A single tree of Juniperus excelsa grew in one of the ravines, and below 14,000 feet a species of berberry, with very small leaves, was common on dry stony ground. I encamped at the small village of Kardong, at 13,500 feet. The cultivation round this village was on a level plain without any terracing.

On the 21st I proceeded to Karsar, a village on the bank of the Shayuk river, distant about nine miles. A few hundred feet above the village of Kardong the alluvial boulder clay had begun to occur in the valley, and around the village, which occupied the end of a lateral ravine, it was already very thick. From Kardong to the Shayuk this alluvium continued in great quantity, forming elevated platforms, sloping very gently from the mountains, and faced by steep, often quite perpendicular cliffs. Where lateral ravines joined the main valley the alluvium was deeply excavated by the little streams which traversed them, and the road descended abruptly by steep and curiously winding paths down the cliffs of clay, and among piles of boulders, to re-ascend to the platform beyond the stream. Such a ravine, of great depth, occurred just below Kardong. After crossing it the road lay over the surface of the clay platform, which was nearly level, and consequently at an increasing height above the bottom of the Kardong valley, which rapidly diminished in elevation. This platform was extremely barren, and quite devoid of water. Here and there isolated rocky masses rose up through the alluvium. The rock was peculiar, being very hard, and, as it were, porphyritic, with a black, basaltic-looking matrix, quite homogeneous, in which numerous white specks were diffused. In hand specimens and boulders, and even on a near view of the hills, this rock appeared quite an igneous rock, but when an extensive section was exposed, it could be seen to be distinctly stratified.

When within a short distance of the Shayuk valley, though still high above it, the road turned to the left, and, leaving the alluvial platform, proceeded among rugged rocky hills, in a direction parallel to that river, at the same time descending somewhat rapidly to a platform of modern lacustrine clay and conglomerate, which filled up the whole of a deep recess in the mountains facing the Shayuk, to a thickness of at least 1000 feet. The village of Karsar, at which I encamped, lies in a deep ravine, excavated out of the clay formation by a considerable stream, on both sides of which, for nearly a mile, there is a belt of cultivation, very narrow where the stream issues from the mountains, but gradually widening as it descends. Owing to the sheltered situation, from the great height of the cliffs of clay on both sides, the crops were exceedingly luxuriant, and fruit-trees were plentiful, principally apples and apricots. Some very fine walnut-trees also occurred.

From the same cause the herbaceous vegetation was particularly rich, and I met with many species which were new to me. The banks of the stream, from the point where it issued from among the mountains, were everywhere bordered by large bushes of Myricaria elegans, now adorned with masses of sweet-scented rose-coloured flowers. In the lower part of the village-lands there were shady plantations of poplar and willow, which seemed to be occasionally irrigated, in order that they might produce a rich natural pasture. In these groves Euphrasia officinalis, species of Gentiana, Ranunculus, Potentilla, and Carum grew most luxuriantly; a tall but very small-flowered Pedicularis was also very common. No less than three species of Orchideæ occurred, a family which more than any other dislikes dryness: these were Orchis latifolia, an Epipactis, and an Herminium. Many of the weeds of the cultivated fields were also new and interesting: a Hypecoum, an Elsholtzia, and some species of Polygonum, were those I particularly noted.

The lacustrine formation of Karsar consists mostly of very pure white clay, horizontally stratified; but at the lower end of the ravine, where it is about to expand into the open plain of the Shayuk, a tolerably solid but still very friable sandstone, the strata of which were also quite horizontal, occurred under the clay. I saw no fossils, but when the clay is examined with care, they will probably be occasionally detected. At all events, as this clay formation is at least a thousand feet thick, if we take into consideration the open nature of the whole valley of Nubra, there can be no doubt that it must have been deposited from the same waters with the very similar clay which I found at Tertse, in lower Nubra, in October, 1847, and that it is therefore lacustrine. If this be admitted, it seems impossible to escape from the conclusion, that the deposits in the Kardong valley, (of which I have given an imaginary section in page 398,) though different in appearance, belong to the same lake. Now, these attain an elevation of 13,500 feet and upwards, as they commence above Kardong: the level of the surface of the Nubra lake can therefore hardly have been less than 14,000 feet; so that it must have extended up the Tanktse valley, almost as far as the low pass by which that district is separated from the Pangong lake.

From Karsar, I marched on the 25th of July, down the valley of the Shayuk, to Diskit. The earlier part of the road, after ascending abruptly out of the Karsar ravine, lay over the clay platform, which was perfectly flat; but after about four miles, it descended nearly to the level of the river, whose wide gravelly plain now extended on the south side to the very foot of the mountains, the lacustrine beds having been entirely removed. The plain was traversed by several small streamlets, apparently derived in a great measure from the river, the water of which seemed to sink among the gravel and sand of its bed, and to spring up again at a distance from the main channel. One of these streams ran at the extreme edge of the plain, close under the cliffs, which here rose almost precipitously to a great height. Its banks were very saline, and in the neighbourhood of Diskit a great part of the plain was encrusted with soda.

The cultivated lands of the village, which is of considerable size, lie on a sloping bank, rising rather steeply out of the plain. Many apricot-trees grow among the houses, some of which were large enough to afford a shade under which a tent could be pitched. The vegetation was in general the same as at Karsar, but a white-flowered Allium was new, as well as a species of Chloris, which was abundant in the pastures. A very small Cyperus, which grew in the water-courses, appeared to be a dwarf state of a species common in the plains of India, and, with the Chloris, which is a tropical grass, was interesting as an indication of the considerable heat of the summer climate in the valley of the Shayuk, notwithstanding its great elevation.

The village of Diskit is almost exactly opposite the place where the Nubra river joins the Shayuk from the northward. In October, 1847, I had crossed the Shayuk five or six miles above Karsar, and descended along its right bank, but during the hot months this route is not practicable, as there are no bridges, and the river is too deep to be forded anywhere except just at its junction with that of Nubra, where the wide gravelly plain of the Shayuk expands to its greatest diameter, and the river is divided into numerous branches.

The greater part of the 26th of July was occupied by the passage of the Shayuk, which was both tedious and difficult, the river being now nearly at its greatest height. The first branch was nearly two miles from Diskit, the intervening gravelly plain being partly swampy, with a few bushes of Hippophaë, Tamarix, and Myricaria. There were four large branches to be crossed, besides several of smaller size. Nearly a mile of sand separated the last large branch from the remainder, and the ford was a most intricate one, each branch being crossed obliquely and at a different point from the adjacent ones. The united breadth of all the streams could not, I think, have been less than half a mile. The velocity of the water was so great, that though the depth nowhere, I think, exceeded three and a half feet, and was more usually about two and a half, people on foot appeared to have the utmost difficulty in retaining their footing, and the loaded men had to be supported by one or two without loads on each side. In the more difficult parts, two men placed themselves on each side of my horse's head, to guide him in the proper road, and two more at each stirrup to give him support in case of need. When in the centre of the current, where, from the necessity of keeping my eye on the horse's motions, I had to look at the water, I found it impossible to avoid a feeling of giddiness, and an impression that horse and rider were being hurried upwards with extreme velocity in a direction contrary to the stream. These very rapid portions, however, were never more than ten or twenty yards broad; the remainder was more moderate and shallower.

After safely effecting the passage with all my party and baggage, I proceeded about a mile over loose sand, and encamped at the village of Lyakjung, situated at the border of the low plain of the river, at the point of union of the two valleys. The Shayuk valley is visible from this place as far as the large village of Hundar, about ten miles, the river running throughout that distance through a wide gravelly plain, but with high rocky mountains on both sides.