From Kulja, across the Tian Shan to Lob-Nor/Chapter 1

by Nicholas Prejevalski

 

TRAVELS TO LOB-NOR.

 

 

CHAPTER I.

 

Departure from Kulja — Valley of the Ili — Crossing the Tekes — Inhahitants — Fertility of Kunges Valley — Abundance of fruit; bears, birds, &c. — Pass to the Tsanma — Fir forests — Autumn in the mountains — The Narat range — Yulduz and its fauna — Hunting — Ovis Poll — Descent of Tian Shan — Yakub Beg's envoys — River Kaidu-gol — Arrival at Korla — Jealousy and distrust of officials — Desert of Lob— Hydrography of Lower Tarim — Barren country — Oleasters — Monotonous scenery.

Another successful step in the exploration of Inner Asia—the basin of Lob-nor, so long and so obstinate a terra incognita—has at length been revealed to science.

As originally contemplated, the starting-point of my expedition was the town of Kulja.^[1] Here I arrived at the end of July, 1876, with my two companions, Lieut. Povalo-Schweikofsky, and a volunteer of the name of Eklon. Adequately supplied this time with funds, I was able to buy in St. Petersburg and Moscow all the requisite stores for so long a journey, and these, together with guns and ammunition (the latter supplied by Government), weighed about two tons. This weight I had to transport from Perm to Kulja on five postal troikas which took me more than a month, delayed by the abominable state of the roads in crossing the Ural.

At Semipalatinsk we were joined by the companions of my last expedition to Mongolia—the Trans-Baikalian Cossacks, Chebayeff and Irinchinoff, who declared their readiness to share with me once again the hardships and privations of a new journey. One other Cossack was also sent from Trans-Baikalia to act as interpreter of the Mongol language, and I took three others at Vemoye from the Semiretchinsk force. Lastly, at Kulja itself I hired a Kirghiz Christian convert, who spoke the Sart language. In this way the personnel of my expedition was formed, but unfortunately I was not nearly so successful in the choice of my companions as I had been on the last occasion.

Nearly three weeks were occupied at Kulja in the final formation and equipment of our caravan, consisting of twenty-four camels and four riding-horses. The latter were bestridden by myself, my companions, and one of the Cossacks. We were all admirably armed; besides fowling-pieces each carried a Berdan rifle swung over the shoulder, and a brace of pistols in our holsters.

Our original plan was to proceed to Lob-nor, explore as much of this lake and its environs as possible, and then return to Kulja, leave our collections here, and taking our remaining supplies, start for Tibet.

On the morning of the 12th of August we took our departure from Kulja, accompanied by the good wishes of our countrymen resident at that town.

Our road lay at first up and almost alongside the bank of the Ili, whose valley is here thickly settled by Taranchis.^[2] Clean, pretty villages with gardens, shaded by lofty silver poplars, follow each other in quick succession. In the intervals are corn-fields irrigated by numerous watercourses, whilst on the meadows along the river's bank large herds of sheep, oxen, and horses are grazing.^[3] The population is everywhere apparently prosperous; the Mohammedan rising never having desolated this part of the valley. The districts which were laid waste lie below Kulja, following the Ili. Here, too, agriculture once flourished, but since the extermination of the Chinese inhabitants by the Taranchis and Dungans, the villages are mostly destroyed, and even such towns as Old Kulja, Bayandai, Chimpanzi, and others are in ruins, the fields deserted and choked with weeds. After crossing to the left bank of the Ili, near the mouth of the Kash (fifty versts beyond Kulja), we continued as before to ascend its valley, in this part twenty versts wide, and having the appearance of a steppe plain with a clayey and slightly saline soil, producing Ceratocarpus, dwarf wormwood, and Lasiogrostis; in the more fertile part astragalus, a few kinds of herbs or plants of the order compositæ and small gnarled bushes; whilst the river bank is fringed with thick cane-brake.

The width of the Ili near the mouth of the Kash is about 500 feet, with a very rapid stream. Taranchi villages continue for twelve versts further up the right bank from the confluence of the Kash—the left bank has no settled population. Here only occasional fields temporarily tilled by the Kalmuks may be seen, and these only nearer the river Tekes. The last-named stream flows from the Mussart, and unites with the Kunges to form the Ili, which empties its muddy waters into Lake Balkash. [See supplementary note.]

The Tekes, here 350 feet wide, with a terribly swift stream, is crossed in small, rotten ferry-boats. On these our baggage was taken across; the horses and camels were fastened behind the boat, and were made to swim to the opposite bank. This swim proved very injurious to the camels, and three soon afterwards died from the effects of it. Beyond the Tekes our road lay continually in the same easterly direction by the valley of the Lower Kunges,^[4] which is hardly distinguishable from that of the Upper Ili, excepting that feather-grass is more abundant. The hills, bordering the valley as before, are covered with grass, rounded in outline, and totally bare of trees as far as the river Tsanma. Here the traveller sees the last of the fields and encampments of the Turgutes;^[5] beyond  his point, as far as the Kara-shahr valley, no inhabitants are to be met with. The flora of the plain we had hitherto traversed from Kulja was very scanty; and the fauna equally deficient. The season too (latter half of August) was most unfavourable for ornithological researches and preparing skins, many of the birds being in the moulting stage. But snakes and lizards were very abundant, and we collected a good number of these reptiles. Of fish we only caught four kinds; Dyptichus, Schizothorax, perch, and gudgeon. According to the Cossacks, who are great fishermen, there are no others in the Ili.

As the elevation of the country rises beyond the Tsanma,^[6] the valley of the Kunges changes its character, and becomes narrower and more fertile. Instead of the clumps of vegetation we had hitherto seen, excellent and varied herbage clothed the undulating plain, growing higher and thicker every ten versts or so as we advanced; the outline of the marginal hills became sterner, and spruce firs began to show themselves, their lower belt marking the limit of the summer rains.

Rain, however, does fall, although perhaps less abundantly, in the steppe zone, where the elevation is 4000 feet, or even somewhat less. At this point larch woods begin growing on the banks of the Kunges itself, interspersed with tall poplars (some 80ft. high, with stems 3ft. and 5ft. thick) and apple-trees; birch and apricot are more rare. The thick underwood is composed of hawthorn, cherry, woodbine, guelder rose, and briar. The islands in the river are thickly overgrown with tall salix or willow, round whose stems the wild hop is often twined, and tamarisk appears on the sandy and stony spots. The woodland meadows and slopes of the neighbouring hills are everywhere clothed with the thickest grass, interwoven with convolvulus and dodder, often 7ft. high and almost impassable in summer. But at the season we arrived on the Kunges (early in September) the grass was withering and dying down, and the trees and bushes had donned their autumnal attire.

After the monotony of steppe scenery, the wooded islands and banks of the Kunges produced an agreeable impression, and yielding to its influence we determined on making some stay in this highly-favoured little corner of the Tian Shan. Here, too, we could reckon on a rich, scientific harvest. Moreover, two of our Cossacks had proved unserviceable for travel, and we were obliged to send them back to Kulja and exchange them for two soldiers, whose arrival could not be expected for ten days.^[7]

We selected for our camping-ground in the forests of Kunges the very spot occupied for some months in 1874 by one of our sotnias of Cossacks. Here the shed they had erected, their kitchen and bath-house were still standing; we too enjoyed a good and final wash in this bath-house before starting for the Tian Shan.

One remarkable characteristic of the Kunges forests, and probably of other wooded glens on the northern slope of the Tian Shan is the great abundance of apple and apricot-trees^[8] producing excellent fruit. The apricots, or as they are here called, uriuk, ripen in July; the apples by the end of August. The latter are about the size of a hen's egg, pale yellow in colour, and with an agreeable bitter-sweet flavour. We were just in time for the apple harvest on the Kunges; the trees were laden with the fruit, quantities of which strewed the ground, where they decay without benefiting any one, or are devoured by wild boar, bears, deer, and goats, which at this season of the year descend in numbers from the adjacent hills. Wild boar and bears are particularly addicted to apples, and the latter are known to indulge to excess in their favourite dainty.

Our sport with the larger game was tolerably successful, and we secured some fine specimens for our collection; amongst these an old dark-brown bear of a species peculiar to the Tian Shan, and distinguishable from the common bruin by the long white claws on the fore-feet—a peculiarity which induced Sévertseff to name it Ursus leuconyx.^[9]

Besides four-footed beasts, the forests on the Kunges contained many a migratory woodcock and thrush (Turdus atrigularis, T. viscivorus), and numbers of corncrakes and landrail on the meadow-land. Many of the nesting birds had departed for the south; of non-migratory we only found an occasional pheasant (Ph. mongolicus), blue tits (Cyanistes cyanus), woodpeckers, and a few others. The autumnal flight is generally very deficient in this part of the Tian Shan, even in small birds.

A range of no great elevation, crossed by a pass 6000 feet high, separates the Kunges from the broad valley of the Tsanma, the river we had already crossed near its mouth. Although not more than eight versts apart, the difference in the height of the respective valleys of the Kunges and Tsanma is nearly 2000 feet. From the pass itself may be seen, as from an opera-box, on one side the comparatively low and deeply indented Kunges valley, on the other the elevated basin of the river Tsanma. The latter is about four versts wide, and thickly clothed with high grass. Along the upper course of the river, commencing at an elevation of 6000 feet, are forests, whose prevailing trees are the Tian Shan spruce^[10] (Picea Schrenkiana), the mountain ash now taking the place of the apple or apricot-trees. Spruce firs are also scattered in clumps over the neighbouring mountains, growing as high as 8000 feet and even upwards above the sea level.

The approach of autumn now began to be felt in the mountains. Not very long ago we had been oppressed by the heat on the Ili plain; now, on the contrary, every morning brought light frosts, snow lay on all the higher mountains, the trees and bushes were shorn of half their foliage. But the weather continued bright and clear, and during the day it would even be hot at times.

After having ascended the Kunges, and afterwards the Tsanma to its source, we moved towards the foot of the Narat range, which, with its western prolongations,^[11] forms the northern buttress of an extensive and lofty plateau situated in the very heart of the Tian Shan, and known by the name of Yulduz.

Before describing Yulduz, let us say a few words about the Narat. This range though nowhere reaching the limit of the perpetual snow-line, presents nevertheless a wild and alpine character. Its solitary peaks with their steep slopes, particularly near the axis of the range, are scored with bare precipitous cliffs forming narrow gloomy chasms. Below these again are the alpine meadows, and lower still on the northern side clumps of spruce fir; the southern slopes of the Narat are treeless.

We crossed this range at its eastern extremity, where the ascent is not particularly steep, though difficult for camels; on the Yulduz side the descent is very gradual. Snow lay in small quantities on the northern slopes during our march, i.e. in the middle of September, whereas on its south side the Narat was completely free of snow. The pass is 9800 feet above sea-level. Near the summit we killed a small boar, preserving its skin for our collection, and its meat for our provision-store.

Descending the Narat, we entered Yulduz. This name signifies "star," and was perhaps bestowed on the country owing to its elevated position among the mountains, or from the circumstance of its being the promised land of cattle.^[12] The pasturage is excellent in every part, and it enjoys in summer an immunity from flies and mosquitoes, "an admirable, cool, and productive country, fit for gentlemen and cattle to inhabit," as the Torgutes described it to us. It forms an extensive depression continuing for some hundreds of versts from east to west. In all probability it was at some remote geological epoch the bed of an inland sea, as its alluvial clay soil tends to prove. Yulduz consists of two parts: Greater Yulduz occupying the more extensive westerly half of the whole depression, and Lesser Yulduz the smaller eastern part. Both of these have the same general features, the difference between them consisting only in their size. Lesser Yulduz, along the whole of which we passed, has the appearance of a steppe-plain extending lengthways for 135 versts, and widening in the centre to thirty versts.

Near the marginal mountains this plain is hillocky, and covered with luxuriant herbage. Here, too, chiefly in its eastern part grow low, stunted bushes of camel thorn, willow, and Potentilla; of trees there are none in Yulduz.

The elevation of Lesser Yulduz is from 7000 to 8000 feet above sea-level.^[13] The marginal ranges on the north and south are wild, rocky, and of

great elevation, not only above the level of the sea, but also above that of the adjacent plain; the southern range, dividing Lesser from Greater Yulduz, rises in several places above the limit of perpetual snow.^[14] Exactly in the centre of Lesser Yulduz, and throughout its entire length, flows the Baga Yulduz-gol, uniting with the Kaidu-gol after the latter has drained Greater Yulduz, and finally emptying into Lake Bagarash.

We forded the Baga Yulduz-gol, but in spring and summer the water is too high to allow of the fords being practicable. Fish are plentiful, both in the Baga Yulduz-gol as well as in its tributaries, but only of two kinds:^[15] Dyptichus, a foot or a little over in length, and gudgeon. About half-way down this river, and for some distance on either side, are marshes (sasi) and lakelets. Here we found in the latter half of September numbers of migrating water-fowl;^[16] most of the other birds nesting in this country had taken wing for the south, and it was only now and then that we saw a few in the mountains.^[17] Non-migratory birds^[18] however are common.

Yulduz is very rich in mammalia; of the larger animals there are brown and tawny bears, Ovis Poli, wild goat, and what are more remarkable, considering the absence of trees, deer and pygargs; numbers of marmot hybernate as early as the middle of September, when they frequently become the prey of the bear, who grubs up their burrows, and extracts from them the half-dormant little animals. Wolves are very common, and foxes particularly so; the latter prey on the innumerable field-mice. Amongst others of the rodent order, Siberian marmot are plentiful, but they were also hybernating, and wild boar are occasionally found in the marshes of the Baga Yulduz-gol.

There are absolutely no inhabitants in either Yulduz, although not above eleven years ago Turgutes lived here to the number of ten thousand kebitkas. Plundered by the Dungans, these nomads retired, partly to Shikho, and partly to the Kaidu-gol to the neighbourhood of Kara-shahr; while some escaped to our lines on the Ili, where they are living at the present day.

Our entrance into Yulduz was marked by an unfortunate incident. My companion Lieut. Povalo-Schweikofsky, who from the very first was unable to support the hardships of travel, fell ill, and as he did not recover, was obliged to return to his former place of service. Fortunately my other travelling-companion, the volunteer Eklon, proved to be an energetic and willing youth, and with a little practice he soon became an invaluable assistant.

We stayed about three weeks in Yulduz, hunting most of the time, and succeeded in obtaining about a dozen fine skins for our collection, including two males of the Ovis Poli. This magnificent sheep, characteristic of and peculiar to the highlands of Central Asia, is often seen here in herds of thirty to forty.

These herds are mostly composed of females, and a few young, full-grown males, acting as leaders and protectors. The old males^[19] hold aloof, and generally roam about singly, or in twos and threes. The favourite resort of these sheep are the spurs of the great ranges, and the smooth slopes leading to the level steppe. They rarely assemble in stern, rocky mountains, where the wild goat^[20] makes his home, and where the latter may also be seen in herds numbering forty and upwards, similar in habits to the arkari, and extremely difficult of approach, both on account of his wariness, as well as from the nature of the localities he frequents.

The deer we saw in Yulduz belong to the same kind as those inhabiting the forests of the Tian Shan. The stags are of enormous size; the does are smaller, but fully equal to the full-grown male of the European deer (Cervus elaphus).^[21] Owing to the absence of forests in Yulduz, the deer frequent the belts of low bushes, climbing the rocks as easily as the mountain sheep, and so like these as to be mistaken for them at a distance. In spring, during the months of May and June, they are eagerly pursued by hunters for the sake of their young horns—so called ‘panti’ which fetch high prices in China. Thus, in Kulja, a pair of large, six-pointed antlers is worth fifty to seventy roubles, in first hands; and even small ones fetch fifteen, twenty, or thirty roubles. The profits derived from this chase induce Russian and native hunters to pursue it with axdour during the spring, throughout the vast expanse of Asia, from Turkestan to the sea of Japan.^[22]

After we had done hunting we turned into the Kaidu valley, crossing the southern slope of the Tian Shan. The ascent of the pass from the Yulduz side is so gradual as to be hardly perceptible, although the elevation above sea-level is at least 9300 feet. But the descent on the other side is extremely difficult. For about forty versts the barely-distinguishable track follows the defile of the Habtsagai, and for twenty-two versts further that of the Balgantai river. Both these ravines are exceedingly narrow (in places not more than 400 feet wide), their beds strewn with débris of rock and pebbles, and their sides walled by huge precipitous cliffs.

The banks of the streams are thickly covered with willow and tamarisk bushes; lower down, at an elevation of about 6000 feet, buckthorn and elms appear; and, still lower, barberry and oleaster; the only grasses found in the ravines are lasiogrostis and reeds. The surrounding mountains are entirely bereft of vegetation, the neighbouring desert having affixed the seal of death on this side of the Tian Shan. Atmospheric precipitations, although plentiful on the northern side of the range where the rain-clouds deposit their moisture, the last drops of which are wrung out by the snow mountains of cold Yulduz, are absent here, and it is exceedingly probable that the whole southern slope of the Eastern Tian Shan is arid and barren.

Upon entering the Kaidu valley we descended to 3400 feet above sea level. The weather became warm, and the morning frosts no longer severe; whilst in Yulduz, towards the end of September, the thermometer marked 10° Fahr. at sunrise, and snow fell occasionally.

At the camping-ground of Kara-moto, where we halted, we were well received by the first Turgute inhabitants that we met. Meanwhile, the report spread rapidly of the approach of the Russians, and alarmed the whole Mohammedan population of the neighbourhood. It was stated that Russian troops were marching into this country, and that their advance-guard had already appeared on the Kaidu. This story gained currency when, on our first arrival, the reports of our fire-arms, as we shot pheasants and other birds, began to be heard, and caused such a panic among the Mohammedans living near Kara-moto as to induce them to leave their homes and fly to Kara-shahr.

Thither notice was of course at once sent of our arrival, but, at first, none of the officials made their appearance. We now sent back to Kulja our guide Tokhta-akhoond, a man devoted to us, but for this very reason hateful to the Mohammedans, he being himself a follower of the prophet born at Korla, whence he had escaped some years previously to Ili. With him we despatched the greater part of our collections, so as not to encumber ourselves needlessly with them.

On the third day of our appearance at Kara-moto, six Mohammedan envoys from the governor of Korla^[23] came to inquire the object of our journey. I explained to them that we were on our way to Lob-nor, and that Yakub Beg was well aware of this.^[24] On receiving my reply the envoys returned to Korla, but a small picket was stationed on the opposite bank of the Kaidu, to watch our movements. The day afterwards the same envoys reappeared, reporting that the governor had despatched a courier to Yakub Beg,^[25] and that until his answer were received, no permission could be given us to proceed. This decision did not disturb us in the least, as the wooded country on the Kaidu abounded in wintering birds and pheasants. The latter probably belong to a new species, very closely allied to Phasianus Shavii, recently discovered in the neighbourhood of Kashgar by the British mission to Eastern Turkestan, and occurring along the whole length of the Tarim, and on Lake Lob.

The Kaidu river is from 200 to 270 feet wide at Kara-moto, with a very rapid stream and a depth of three to four feet at the fords, which, during summer, are entirely impassable. Fish are plentiful in the river, but I cannot say of what kinds, for neither in going nor coming had we the opportunity of catching any. Fish too are said to abound in Lake Bagarash, into which the Kaidu empties. This lake lies not far to the west of Kara-shahr, and is very large and deep.^[26] It would have been very interesting to have explored it; but, alas, we could not manage this, either in going or returning.

After a halt of seven days at Kara-moto, we at length received permission to proceed to the town of Korla (but not to Kara-shahr), through which lies the road to Lob-nor. The distance from Kara-moto to Korla is sixty-two versts, and we accomplished it in three days, escorted by the same men who had a little while back first visited us. At each station on the road they brought us a sheep and some fruit. Before Korla could be reached, it was necessary to cross the last spur of the Tian Shan by a defile, through which rushes the Koncheh-daria, flowing out of Bagarash into the Tarim. At either end of this defile, which is ten versts long and very narrow, stands a mud fort, garrisoned by a small force.

No sooner had we arrived at Korla, and established ourselves in a house prepared for us outside the town, than a guard was placed over us, on the plea of protecting us; but, in reality, to prevent any of the townspeople, who are extremely dissatisfied with Yakub Beg's rule, from cating with us; and, in the same way, they forbad our entering the town, for they said, "You are our honoured guests, and must not be troubled with anything; all you want, we will bring you." But these honeyed words were mere phrases; they certainly brought us a sheep, bread, and fruit daily, but here their hospitality ended. All that could interest us, or advance the objects of our journey, was denied us; and we were not allowed to know anything beyond the gate of our enclosure. To all our questions as to the town of Korla, the number of its inhabitants, their trade, the features of the surrounding country, &c., we received the curtest replies, or absolute falsehoods; and this continued during the whole of our six months' stay in the dominions of Yakub Beg, or, "Badaulat," i. e. the happy one, as he is termed by his subjects. Nor was it until afterwards on the Tarim and Lob-nor, that we succeeded in occasionally ehciting some information in a quiet way from the inhabitants, who, though generally well disposed, feared showing their feelings. From the people on the Tarim, we learned that Korla and its neighbouring district numbers about six thousand inhabitants of both sexes. The town itself consists of two parts, each surrounded with mud walls: the old commercial town, and the new fort occupied only by troops, of whom very few were left at the time of our visit, most of them having departed for Toksum, where Yakub Beg was superintending the erection of fortifications, to protect himself against the Chinese.

The day following our arrival at Korla one of Badaulat's personal suite, a certain Zaman Beg, formerly a Russian subject, born at Nukha^[27] in Trans-Caucasia, and probably of Armenian extraction, paid us a visit. Having been actually at one time in the Russian service he spoke Russian fluently, and at once informed us that he had been sent by Badaulat to accompany us to Lob-nor—a piece of news that disconcerted us not a little, for I well knew that he was sent as a spy on our movements, and that his presence would be rather embarrassing than otherwise. Zaman Beg was, however, personally disposed to be friendly, and showed us all the attentions he could, for which I cannot be too grateful. Indeed, we got on better with him at Lob-nor than with any other of Yakub Beg's officials.

We left Korla for Lob-nor on the 4th of November. Besides the members of our own party the caravan included Zaman Beg, his servants, and a hadji. Hardly had we started than our companions showed us how disagreeable they could make themselves. In order to prevent us seeing the town they led us by a circuitous path across the fields and were bare-faced enough to assure us that there was no better road. However, there was no help for it but to feign ignorance, as we also did on many subsequent occasions, however distasteful such a line of conduct was to persons like ourselves engaged in scientific inquiries of the highest importance. They suspected and deceived us at every step; the inhabitants were forbidden to hold any intercourse with or even to speak to us. We were in fact under surveillance, and our escort nothing but spies. Zaman Beg evidently felt the irksomeness of the situation at times, but he could not alter his demeanour towards us. Eventually at Lob-nor, when they became tired of watching us, their former distrust wore off a little, but at first the police inspection was of the strictest, and not a week passed but that a courier arrived either from Badaulat or the Tokhsabai "to inquire after our well-being," as Zaman Beg naively expressed it.

Everything tended to show that our journey to Lob-nor did not please Yakub Beg, though he could not refuse General Kauffmann, and a quarrel with Russia on the eve of a war with China would have been impolitic on his part.

Probably with the view of inducing us to renounce our further journey, they led us to the Tarim by the most difficult road, obliging us to cross two large and deep streams—the Koncheh and Inchikeh-daria—by swimming. Reference to the map will show how easily we might have kept along the right bank of the former without having to cross it twice unnecessarily. We could only suppose that they wished to exaggerate the difficulties of the route by obliging us to swim in frosty weather with the thermometer at 4° Fahr. at sunrise. The crossing of both these streams was satisfactorily accomplished, though the camels suffered seriously from their cold-water bath, and when our guides convinced themselves of the hopelessness of their attempts to thwart us, they set to work and constructed rafts and landing-stages at the crossings.

Before reaching Lake Lob we had to march due south and strike the valley of the Tarim at a point eighty-six versts distant from Korla. For some way the country has the appearance of an undulating plain covered with a pebbly or gravelly soil, and totally devoid of vegetation, forming a belt twenty to twenty-five versts wide, more or less, running parallel to and at the foot of the Kurugh-tagh, a low, waterless, and barren range forming the last arm of the Tian Shan in the direction of the Lob-nor desert. This range, as we are told, rises on the southern shore of Lake Bagarash, and after continuing for nearly two hundred versts to the east of Korla merges in the low clay or sand hillocks of the desert.

Beyond the stony margin lying next to the mountains, and as it appears to me distinctly defining the shore-line of an ancient sea, lies the boundless expanse of the Tarim and Lob-nor deserts. Here the soil is loose saline loam or drift-sand remarkable for the absence of organic life. The Lob-nor desert is indeed the wildest and most barren of all the deserts I have seen, surpassing in this respect even that of Ala-shan.^[28] But before proceeding to a more detailed description of these places, I will briefly sketch the hydrography of the Lower Tarim.

As already stated, on our road from Korla to the south we had to cross two streams of considerable size—the Koncheh-daria^[29] and Inchikeh-daria. The first of these flows out of Lake Bagarash, forces its way through the last spur of the Tian Shan near Korla, and after taking a slight bend to the south, flows in a south-easterly direction, and falls into the Kiok-ala-daria, an arm of the Tarim. Owing to the velocity of their current and the loose clay soil through which they pass, the Koncheh-daria as well as the Tarim and all its arms and tributaries have worn for themselves deep trough-like channels. The width of the Koncheh-daria where we crossed it for the second time is fifty to seventy feet; depth ten to fourteen, and even more in places. Less than ten versts to the south of the Koncheh-daria, the Inchikeh-daria lay across our road; the latter river after a short course to the east loses itself in salt-marshes, perhaps uniting with the Koncheh at high water. After many inquiries we ascertained the Inchikeh to be an arm of the Ugen-daria, which falls into the Tarim close by, after rising in the Muzart and flowing past the towns of Bai and Sairam. In the meridian of the town of Bugur an arm separates from the Ugen-daria, uniting with the Tarim on the right, and a little fiirther down the Inchikeh-daria branches off to the left.

We struck the Tarim at the point where it is joined by the Ugen-daria with a stream 56 to 70 feet wide. The Tarim itself is here a considerable river from 350 to 400 feet wide, with a depth of not less than twenty feet. Its water is clear and stream very rapid. The river flows in one channel, and at this point reaches its furthest northing; hence it continues in a south-easterly course, then almost due south and before finally emptying into Lob-nor debouches in Lake Kara-buran. The natives rarely make use of the name Tarim in speaking of this river, which is more generally known as the Yarkand-Tarim or Yarkand-daria, after its principal feeder the river of Yarkand. The name Tarim, as we were told, is derived from "tara," i. e. field, owing to the circumstance of the water of this river in its upper course being mostly utilized to irrigate the fields.

Fifty versts below the mouth of the Ugen-daria a large arm, the Kiok-ala-daria (about 150 feet wide) separates from the Tarim and flows in an independent channel for about 130 versts before reuniting with the parent river. Into this arm flows the Koncheh-daria from the north.

With the exception of the Kiok-ala-daria the Tarim has no important subsidiary channels in its lower course, and is mostly contained in one channel. Along its banks to the right and left of its course are scattered marshes and lakes. These are for the most part artificially formed by the natives for purposes of fishing and pasturage—reeds being the only food for cattle in this wretched country. The river itself assists in the irrigation of its own valley. Fine sand and dust driven by the wind-storms prevalent in spring are caught and retained by the trees, bushes, and cane-brake growing on the banks, so as gradually to raise their level above that of the adjacent land, which is constantly diminishing under the influence of the same causes. Hence it becomes only necessary to bore through the bank for the water to pour out of the river and inundate a more or less extensive tract of plain. With the water come fish, and in a little while reeds begin to grow. After a time the channel gets silted up, the lake grows shallower, the fish are easily taken, and the recently submerged land affords pasturage for sheep. When the reeds are all fed off, the operation is repeated, and a fresh supply of fish and pasturage obtained.

The general character of the Lower Tarim is very much as we have described it. Along the right bank and not far from the river lie bare hillocks of drift-sand twenty to sixty feet high. These sandy wastes continue the whole way down the Tarim to its confluence with Lake Kara-buran, then up the Cherchen-daria in a south-westerly direction, almost as far as the town of Keria, and a long way up the Tarim from the mouth of the Ugen-daria. Indeed the whole country between the right bank of the Tarim on the one side to the oases at the foot of the Kuen-lun on the other is described to be filled with drift-sand and positively uninhabitable.

On the left bank of the Tarim the sands are much less frequent and not nearly so extensive. Here the soil consists of loose saline clay in some places entirely bare, in others again overgrown with rare bushes of tamarisk and occasionally patches of Haloxylon. These plants bind the yielding soil with their roots, the intervals being subjected to the full force of the wind, which accumulates the drift round the bushes so as gradually to form a hillock seven to fourteen feet high beneath each of them; and such hillocks cover vast areas, as they do in Ordos and Alashan.

On the banks of the Tarim itself, as well as on its arms and tributaries, vegetation is somewhat more varied, though scanty in the extreme. First of all, in the narrow wooded belt we notice the poplar (Populus diversifolia) a crooked tree attaining a height of between twenty-five to thirty-five feet, with an almost invariably hollow trunk from one to three feet thick; the oleaster in small quantities; the Halimodendron, Asclepias, and two other kinds of bushes of the bean family, covering vast areas, whilst tall cane-brake and Typha obstruct the lakes and marshes on both banks of the Tarim, and as a rarity, wild pea and Astragalus, with two or three representatives of the genus Compositæ growing here and there on the damper ground. These complete the list of plants of the Tarim and Lob-nor.^[30] No meadows, no grass, not a vestige of a flower is here to be seen.

It would indeed be difficult to picture to oneself a more desolate landscape; the poplar woods, with their bare soil, covered only in autumn with fallen leaves parched and shrivelled with the dry heat, withered branches and prostrate trees encumbering the ground, cane-brake crackling under foot, and saline dust ready to envelope you from every bough that you brush aside from your path.^[31] Now, again, you come to acres of dead poplars, with broken boughs, shorn of their bark, lifeless trunks never decaying, but crumbling away by degrees, to be hidden in layers of sand.

But cheerless as these woods are, the neighbouring desert is even more dreary. Nothing can exceed the monotony of the scenery. Whichever way you turn, an ill-favoured plain meets your eye, covered with what seem to be large mounds, but which are really hillocks of clay surmounted by tamarisk, between which the path winds, every surrounding object shut out from sight, and even the distant hills barely visible in blue outline through the dusty vapour which fills the atmosphere like fog. Not a bird, not an animal, nothing but the occasional tracks of the timid gazelle.

1. [There were two towns of this name, about twenty-five miles apart. The one mentioned in the text is the old Tartar town, now the head-quarters of the Russian administration of the province of Ili; the other, New or Manchu Kulja, was a flourishing Chinese city of about 75,000 inhabitants until the late Mohammedan rising, when it was taken by the rebels, the whole population put to the sword, and the city reduced to ashes. See Schuyler's Turkistan, ii. 162 et seqq.—M.]
2. [These were agricultural colonists from Eastern Turkistan, of whom Sir D. Forsyth has spoken in his introductory remarks. According to M. Radloff, quoted by Schuyler, their language is more specifically Turkish than that of any book published at Constantinople.—Turkistan, ii. 169 seq.—M.]
3. [Schuyler, who visited the Ili valley in 1873, thought it the richest part of Russia's recent acquisitions in Asia.—ii. 198.—M.]
4. [Colonel Yule informs me that the route followed by Colonel Prejevalsky seems to be the same as that of Shah Rukh's embassy to China in 1420, which went by the Kunges and Yulduz to Turfan.—M.]
5. [The Turgutes or Torgutes, as already mentioned (vide supra, introductory remarks), are the Kalmuks of the present day, of whom remnants still exist on the Lower Volga.—See Wallace's Russia, ii. 52; and see also pp. 169—186 of this work.—M.]
6. Kulja is about 2000 feet above sea-level. It should be noticed that although the heights have all been measured barometrically, the results obtained have as yet only been worked out approximately.
7. Our Kirghiz interpreter also proved worthless, and he had also to return to Kulja and be replaced by a new one.
8. [Compare Aristof 's description of the Kunges valley, quoted by Schuyler.—Turkistan, ii. 199 seqq.—M.]
9. Sévertseff identifies his Ursus leuconyx with U. isabellinus Horsf. from the Himalayas. But in my opinion they are two distinct species. The Himalayan bear is also met with in the Tian Shan, where it is only known to inhabit the elevated plateaux devoid of trees and the Alpine region, never entering the forest zone. Besides, U. isabellinus is of a tawny colour; U. leuconyx, on the other hand, is dark brown, like the European U. arctos.
10. This tree attains a height of "seventy to eighty feet, with a thickness of stem two, three, and often four feet in diameter. It grows very much in the sugar-loaf shape, its thick branches hardly projecting from the general mass, so that the whole tree has the appearance of having been cropped by a barber.
11. The western prolongations of the Narat range, taking them in their order, are the Dagat, Kara-nor, Koko-stmg, and Djamba-daban ranges; the three last-named are said to be capped with eternal snows.
12. [According to Bellew, Yulduz was the son of Manglai, the son of Timurtash—"Ironstone"—a descendant of Kaian. He raised the Mongol name to the highest fame, and was the ancestor of all the Mongol Khans. (Report of a mission to Yarkand in 1873, p. 136.) May not the country have derived its name from him, for it was all under Mongol dominion?—M.].
13. The lowest parts are on the lower course of the Baga Yulduz-gol; on its upper stream and nearer the marginal mountains the country is higher.
14. This range, as well as the northern, has no general name among the inhabitants, who distinguish parts by specific names.
15. At all events, we did not catch any other kind of fish, either in autumn or in spring, on our return journey.
16. Common wild duck, gadwall, teal, red-crested pochard, red-headed pochard, and garrot.
17. Redstart, accentor, mountain finch, and Brandt's finch (Leucosticte Brandtii) the two last-mentioned generally in flocks.
18. Snow vulture, black vulture, wall creeper, rock partridge; and shore-lark (Otocoris albigula) on the steppes.
19. The horns of these old males are of colossal proportions. Those in my collection measure 4 feet 8 inches in length, taking the outside of the curve, and are 1½ feet thick at the base, their weight is about 36 lbs.
20. In all probability this is Capra Skyn, not Capra Sibirica the horns approaching at the tips and turning inwards; the colour of the hair is a tawny-grey, belly white. The longest horns I saw measured 4 feet.
21. A two-year old buck killed by me on Yulduz, measured 6 feet 1 inch in length, 4 feet 3 inches in height at the shoulders. A full-grown doe, killed in the same place, measured 7 feet 4 inches in length from the nose to the tail, and stood 4 feet 3 inches from the ground.
22. Compare Mongolia, i. 170.
23. This town is fifty versts south-east of Kara-shahr.
24. Before our departure from Kulja, Yakub Beg wrote in answer to the Governor-General of Turkestan that the Russians going to Lob-nor would be hospitably received in his dominions.
25. Who was then at Toksum, not far from Turfan.
26. According to the Kalmuks, it is eight or nine days' ride round Bagarash.
27. [Nukha rose to be a place of some note about the middle of the eighteenth century, when it was the capital and place of residence of the khans of Shekin, who are reported to have turned back Nadir Shah's victorious army. In 1805 Nukha was taken by the Russian general Nebolsin, and finally annexed by Russia in 1819.—M.]
28. For a description of Ala-shan see the author's last work, Mongolia, &c. vol. i. ch. vi.
29. Incorrectly marked on existing maps, both as to name and direction.
30. Moreover the poplar and elæagnus only grow along the Tarim, not on Lob-nor. [Henderson remarks that the latter is one of the most common trees in Yarkand, where it is cultivated as a tall hedge and for its fruit along roadsides. (Lahore to Yarkand, p. 335). The name is derived from ἐλαία, an olive, the tree having a striking resemblance to an olive-tree.—M.]
31. The poplars are so saturated with salt that on breaking a bough a saline incrustation may be seen on the wood.