His eye might there command wherever stood
City of old or modern fame, the seat
Of mightiest Empire, from the destined walls
Of Cambalu, seat of Cathaian Can,
And Samarchand by Oxus, Temur's throne,
To Paquin of Sinæan Kings; and thence
To Agra and Lahor of great Mogul…

—Paradise Lost, Bk. XI.

THE area over which Mirza Haidar's history extends is a wide one. Expressed in the geographical terms of our times, it may be said to deal with Western Turkistan, Bokhara, Farghána, the Russian province of Semirechensk (or the seven rivers), the Chinese province of Ili (or Zungaria), Eastern Turkistan, Tibet, Ladak, Baltistan, Gilgit and the neighbouring states, Chitral, Wákhán, Badakhshán, Afghanistan, Kashmir, and Northern India; while references are frequently made to countries lying even beyond these regions. But it is a history, especially, of the eastern branch of the Chaghatais—i.e., the Moghuls proper —and, therefore, the chief scene of action lies in and immediately around their home-land. The situation and extent of this region are not difficult to describe, but it is far from easy to give a name to it as a whole.

Its limits were not very clearly defined at any period, and were seldom the same for twenty years at a time, while even the names of “Jatah” and “Moghulistan,” by which a portion of it was known, are now not only obsolete, but have hitherto been subject to some doubt regarding the exact locality to which they were applied. Moreover, there was at no time any one name in use, which served to designate the entire Khanate. Mirza Haidar usually speaks of ‘Moghulistan’ and ‘Kashghar,’ but it is not always clear whether, by ‘Kashghar,’ he means only the city and district of that name, or the entire province of Alti-Shahr—the Six Cities* of Eastern Turkistan—a region which, he tells us more than once, was, when combined with Farghána, termed Mangalai Suyah or “Facing the Sun.”* This territory would almost exactly correspond to the provinces of Farghána and the Chinese Turkistan of modern times, less the districts of Karashahr, Turfán and Hami in the extreme east; or, in other words, to Farghána and Alti-Shahr. But even if we were to give the entire country the double name of “Moghulistan and Mangalai Suyah,” there would still remain some difficulties of definition. At first sight it would appear that the author describes the limits very exactly; but this is not quite the case, and for two reasons. In the first place, he sets forth the provinces that composed it on several occasions, but does not always make them the same: the other is that, in common with all Asiatics who attempt to describe an area, he names a district or a geographical feature as a boundary, but does not mention whether it should be included or excluded— whether the limiting district, range or lake lay beyond or within the area he is describing. In addition to these uncer­tainties there is also the inconsistency that Farghána, as a whole, was seldom included within the actual possessions of the Khans of Moghulistan. They always regarded it as theirs by right, but they rarely held more than a few positions, or districts, within its limits, and even these they were usually unable to keep for any length of time. Practically, therefore, Farghána can scarcely be held to have formed a part of their dominions, although it may have been comprised in the geo­graphical term “Mangalai Suyah.” With this reservation, however, and in order to show what the author describes, it would seem as well that Farghána should be included nominally with Moghulistan and Alti-Shahr; so that, after making due allowance for the fluctuations that occurred at different periods, the following may be regarded (as nearly as possible) as a statement of the extent of the dominions of the Moghul Khans, from about the middle of the fourteenth century to the middle of the sixteenth.

There was no central division, but the province of Moghulis-tan proper—or Jatah, as it was also called during the early part of that period—being a “steppe” or pastoral country, and the homeland of the dominant tribe, was therefore the principal division. Its western boundary marched with the province of Shásh, the modern Táshkand, which seems to have contained the whole of the lowlands of the valley of the Sir, from a little above Khojand to about the Arys tributary, and included such towns as Sháhrukhia, Táshkand and Sairám. Immediately to the east of this level agricultural stretch, rise the hills which separate it from the valley of the Upper Tálás, and it was this line of hills, or uplands, which seems to have stood usually, and in a general way, for the boundary of the Moghuls. To the north of Shásh lay the province of Turkistan, with the Káratau hills between it and the Lower Tálás, and here again the hills appear to have been the western limit of the nomad tribes. Turning towards the north-west, a line drawn from the Káratau to the southern extremity of Lake Balkásh, and continued again from its other extremity to the Tárbágatai mountains, may be taken roughly to have been the frontier in that direction. We hear, at any rate, of no transactions of the Moghuls, as a tribe, anywhere to the north-west of the Balkásh; nor do we trace them anywhere to the north of the Imil river, which is fed from the Tárbágatai mountains, except when flying before Timur's avenging army in 1389 and 1390, they crossed the range into the valley of the Irtish. But this was an occasion when danger led them to seek refuge beyond the bounds of their own country. From the Tárbágatai range, the limiting line would probably bend south-eastward to some point at the northern foot of the Tian Shan, near the present Urumtsi; but this is somewhat uncertain. All that is clear is that the tract now known as “Zungaria” (or the land of the Zungár, or Jungár, Kalmáks) formed a part of the Moghul dominion, but how far precisely, “Zungária” extended towards the east, there is nothing to show. Probably it included Lakes Ebi Nor and Ayar Nor, and had for its central feature the upper course of the Ili river. On the south, the main range of the Tian Shan, as far west as about the head of the Nárin river, divided Moghulistan from Kuchar, Aksu, etc., while westward, again, the water-parting ranges between the Nárin and Lake Issigh-Kul, continued up to the heads of the Tálás, would seem, approximately, to have been the line of separation from Kash-ghar and Farghána.

The boundaries of Alti-Shahr were better defined by natural features than Moghulistan. It may be said, generally, to have embraced the whole of the system of the Tarim, together with some of the upper waters of the Sir. On the north it marched with the southern limit of Moghulistan, as described above. On the east it included the town and district of Kuchar (which was usually a dependency of Aksu), and probably the region of Lake Lob; while it excluded Karashahr—then known as Chálish— and all to the east of it, which constituted, as we shall see, a province that bore the ancient name of ‘Uighuristán.’ On the south, along the whole length of the country, the mountains forming the scarp of the Tibetan highlands—the Kuen-lun and the Altyn Tágh—shut it off from all beyond. Towards the west the Pamirs, generally speaking, constituted its extreme limit, till these abut northward on the southern confines of the valley of the river Sir; for these uplands, then as now, seem to have divided the Kashghar district from Farghána. What the precise limits in the Pamir region were, there is nothing to indicate, but in speaking of Sárigh-Kul, Mirza Haidar implies that that district, at least, lay within the pro­vince of Alti-Shahr, and for a time also, we find Wakhán and Shighnan described as territory dependent on Kashghar, though this was not usually the case.

But if these were approximately and usually the limits, it does not follow that they were, on the one hand, never over­stepped, or on the other, that the area they included was always held intact. As a matter of fact, they varied considerably from time to time. Before the rise of Timur, for instance, invasions, by the Moghuls, of Shásh, Turkistan and even Mávará-un-Nahr were of common occurrence, while at times in their later history, they extended their sway over districts in the east which did not properly belong to Alti-Shahr. In the same way, when the affairs of their neighbours were in the hands of strong rulers, portions of Moghulistan were cut off for a time, and numbers of the inhabitants seem to have had no scruple in joining the service of the successful conqueror of the time being.

The section known as Moghulistan differed widely, in most respects, from its companion province on the south. It was a land of mountains, streams and lakes, of upland pastures and steppes, of wooded valleys and even forests; for while it lay north of the regions which can only become productive if reached by the monsoon from the southern seas, or if irrigated by the art of its inhabitants, it was yet far enough from the blighting snows and sunless days of Siberia, to be in most parts clothed with natural verdure of some kind. Its altitudes were moderate, and its climate, therefore, as Mirza Haidar describes it, cool and invigorating, though to Europeans, accustomed to live within the modifying influence of the sea, it would appear to be subject to extremes of temperature. Deserts in the proper sense of the word—sandy or stony wastes, with little or no vegetation or water—nowhere existed, except on the ex­treme north-western confines, and wherever the word ‘desert’ occurs in the text, when referring to Moghulistan, it is because the author has used the Persian or Turki equivalent, though the real meaning would be ‘the open country,’ or the ‘country devoid of towns and cultivation’—the ‘steppe’—a feature which no English word will describe.

However this may be, it was a land in every way suited to the habits and customs of a sparse population of nomadic graziers and shepherds, and it accordingly evolved, or at least attracted, a race whose requirements it fulfilled. But the peaceable pursuits of raising flocks and attending herds were not the only avocations of a people with the traditions of the Moghuls. Perhaps their chief requirement was a land whence they might raid on their settled and more wealthy neighbours, and whither, if beaten, they could retire and find a refuge—a land, in short, so inaccessible and unproductive to all but them­selves, that it formed, at once, a base for their own description of warfare, a secure retreat, and an inhospitable waste for the pursuing enemy; for where they moved, the whole resources of the country—the food supplies, the transport, the shelter— moved with them, and were used to meet their wants alone. There could have been no forts or towns or immovable pro­perty, worthy of the name, for an invader to destroy, and no stationary population, left undefended, upon whom he might wreak his vengeance; for the women and children and the aged all formed part of the expedition, and were doubtless employed or disposed of, in much the same way while the tribe was on the march, as while at home in their own encampment. In times of peace—or rather of inactivity—they probably bred, besides the camels and sheep, which were their principal food-pro­ducers, large numbers of ponies, for it was on these that all depended, when wars or forays were on hand. Mobility must have been the quality they relied on more than any other, both in attack and retreat, and we find them baffling their enemies more by their movements than by their fighting power. Indeed, fighting in its proper sense must have been with them, as with most of their neighbours, a pursuit very sparingly indulged in. We read, it is true, of armies counted by hundreds of thousands, and of pitched battles when thousands were killed on either side, but apart from the facts that populations such as those in question could not have put such masses of fighting men in the field, and that numbers among Orientals are at all times used as mere figures of speech, it is remarkable that where a particular battle or other special incident is described in detail, there are usually indications that the numbers engaged were very small indeed.

This must have been more especially the case with the tribe of Moghuls and the other nomads who allied themselves with them, after the first quarter of the sixteenth century. During Amir Timur's reign, the Moghuls under Kamar-ud-Din, one of their best leaders, seem to have been always beaten when met by the Amir's troops, yet they were never thoroughly repressed until the great conqueror had put forth all his strength and resources in following them up, in separate bodies, to the farthest confines of their territory. His problem was not how to beat the Moghuls in battle or to invade their country, but how to catch their mobile forces in sufficient numbers, to make an impression on the nation at large; while, on their part, the Moghuls never seem to have attempted an incursion into Timur's dominions, except when he and his troops were engaged in prosecuting a war elsewhere. Later, the same difficulty occurred to Ulugh Beg Mirza, who only succeeded in dealing them a heavy blow, through the accident of a piece of treachery on the part of one of their own people, by which he was afforded an unlooked-for opportunity. And later again—within the sixteenth century—when the Kirghiz and Kazáks had to a great extent supplanted the Moghuls in what had been the latter's own land, and the nominal Khans of the country (Sultan Said and his successor) had their headquarters at Kashghar, it seems evident, though Mirza Haidar says little about it, that the tactics of the nomads left them practically masters of the situation. Yet even in those days, when brought to battle, they are said usually to have been beaten. Perhaps the only power which the Moghuls stood in fear of, after the days of Timur, was that of the Uzbegs, when these were first rising to power. Under Shaibáni Khan the confederated tribes of Uzbegs still possessed the characteristics and qualities of nomadic nations, and it is not a little remarkable that the Moghuls, so far from dealing with them as they were accustomed to do with others in the low countries, enlisted under Shaibáni in large numbers, and assisted him against the more civilised forces of Baber and the Khorasáni Mirzas. They seem to have feared to measure themselves with those who could use their own tactics against them, or fight them, indeed, with their own weapons.

In many places in Mirza Haidar's history, as well as in the Zafar-Náma and other books, mention is made of the ‘cities’ or ‘towns’ of Moghulistan; but as the same words must necessarily be used when speaking of the settled countries of Mávará-un-Nahr, Turkistan, and Alti-Shahr, they are somewhat misleading terms to apply to the auls, or encampments, of a nomadic people. One native writer, whose book dates from the first half of the fourteenth century, presents, in a few words, a telling picture of Moghulistan in his day—or part of Turkistan as it was then still called. “Since the region has been devastated by the arms of the Tatars,” he writes, “it is inhabited only by a scanty population. According to what I have been assured by a man who has travelled through the country, there is nothing to be seen in Turkistan but ruins, and more or less obliterated remains. From a distance one sees a well-built village, the environs of which are covered with beautiful verdure; but on approaching, in the hope of meeting with some inhabitants, there are found only houses completely de­serted. The population is composed entirely of nomads—that is, of shepherds and graziers who never occupy themselves with cultivating the land or sowing crops. There is no other verdure but that of the steppes, which grows naturally.”* That towns, in the true sense of the word, had existed in the land is thus correct, but they had been built when others possessed and governed it, and before it had become the home of the Mon­golian nomads. The Uighurs, a Turki tribe of considerable cultivation by comparison, had owned the greater part, if not the whole, of the country up to less than a century prior to the rise of the Mongols, and were probably the founders of several towns of more or less importance; while the whole of Moghul-istan had, during the interval, been occupied by the Kara Khitai, whose people, although perhaps much mixed with nomad tribesmen, seem also to have been capable of building cities and carrying on cultivation. The advent of the Mon­golian hordes, however, under Chingiz and his successors, put an end to all such practices, and from that time till the date when Mirza Haidar's history closes (and probably for long after also), the country reverted to a purely pastoral condition. When, therefore, we read of the cities of Taráz, Bálásaghun, Aimal, Bishbálik, Almáligh, etc., within the Moghul period, it can hardly be that Moghul cities are intended, but rather encampments—some of them, perhaps, central in situation and well inhabited—standing on or near the sites of the remains of these places.

In the more advanced of the countries conquered by the Mongolian armies—in Persia, Mávará-un-Nahr, Turkistan, etc. —no obliteration or even systematic destruction of towns (except in the course of the wars), and no reversion to a nomadic level, seems to have taken place; but the difference in the case of Moghulistan was that, in that country, the nomadic tribesmen of the steppes immediately to the eastward —the true Mongolia—pressed in, and appropriating the land for their own habitation, took root, while in the lower countries they settled as rulers only. Those of the Mongols who, after the first invasion, stayed in the conquered countries with their governing Khans or chiefs, probably intermarried, after a time, with the settled population, and were soon absorbed; while in what became known—and partly for this very reason—as ‘Moghulistan,’ or the ‘land of the Mongols,’ the invaders found a suitable home, and establishing themselves as one of the nations of the soil, became, for a time at least, the dominant one. As generations passed, they tended, no doubt, to lose their identity by intermarrying with other races already sparsely inhabiting the region, but in this instance their absorption would be a slow process, as compared with the few left among the overwhelming populations of the lower countries in the west. The aul was probably a tribal community, and the number of the Moghuls was perhaps greater than that of their neighbours, while the life of the steppes rendered a certain degree of isolation inevitable. All these circum­stances would combine to retard a fusion of races, though it may not, as far as the evidence goes, have obviated it in the end.

Here, then, no cities sprang up, while those already in existence soon fell to ruin. But the Musulmán writers, who constantly confuse the words for ‘city’ and ‘country,’ and even ‘nation,’ would be unlikely to draw any distinction between a built and permanent town, and an encampment of felt tents—an urdu or an aul, as the Turki words are. In several cases Mirza Haidar mentions towns of Moghulistan as existing in the form of ruins only, and he is explicit on this point. But he nowhere describes one as an inhabited centre at his own time, though it is only reasonable to suppose that he would, at least, have made some mention of them had they existed, in the same way that he speaks of, and even describes, those of Alti-Shahr. The aul, or collection of felt tents, pitched without order or any view to permanency, near the banks of a stream, and in the centre of some district where pasture was near at hand, was probably the nearest approach to a town at the period our history belongs to. Here, possibly, a square or oblong shed of brown mud bricks, ornamented with yaks' tails, antelopes' heads, and rows of small, coloured flags, may have stood to represent the urdu proper, or reception-room and court-house of the chief; while round it were scattered the dome-shaped tents of willow laths, covered with sheets of felt—all grimy and greasy—and ready at any moment to be taken down by the women of the tribe, and packed, with the rest of their domestic belongings, on the backs of the camels. Of forts, walls, or streets there could have been no sign. In the daytime, the ground on which the encampment stood would have been black with the dried droppings of sheep, a foot in depth, which, whirled into the air by the west wind, would pervade, with its pungent smell, the valley for a mile round, and cover everything, even the surface of the river, with a film of black. By the evening, this unsavoury carpet would be overlaid by thousands of sheep, driven in from the neighbour­ing glens and packed close, in scarcely separated flocks, for the night, while outside these, long rows of camels would kneel at their tethering-ropes, and groups of shaggy ponies stand fastened to the doors of their masters' tents. Near at hand, it may be, some ruined walls or weather-worn mounds pointed to the remains of an Uighur town, or fort, destroyed hundreds of years ago, and having no more connection with the life of the people of the aul than have the ruins of an Elizabethan castle, or a Norman keep, with the inhabitants of a neighbouring county town in England at the present day.

Encampments such as these would not only leave no trace of where they stood, but even their names would be unlikely to endure in history. Such were, no doubt, At-Báshi, Kuchkar, Jumgál, Jud Kuzi, and others, so often spoken of in the Second Part of the Tarikh-i-Rashidi, and several more that are mentioned in the Zafar-Náma and the Tarikh-i-Jahán Kushai, now impossible to identify. They were typical of the Moghuls as a race—of a nation devoid of constructive instincts, destined only to fallow the land and then make place for others.

The period subsequent to the conquest of Chingiz's successors was one when disorder and intolerance prevented European travellers, who might have left a description behind them, from traversing the country of the Moghuls; but a side-light is shed on the subject by a brief mention in Rubruk's narrative of his visit to Mangu Kaán (Chingiz's grandson) at Karakorum in the year 1253, and consequently only just at the outset of the establishment of the Mongols in the region in question. Kara-korum was then the Mongol capital: it numbered among its inhabitants many Chinese, Uighurs, and other comparatively cultivated people, and was, presumably, if not the only per­manent Mongolian town, at any rate by far the best of them. Yet the walls only measured about a mile in circumference, and Rubruk relates of it: “You must understand that if you set aside the Kaán's own palace, it is not as good as the borough of St. Dennis; and as for the palace, the abbey of St. Dennis is worth ten of it! There are two streets in the town, one of which is occupied by the Saracens, and in that is the market­place. The other street is occupied by the Cathayans, who are all craftsmen … There are also twelve idol temples belong­ing to different nations, two Mahummeries, in which the law of Mahomet is preached, and one Church of the Christians at the extremity of the town. The town is enclosed by a mud wall and has four gates.”* The Chinese travellers of the thirteenth century give no description of the inhabited centres of Moghul-istan which they passed through, though one of them, Chang Té (who seems to have had an eye for irrigation) mentions briefly that at Almáligh there were reservoirs in the market­places, “connected by running water.” Farther westward also, in the valley of the Chu, he remarks that the country was intersected in all directions by canals which irrigated the fields, while numerous ancient walls and other ruins were seen which he attributed to the days of the Kitan or the Kara Khitai.* But all these marks of civilisation had been swept away in Mirza Haidar's time, as he himself implies in his description of the antiquities of that part of the Khanate which, he says, was formerly known as ‘Kara Khitai.’

In attempting to make clear the condition of Moghulistan and the neighbouring regions of Central Asia, perhaps the chief perplexity is experienced in unravelling the nomenclature of places and people. The names of countries and towns not only changed with time, but different nations applied, frequently, a different designation to one and the same place. Thus, names often arose at a certain period, were employed by writers for a time, and again fell out of use. The Mongols, for instance, during their ascendency, gave names of their own to many places which, after the decline of their power, became obsolete. In the same way, the conquests of Timur seem to have given birth to names that are peculiar to that period alone, and were perhaps only in vogue among those connected with the con­queror's court or his armies. This circumstance, in addition to the habit of applying nicknames to tribes and nations, may account for many of the difficulties that surround the identifica­tion of names mentioned by various authors, and should act as a warning, in the case of the tribes, not to attach too readily a racial significance to every name that is met with.

To the Chaghatais of Mávará-un-Nahr and the west, Moghul-istan was known, in the 13th and 14th centuries, by the name of Jatah, and though this was only a term of depreciation, or a nickname (as will be explained below), it is employed in the gravest way by several Persian authors of the Timuri period, whose works have become standards of historical reference. What is perhaps more curious to remark is, that the name of Bishbálik, which so often occurs in mediæval histories and travels, and in the Chinese historical annals, is that by which the Chinese knew the Khanate of Moghulistan, during the earlier part of the period over which Mirza Haidar's history extends. This name had originally no connection with the Moghuls or their dominion, but was a survival from the days when the region had belonged to the Uighurs. Properly it was the name of a city only, which had been built by the Uighurs, and, having become their capital, had lent its name to the whole kingdom. The meaning, in Turki, is ‘Five Cities,’ and seems, possibly, to have indicated the capital of the five divisions, or provinces, into which the country of the Uighurs, at that time, (about the middle of the ninth century) was divided; or other­wise, it may have meant that the tribe was divided into five sections, or the town (as one authority has it) into five quarters.* However this may be, the Chinese knew the country by its Turki name (which they sometimes translated into its Chinese equivalent—Wu-chêng), while they gave the city itself the Chinese style of Pei-ting, or ‘Northern Court’; and subse­quently (early in the fifteenth century) changed that of the whole country from ‘Bishbálik’ into ‘Ili-bálik.’

The town of Bishbálik was situated on, or near, the site of the modern Urumtsi, and the country of which it was the chief place, extended to the westward and north-westward, as well as beyond the southern slopes of the Tian Shan. Like the rest of this part of Asia, it fell into the empire of Chingiz Khan, and, after his death, passed to his son Chaghatai. Later again, in the time of the Chinese Mings, the official historians of that dynasty described the limits of the region in such a way, as to leave no doubt that the country they termed Bishbálik was, indeed, Moghulistan. “Bie-shi-ba-li,” says the Ming Shi,* “is a great empire in the Si Yü [countries of the west]. It is bordered on the south by Yü-tien [Khotan], on the north by the country of the Wa-la [the Oirát Kalmáks], on the west by Sa-ma-rh-han [Samarkand], and to the east it is contiguous with Huo-chou [Kara Khoja]. It is distant [probably the urdu of the Khan is meant] from Kia-Yü-Kuan in the south­east, 3700 li. It is believed that Bie-shi-ba-li occupies the same tracts as, in ancient times, Yenki or Kui-tsz.”* As a description of the land and people, the Ming history adds:— “The country of Ili-ba-li is surrounded by deserts. It extends 3000 li from east to west and 2000 li from north to south. There are no cities or palace buildings. The people are nomads living in felt tents, and exchanging their abode, together with their herds, in accordance with the existence of water and pasture land. They are of a fierce appearance. Their common food is flesh and kumis. They are dressed in the same fashion as the Wa-la.”

Many embassies are recorded in the Ming Shi as having passed between Bishbálik, or Ili-bálik, and the Chinese capital, which make it appear that the Khans of Moghulistan and the Dughlát Amirs paid tribute to China. Whether the position of tributaries was imposed upon them by superior force, or whether, as is far more probable, the missions were sent to cultivate the friendship of a powerful neighbour, and to profit by an exchange of presents, is nowhere intimated; but the result remains, that from the time of Khizir Khwája, about the year 1391, down to the reign of Isan Bugha II. in 1456, each successive Khan (as we have seen in Section II.) sent one or more tribute-bearing missions to the Ming court. After the latter date, it appears to have been settled that ‘Ili-bali’ was to send tribute every three years, but no further mention is made of any special mission, and it is possible that not long afterwards, the growing weakness of the Mings caused the custom to fall into disuse.

It may be thought strange, perhaps, that Mirza Haidar's his­tory nowhere speaks of intercourse with China, or mentions that the Moghul Khans performed these acts of homage to her Em­perors. Whether he omitted any allusion to them, from a feeling that the payment of tribute was derogatory to his ancestors, or whether he thought the subject not worth recording, must remain a matter of conjecture. In all likelihood the latter was the reason, as we shall see, further on, when referring to similar missions from Uighuristán. The proceeding was, presumably, looked upon as a mere form, or indeed a farce, and therefore attracted no attention on the part of the historian. Still, his silence on the point cannot be taken to disprove the statements of the Chinese, for these are explicit and persistent, and can hardly be otherwise than correct as records of bare facts. What is remarkable, however, is that the same Khans and Amirs who were bowing the knee to China, whether in good faith or other­wise, had no hesitation in measuring their strength with so great a soldier as Timur. The fact that his power was near and visible did not inspire them with respect, or deter them from raiding into his territory and otherwise provoking his vengeance. But the Chinese, then as now, seem to have possessed the art of attracting the outward forms of submission from distant States, though they had no power to exact the reality.

Passing now across the mountains to the south-east, an entirely different land and people present themselves, in the province that may be most appropriately and correctly called Alti-Shahr, or the ‘Six Cities’ of Eastern Turkistan. Here the low ranges and open valleys of the steppes, are changed for gigantic mountains on the one hand, and sandy deserts on the other; the aul of felt tents for the town of brown mud-bricks and close-packed bazars; the grazing grounds and hill-side torrents for cultivated fields and irrigation canals; while, above all, the thriftless, irresponsible nomad is replaced by the culti­vator and artisan, with all the elements of stability that their industry confers upon a people. Though the area is large, the culturable and habitable spots in it are, out of all proportion, small. One modern traveller describes it as a huge desert fringed by a few small patches of cultivation. Another tells us that a bird's-eye view of the country would show a huge bare desert, surrounded on three sides by barren mountains, along the bases of which would be seen some vivid green spots, show­ing out sharp and distinct like streaks of green paint on a sepia picture. At the western end, the cultivation is of greater extent and more continuous than in the eastern half, where the oases are small and separated from each other by stretches of desert, which increase in length as the traveller passes eastward; while the eastern extremity is desert pure and simple. The oases, however, are fertile enough in themselves, for every drop of the water brought down by the streams from the mountains, is drawn off into irrigating canals, and made to reach as far as possible toward the desert, for agricultural purposes.

All except the shifting sands of the central waste, appears to require only water to render the ground fertile; but water is precisely the boon that is withheld. Though the monsoon clouds roll in every summer across the mountain masses on the south, they seldom do more than tantalise the cultivator, who watches them in the hope of rain. Indeed, rain but rarely falls, and a Chinese traveller of ancient days* has recorded the incredulity of the people, when told that water for cultivation fell from heaven, onto the favoured soil of his country, and rendered it independent of melted snow from the mountains. They laughed, and cried: “How can heaven provide enough for all?”* Snow may be less of a rarity, but so dry is the atmo­sphere, that when a fall occurs, it evaporates after a few hours, and leaves the surface of the ground scarcely moistened.

That a land of this nature should support only a small population, and be too poor, as Mirza Haidar tells us, to main­tain an army on its own produce, is not surprising. Whether its weakness as a State is owing to this or to whatever other cause, it has always been an easy prey to invaders, and has seldom had a native ruler within historic times. Its population has been a Turki one for ages past, and the Uighur branch of that race may be regarded (as far as historic times are concerned) as the original owners of the soil, and the parent stock of the bulk of the present inhabitants.* That in later times, at least, they were not an aggressive race appears evident from the little we hear of them, and that they had some capacity for crafts and literature seems also to be established. No doubt the tendency of such a people would be to live peaceably under any govern­ment strong enough to repel external enemies; so that when Mirza Haidar tells us that Alti-Shahr was “free from the dis­cord of men and the trampling of hoofs, and became an asylum for the contented and the prosperous,” he is probably drawing a picture of the country not only true of his own time, but one that serves for several centuries both before and after it.

During the periods that the Dughlát Amirs and Moghul Khans held sway, we hear of expeditions being sent to overrun Badakhshán, Ladak, and other weak States, but these were evidently undertaken by foreign rulers with their foreign troops, and not by the people of the country; indeed, we come much more frequently upon records of invasions which they themselves underwent at the hands of various enemies, such as the Arabs, the Mongols, the Kara Khitai, and even the Kalmáks. In the raids of the Moghuls into Western Turkistan and Mávará-un-Nahr, in their wars with Timur and Ulugh Beg, and their long campaigns with the Uzbegs, it is probable that the natives of Alti-Shahr took little part, for they are never mentioned as combatants. They had, in short (and have still), all the attributes of a lowland and unwarlike people, whose wealth excites the cupidity of aggressive neighbours, but the nature of whose country and customs prevent them from becoming themselves aggressive.

It would be interesting to learn what the armies were com­posed of, that invaded, in the reigns of Abá Bakr and Sultan Said, Badakhshán, Chitral, etc., Ladak, Tibet and Kashmir. In all likelihood the numbers were very small—to be counted in some instances by hundreds rather than by thousands—while most of the men were probably mercenaries from countries other than Alti-Shahr. Mirza Haidar nowhere specifies the races which furnished the rank and file of these forces. When entering on the conquest of Kashghar, in 1514, he gives an analysis of the chiefs of Sultan Said's army, nearly all of whom were Moghuls of various clans, or members of tribes who had long previously thrown in their lot with the Moghuls, and the number of tribal followers that each chief brought with him is specified in each case. If the figures given are correct—and as they are not mere round numbers, they appear as if intended to be exact—it is evident that the tribal following which each chief could muster was a mere handful, for the total of the tribesmen mentioned does not approach that of the entire army of 4700 men, as he states it.* The remainder must have been mercenaries and adventurers who were, no doubt, to be found in abundance all over Central Asia in those times, in the persons of Kipcháks, Turkomans, Afghans, Kárluks and what not. On this occasion, too, a great effort was being made and a prize worth winning was at stake; the army was raised, moreover, in Farghána and Moghulistan, and not in peaceful Alti-Shahr. Thus it was probably a much more numerous one than those afterwards employed on distant expeditions beyond the mountains, though it may be fairly conjectured that the composition was very similar in all cases. In the expedition of Sultan Said against Ladak, Kashmir and Tibet in 1532, the author puts the total of the army at the round figure of 5000 men, but in this instance he gives none of the minute parti­culars that he records with regard to the 4700 and their supports, who invaded Kashghar. The round number is likely, therefore, to be one of the many similar exaggerations in which his book abounds; for it is improbable that as large a force would have been thought necessary for this enterprise as for the wresting of Kashghar and the whole of Alti-Shahr from so formidable an enemy as Mirza Abá Bakr. He tells us, it is true, that Ladak was incapable of supporting the Khan's army, but this might have been the case with even half 5000 men and their complement of horses.

Perhaps the most noteworthy feature of this land of the Six Cities, and the one that has chiefly struck the imagination of both ancient and modern writers, is the central desert with its moving sands and buried towns. It is referred to in Chinese writings of more than 2000 years old, by travellers who gave the region the name of Liu Sha, or ‘Moving Sands,’ from its chief characteristic and most obvious peculiarity;* and it was made known to Europeans through the graphic accounts of it which Marco Polo left on record. The phenomenon of the shifting sands could hardly have escaped Mirza Haidar, and the story he tells of the overwhelming of Katak, with its mosque and minaret, is one of the best pieces of description in his book. It is almost an exact counterpart of that told by Hiuen Tsang in the eighth century, of a town between Khotan and Pima (Pain?) which was said to have been overwhelmed by the same agency, some hundreds of years previously. In this case, neglect in the proper worship of a Buddhist idol was the cause, while in the later one the Musulmans detected the wrath of God. The earlier calamity too, is said to have been predicted by a pious Arhát seven days before it occurred. At first a great storm of wind arose, which carried sand and soil before it, while “on the seventh day,” continues the narrative, “in the evening, just after the division of the night, it rained sand and earth and filled the city … The town of Ho-lo-lo Kia is now a great sand mound. The kings of neighbouring countries, and persons in power from distant spots, have, many times, wished to excavate the mound and take away the precious things buried there; but as soon as they have arrived at the borders of the place, a furious wind has sprung up, dark clouds have gathered together from the four quarters of heaven, and they have become lost.”*

Similar stories are in the mouth of nearly every native of the country down to the present time, and several have been recorded by Dr. Bellew and Sir Douglas Forsyth. These travellers themselves visited some of the sand-buried ruins in the neighbourhood of Yangi Hisár. One of them was the fort of a Uighur chief called Tokhta Rashid, which had been destroyed about the eleventh century by Arsalan Khan, and afterwards overwhelmed by the sand. Another was the Mazár, or shrine, of one Hazrat Begum, which had been first swallowed up, and again, at a later date, left free by the receding dunes. The neighbourhood of the latter ruin is described as “a perfect sea of loose sand, advancing in regular wave lines from north-west to south-east. The sand dunes are mostly from ten to twenty feet high, but some are seen, like little hills, full a hundred feet high, and in some spots higher. They cover the plain, of which the hard clay is seen between their rows, with numberless chains of two or three or more together in a line, and follow in successive rows one behind the other, just like the marks left by wave ripples on a sandy beach, only on a large scale. Towards the south-east these sand dunes all present a steep bank in the shape of a crescent, the horns of which slope forwards and downwards, in points, to the ground …” The process of submergence, Dr. Bellew found to be usually a very gradual one, until the symmetry of the dune, becoming broken by an obstructing object, its loose materials subside, and thus overwhelm the obstruction. In the instance of one of the buildings inspected, it was found that “a chain of three crescentic dunes, side by side, had advanced in line across the plain, till one of the outer crescents had struck the walls of the court of the tenement, and growing up, had, in time, over­topped, and thus overflowed and filled its area by its downfall; whilst the other two crescents at its side, continuing their unobstructed course, maintained their proper form uninjured.”* The rate of progression the writer was unable to determine, as it depends on the varying force of the propelling power, the slope of the land, and the obstructions on its surface. The operation, however, is the same as in the well-known instance of Eccles church, on the coast of Norfolk, only on a larger scale. By 1839 the whole of the church, except a portion of the tower, had been buried; by 1862 the tower had nearly emerged again, while in 1892 the whole building rose free from the level of the strand, the dunes having passed to its landward side.

The phenomenon thus seen in operation, explains how the town of Katak, and others mentioned by Mirza Haidar, became engulfed, and confirms the stories still current in Eastern Turkistan of ruined towns, or buildings, now and then appear­ing for a while and being again submerged.* In the extreme east of the country, the sandy desert is found at its worst, and it is in connection with this quarter that most of the tales of weird horrors have their origin. How deeply the superstitious mind of the Asiatic may be impressed by these wastes of moving sands, and how little reason there is to wonder at the stories of ghosts, demons, and visions with which he has in­vested the region, may be judged by General Prejevalski's vivid description of it. “The effect of these bare yellow hillocks,” he writes, “is most dreary and depressing when you are among them, and can see nothing but the sky and the sand; not a plant, not an animal is visible, with the single exception of the yellowish-grey lizards (Phrynocephalus Sp.) which trail their bodies over the loose soil and mark it with the patterns of their tracks. A dull heaviness oppresses the senses in this inani­mate sea of sand. No sounds are heard, not even the chirping of the grasshopper; the silence of the tomb surrounds you.”*

Hiuen Tsang's description scarcely varies from that of the Russian traveller. “These sands,” he says, “extend like a drifting flood for a great distance, piled up or scattered before the wind. There is no trace left behind by travellers, and oftentimes the way is lost, and so they wander hither and thither quite bewildered, without any guide or direction. So travellers pile up the bones of animals as beacons. There is neither water nor herbage to be found, and hot winds frequently blow. When these winds rise, both man and beast become confused and forgetful, and then they remain perfectly dis­abled. At times, sad and plaintive notes are heard and piteous cries, so that between the sights and sounds of this desert, men get confused and know not whither they go. Hence there are so many who perish on the journey. But it is all the work of demons and evil spirits.”*

And if the superstition of the Asiatic is moved by the mystic scenes of the desert, his cupidity is also stirred by the legends of buried riches which the submerged cities are supposed to contain. Traditions lose nothing from age or from being often repeated, and no doubt, the stories of hidden treasures are now —and, indeed, were in Mirza Haidar's time—ancient enough to acquire a very strong influence on numbers of the population. From time to time ornaments, vessels, images, and coins of great curiosity are unearthed, but their value to the finders, whose only interest lies in the worth of the metal they are made of, can scarcely be great. Perhaps the only systematic exploitation of the ancient sites, ever undertaken, is that of Mirza Abá Bakr, Amir of Kashghar, so fully described by our author. It may be dated about the end of the fifteenth and beginning of the sixteenth century, and we may infer that nearly everything of intrinsic value was brought to light, while much that was of antiquarian interest was destroyed, so that when, at some future time, civilised explorers come to investigate the ruins, and find little to reward their labours, they may feel themselves indebted to the cupidity of Mirza Abá Bakr for their disappointment. The tales which the author tells of the riches accumulated by the Mirza, may safely be regarded as, in a great measure, fabulous; but it is precisely tales such as these that have given rise to the inflated estimates of buried wealth so common in the country, even at the present day.

Here and there valuable records of the past may still be forthcoming from the submerged towns, like those obtained in 1874, by Sir D. Forsyth, who enumerates a figure of Buddha of the tenth century, a clay figure of the Hindu monkey-god Hanuman, and Hindu women's ornaments, all pointing to that close intercourse with India which we know, from other sources, to have existed in times before Muhammadanism prevailed and crushed it. He also obtained several Greek coins of great antiquity and interest. Among these is mentioned especially one of Antimachus, dating about 140 B.C., and another, of Menander, from about the year 126 B.C., while a third, an iron one of Hermæus, might, it was thought, prove even older than either of these. But it is not necessarily among the ruins buried by the shifting sands, that relics of remote ages will be found. Very ancient remains are known already to exist at various points along the southern spurs of the Tian Shan, though nothing has yet been ascertained as to the age they belong to. It was near the town of Kuchar—the ancient Kuitze of the Chinese—that Captain Bower found the famous birch-bark manuscript, written in Sanskrit and dating from the fifth century, while he also points to ruins of cities which, though buried beneath the present level of the country, have no connection with the shifting sands.*

But it is time to turn from the land of the man to the man of the land.