RUINS OF DESERT CATHAY

Sir Aurel Stein KCIE
Macmillan and Co
St. Martin's Street, London, 1912
2 volumes, 546 and 517 pages, index, 333 illustrations, color plates, maps

This is the Personal narrative of Stein's second exploration in Chinese Turkestan - 1906-1908 and describes the same expedition as he does in the official report, Serindia. Thus, while the events are the same the writing style is more popular and Stein dispenses with the elaborate detailed descriptions of each and every artifact he found. It does not have the 94 excellent maps, but several others instead. The printing reproduction resulted in many of the photographs being of much higher quality. Since I have made a summary of Serindia, I here focus on Stein's comments not found in that massive work. The two works overlap of course but each contains enough separate information to make it worth while to read both. The differences are quite striking. For instance in this personal narrative Stein described in detail his reception in Afghan Wakhan valley and names the Afghan military and civil officials who were such gracious hosts. But in Serindia the chapter describing his passage through the Wakhan does not even mention that it is in Afghanistan. In this personal narrative Stein describes in much more detail and vivid prose the intense weather conditions, both winter cold and summer heat, and its impact on himself and his assistants. He is effusive in his descriptions of his beloved high mountains. But everywhere he describes the soil, vegetation, hydrology and everything in view. Here also we read about receptions and meetings, including 'Dastarkhans' with the Chinese Ambans and Mohammedan Begs.
Here is the table of contents for Ruins of Desert Cathay .

Preface
Stein begins with comment that this volume is designed as a popular account of his personal record for explorations of 1906-08. He refers to his previous book, Sand-based Ruins of Khotan for the personal account and Ancient Khotan, the official detailed account of the first expedition over part of the same territory. He expresses his gratitude to the government officials who authorized, supported and helped fund these expeditions. He names many of his important assistants and credits them for their extensive services. He also gives great credit and thanks to the many Chinese officials whose attentive help was essential. Then he also cites the important scholarly assistance received from many European specialists in the many and various fields needed to understand the material he brought out of China.
The first expedition was limited to an area close to Khotan, while this one ranged widely throughout the Tarim Basin and into western China proper. But even around Khotan Stein proved the significant influence of Indian as well as Chinese and classical art and culture. He notes that the geographical features of the area have such significance on its history that he will include description of this as well as archeology and culture.
He then gives a brief synopsis of the expedition's route and names some of the more important sites. This preface would serve the casual reader as a summary of the two volumes.

Chapter I - Between Hydaspes and Indus
The departure from Shrinagar had to be carefully times, to early would make crossing the snow covered passes impossible, but too late would make passage through gorges filled with raging rivers swollen by snow-melt impossible. With the text of Ancient Khotan nearly completed he was eager to get going. Remarkably he was able to continue checking proofs even when in desolate deserts and high mountains during the expedition. He chose to take a different route though the mountains of North-west India than he followed on the first expedition in order to explore new regions. These required special approvals of the remaining semi-independent local rulers in these yet unconquered mountain valleys. This route was further west, through Swat and Chitral only recently pacified to some extent.
Stein departed Srinagar on 2 April 1906 and went to Peshawar on 10 April. Thence he assembled his party at Abbottabad ( recently in the news in 2011) on 17 April.

Chapter II - Through Swat and Dir
On 24 April the heavy baggage departed Abbottabad with Stein following on the 26th by train for Dargai and then by vehicle to the famous fortress at the Malakand Pass. In this account Stein provides much more local color and discussion of the recent military campaigns than he does in the official report along with photographs of Malakand and Chakdara forts. He spent 28 April with correspondence. (Another amazing activity is his massive correspondence sent to and received from England and India throughout the expedition.) Already in the Swat valley Stein was examining Buddhist shrines and considering the routes Alexander the Great took when entering India. The area was called Udyana in ancient Buddhist times so Stein uses that name. He crossed the Katgala pass from the Swat to Panjkora rivers observing ancient hillside forts and towers along the way. He comments on the military outposts and patrols encountered along the road, set out to insure his safe passage. He rushed past military garrisons and small forts in his hurry to reach the Lowarai Pass. Along the way he made the required customary stops at the capital - fort of each local ruler. A downpour forced a temporary halt at the miserable capital of Dir.

Chapter III - Across the Lowarai
Finally on 3 May, Stein was able to march ahead to Gujar, a tiny military outpost, at the foot of the Lowarai, already at an elevation of 7,800 feet. At each pass he was met with recruited carriers on the near side and then another team of workers at the summit for the trip down the other side. For such crossings in heavy snow but with spring sunshine already causing melting, Stein crossed at night -usually from about 1AM to dawn. This reduced the chance that the passage of loaded men would cause an avalanche. He reached the crest - 10,200 feet - at 5 AM with most of the teams still following. at 8 AM the loads were transferred from the Pathans from Dir to Chitrali carriers. The route down from the Lowarai was more difficult and dangerous than the ascent. But they made it safely into Ziarat in the Chitral valley. There he entered a "different world" inhabited by Dard tribes. He was entertained by the British officers at their mess.

Chapter IV - In Chitral
On 5 May he continued on to the local capital, all the while noting the scenes of fierce battles of the recent past when Chitral was besieged in 1895. During his three day halt at Chitral he accomplished his usual administrative duties including anthropometric measurement of selected local tribesmen, but was also entertained by another British garrison and by the local chieftain - Mehtar at his castle. He collected not only physical data but also linguistic and legendary information.

Chapter V - Through Mastuj
Stein hated to leave the hospitality found in Chitral, but time was pressing and already the Afghan army unit was waiting for him in the Wakhan Valley at Sarhad. He left on 9 May. It took three days to reach Mastuj through narrow mountain gorges. But he did not fail to stop long enough to make a cast of a Buddhist inscription found high on a sheer rock wall. They passed more locations that were scenes of the fighting. He reached Mastuj on 11 May and found another Sanskrit inscription and Buddhist stupa. Then it was on to a series of small mountain villages before reaching Sanoghar fort and then Mastuj fort. On 13 May Stein continued northward toward the next pass. The next significant village was Miragram, where he was met and entertained by another local chieftain, Obaidulla Khan. By evening of the 14th of May he reached Shuyist and more forts and watchtowers.
.

Chapter VI - On the Darkot Pass
Now Stein was getting into the real mountains. From Kankhun-kuch it took two more days to reach the Baroghil Pass which is low - only 12,400 feet elevation. But as ever Stein was not content with this simple route. No, he had to at least stand in the Darkot Pass as well, because that was the pass used by one of his Chinese heroes in the 8th century. That one is only 15,400 feet high and the location of a glacier to boot, but was used by general Kao Hsien-chih in 747 AD when he was sent by the emperor to drive the Tibetans out of Yasin, Gilgit and Hunza. So on 17 May Stein took 12 men with minimal equipment for his personal excursion to mount the Darkot Pass and back starting out at 3AM. Stein used ropes and all the skill of mountain trekking to cross the dangerous glaciers. Naturally Stein writes that he was thinking of Kao Hsien-chih all during this dangerous excursion and stopped to write a note to his French Sinologist, Chavannes. At 3 PM they started back down under worse conditions due to a day of melting of the snow. They made it back to camp at 8 PM. On the 18th they turned to cross the Baroghil but were unable to reach the crest due to the deep and soft snow. They camped in the open but were encouraged when an Afghan Beg arrived with news that the Afghan army was coming on the following day to assist them down from the pass. On his first expedition Stein had promptly on reaching the Pamirs insisted on a side trip to climb Muztagh-ata. And during the very last days of this trip he took another side trip to satisfy his desire to climb another glacier, with disastrous results we shall learn in the last chapter.

Chapter VII - In Afghan Wakhan
Stein got underway at 6 AM on the 19th. That early the snow was again frozen, making passage so much easier. They reached the level of the saddle that divides the flow of water south to the Indus from that north to the Oxus. There they were met with teams of Wakhis sent to assist. Once down in a valley and able to halt, Stein was suddenly met by two Afghan officers ready to great him in the name of their commander and the King. He had to hurry and don appropriate formal dress for the occasion. Stein was chagrined to learn that the large infantry and cavalry detachment of the Afghan army sent into the Wakhan from Badakhshan had been waiting for a full month. But the Afghans paid no attention to their discomfort. The hosts were Colonel Shirfin-dil Khan the military commander and Hakim Mansur Khan the civil governor of the region of Upper Wakhan. Stein notes that he recognized the difficulties that the Afghans had endured. The military unit was subsisting in inadequate field shelter and short rations, while the scanty resources of the poor local civilians were being exhausted as well. Therefore Stein determined to move east into China as soon as possible. Stein spent 20 May at Sarhad paying the Mastuj transport laborers and then sent them back across the pass. Before leaving on the 21st he naturally had to climb a low hill to inspect the ruin of a fortress that he claimed was the one the Tibetans had defended against Kao Hsien-chih in 747. Again the daily marches up into the mountains were most difficult. Stein never fails to provide detailed descriptions of the vegetation, soil, flowing rivers, rocky gorges and every picturesque scene.

Chapter VIII - To the Source of the Oxus
The journey on 22 and 23 May was ever upward and across a saddle at Dash-i-Miza Murad. Northeastward were the Little Pamirs. To the eastern horizon was the Wakhjir Pass and source of the Oxus River. At the Bozai-gumbaz they found Kirghiz nomads who had been assigned to prepare felt tents for their shelter. They halted at the Kirghiz encampment on 24 May to allow for needed rest. Stein as always used the day for geographic - topographic study to Lake Chakmaktin at 13,000 feet elevation. The party moved on the following day. For each night's camp the Kirghiz had sent ahead their felt tents. On 26 May they reached the last point below the Wakhjir Pass at about 14,000 feet. For the crossing the Afghan colonel demanded full effort from the reluctant Kirghiz and Wakhis. The baggage by then was largely on yaks. Early on the 27th they set out to cross the Wakhjir in deep snow and a temperature of 25 below freezing. Soon even the yaks foundered in the softening snow and had to be left behind. This required all baggage to be carried by the men - at over 15,000 feet elevation. They could carry only a part of the load on each trip and then return to carry another load. Not until 2 PM did they make it to the crest of the pass at about 16,200 feet. Going down again proved extremely difficult. In mid-afternoon they finally were met by Munshi Sher Muhammad, coming with yaks from Sarikol. But it took all night for the baggage to arrive at the rest camp. Stein was delighted to be back at a camp where he had crossed in 1900. There Muhammad Yusuf, the Beg of the Taghdumbash valley, arrived with more yaks and ponies. The exhausted Kirghiz and Wakhiris were paid generously and sent back across the pass to the waiting Afghan Colonel.

Chapter IX - From Sarikol to Kashgar
Stein passed through Sarikol during his first expedition. He renewed friendships and commented on familiar scenes. This time he paused to study several fortifications that he missed during the first tour. He started down the Taghdumabash Pamir on 28 May and stayed at Tigharmansu with Muhammad Yusuf Beg who recounted to him events and his (Muhammad's) successes since 1900. At Mintaka Karaul Stein fortunately met the passing dak runners on their way from Kashgar to India and was able to add his extensive correspondence to their mail bags. His photograph gives a clear idea about these intrepid mountaineers. The weather was still very cold and snow deep. He found a fortress that he was delighted to presume was mentioned in 642 by his 'patron' Hsuan-tsang. Of course he repeats the fanciful legend about the princess who was placed there. The fortress was named Kiz-kurghan and was situated on a steep spur 700 feet above the Taghdumbash River. But Stein managed the assent with his Indian surveyors. His photo ( 35) shows the wall and his text description is detailed. The near impregnable fortress guarded the caravan route through the Pamirs and the cultivated fields of Sarikol. On 31 May he continued down the valley, 40 miles to Tash-kurghan, capital of Sarikol. He stayed in the rest house in this village completing administrative affairs. Its town fortification was in ruins but an extensive modern Chinese fortress loomed nearby. He found that the Chinese governor (Amban) , Mushi Sher Muhammad, was now host to a Russian officer, Captain Bobusheff, with Orenburg Cossack garrison in his own small fort, ostensibly to stop smuggling. Stein was much pleased that there was a new official guest house. At Tash-kurghan Stein sent Rai Ram Singh off to survey the mountains around and then north toward Muzagh-ata. Stein hurried on to Kashgar via the Tiznaf River route, which would be come impassable once the snow turned into a raging torrent. On 4 June Stein continued from Kara-kapa through the gorge to the Chichiklik plateau, near 15,000 feet elevation. Again, the topography matched in detail the description left by Hsuan-tsang. That night he made it to Tar-bashi, 3000 feet lower where he could rest with a Kirghiz Beg. On the 5th he entered the Tangi-tar gorge. On the 6th they crossed the Kashka-su Dawan ridge line, near 13,000 feet. At the next Kirghiz camp he exchanged the yaks, critical for mountain transport, for ponies suitable for more rapid movement at lower elevations. Along all this route Stein describes the flora and spectacular views of mountain ranges near and far. After moving through more defiles Stein reached the extensive fortification erected by Yakum Beg only 40 years previously. There the head-man of the next oasis, now in Turkestan, came to welcome him to Ighiz-yar. On the morning of 8 June at 4:30 Stein set out for the final ride of 60 miles into Kashgar. Along the way he was met by more delegations of local VIP's including Indian money lenders living in Yangi-hissar. It was evening before Stein reached the Kashgar city walls. He had to proceed in darkness by feel and old memories the remaining miles to the British consulate, which he reached around 10 PM.

Chapter X - At Chini-bagh, Kashgar
This was the name of the British consulate that Stein depended on so much for both logistic assistance and entre to Chinese officials. Stein noted the extensive expansion and improvements made since his stay in August and September 1900. at Kashgar Stein had to hire men, ponies, donkeys and especially hardy camels for the coming fall and winter expedition. He notes the difficulties caused by the 'indolent' work habits of the local artisans. He also had to obtain cash useful for local payments, in gold and silver by cashing supply bills in the Indian government to local money lenders. He spent some of the forced delay in checking proofs of Ancient Khotan.. He notes his delight that Kashgar was now only 20 days or so by post from London via the Russian railway system. The Anglo-Russian agrement was bearing fruit. He was also delighted when his former camel-man, Hassan Akhun, arrived to rejoin his team. Another critical addition was the Chinese secretary, Chiang-ssu-yeh (Yin Ma Chiang) who proved to be indispensable throughout the following years. Stein visited yet another ancient ruin at Och-merwan on 21 June, another stupa and two small forts.

Chapter XI - To Yarkand and Karghalik
Stein got away from Kashgar in the afternoon of 23 June, after attending a sad funeral. The move to Yarkand took 5 days, or rather nights, as the terrific heat of the day made movement then too oppressive. He then spent daylight at the homes of local head men in each village along the way, such as Yangi-hissar, Kizil and Kok-robat. He reached Yarkand on the 26th. Stein remarks on the improvements in agriculture since 1901. Outside town he was met by Tila Bai, summoned to join again. And then just outside Yarkand a large party of all the local Hindus were lined up awaiting him under their leader (Ak-sakal), Pandit Butha Mal. The Indians resident in Chinese Turkestan were always eager to pay their respects to Stein, counting him as an official of the government on which they depended for so much. They insisted on taking Stein around the city to show off to the locals their 'big shot' from the home country. Ram Singh rejoined the team, having completed extensive survey of the mountain ranges. Stein made sure to meet extensively with the local Amban, P'an- Ta-jen, whose official orders would be critical for obtaining assistance throughout. Stein mentions also much additional activity. He departed for Karghalik on 2 July.
They crossed the Yarkand and Tiznaf Rivers and on 4 July visited Kiziljai. Stein continued eastward on the 5th and 6th to reach Karghalik and was met again outside the city by the local contingent of Indian traders and money-lenders (whose usury Stein deplores). .

Chapter XII - Stay at Kok-yar
I mentioned already how surprised I am that Stein was able to remain in such close contact by dak runner with London and India. He could entrust the copies he continued to proof of his previous manuscript to these hardy mountain folk. Some correspondence with friends in London he could send via Kashgar and then over Russian post. But for official correspondence and critical texts he preferred to send by Indian post means even though it took much longer over the Karakorum or Darkot passes. Since exploration in the desert was out of the question during the worse of the late summer, he organized his calendar to spent the hot season in the mountains. Thus on 7 July he departed Karghalik going south to Kok-yar, situated at 6400 feet elevation. He describes the terrain along the Tiznaf River. He used this quiet ride to learn more Chinese from his Secretary, Chiang, but comments on how difficult it was for him. And he already knew over half a dozen languages. At Uruk-langar he found a caravan of Yarkandi traders preparing to cross the Karakorum via the upper Yarkand River. Once settled he sent his camels and ponies off to grazing ground. He sent Rai Ram Singh and his cook, Jasvant Singh, with a Chinese daroga to do extensive surveying along the Kun-lun. Stein used the opportunity also to ask the local officials to assemble a group of mountain Pakhpu tribesmen for anthropometric study. A photo shows this group of reluctant men called down from their high valley. Stein considered these representatives of an obscure tribe to be members of "Homo Alpinus" that is of Iranian stock remaining in many isolated parts of the Pamirs. Meanwhile Stein completed work on Ancient Khotan and wrote the Introduction before sending the manuscript off by dak to Kashgar. On 24 July his camels and ponies returned to be reloaded for the next expedition.

Chapter XIII - Along the Foot of the Kun-lun
On 25 July Stein departed Kok-yar to Khotan via a route through the mountains where he could conduct further surveying. He noted that the local Chinese officials went out of their way again to give him a departing "Dastarkhan" a honorary meal. And there was another 'Dastarkhan' at every village along the way. One should note that Stein's official status as a representative of the British and Indian government gave him remarkable status and provided him with much essential official support from the local authorities at every turn. He recorded his observations on this route through Ushak-bashi and Hassan Boghra Khan and Kilian, which he reached on 27 July. At these stops he set up his tent in orchards and noted the apricots were falling on and around his tent. Kilian is on another direct trade route over the Karakorum, so Stein took the opportunity to write and send another letter direct to Leh. He continued through the hills and valleys to Sanju. Near Puski he was shown a ruined stupa made of sun-dried bricks .While there he was joined by Musa who by prior arrangement brought the heavy loads of equipment and supplies over the Karakorum with the first caravan of the year. All the supplies arrived, but the transport contractor had lost many ponies while going over the pass. On 2 August Stein moved on from Puski to Duwa. On 3 August Stein rode on toward Khotan by then through burning desert. By nightfall they reached the 'Pigeons' Shrine', Kum-rabat Padshahim (My Lord of the Sands), just west of Khotan.

Chapter XIV - My Return to Khotan
Stein found the temperature at 4 AM when he arose to be 60 degrees. He made the usual food offering to the sacred pigeons. Stein was met by a cavalcade of local Begs and their attendants outside the city and given a royal welcome. Fruit and tea were served at a rest stop along with fodder for the ponies and more food for the men. With the Kara-kash River in flood stage Stein had to move down stream to Kark-kash village to find boats causing a day's delay on entering Khotan itself. More Begs and more "Dastarkhans", as well as old friends including Islam Beg and Badhuddin Khan (chief Indian merchant), who rushed to greet him, filled Stein's day and evening. Finally, on 5 August Stein rode into Khotan in the midst of a formidable entourage. The temperature at 10 AM he recorded at 100 degrees. Then came the 'official' welcoming at a road-side hall. A host of Chinese officials lead by the military Amban, T'ang Tajen, and a military guard of honor in resplendent costumes was assembled. On the following day Stein met with the newly returned civil Amban, Ch'e Ta-jen. Two days later Rai Ram Singh arrived having successfully surveyed a large area of the Kun-lun as far south as 17,400 feet up into the Hindu-tash and then between Kok-yar and Khotan. Stein spent several days preparing both for the immediate return expedition into the mountains and the fall expedition eastward into the desert. On 11 August they set out southward to Langhru near the Kara-kash river. There he found the remains of another fort of sun-dried bricks designed to guard the opening of the pass from the mountains to the south.

Chapter XV - To the Nissa Glaciers
Stein and Ram Singh wanted to begin this new survey at the last station the latter had just used enroute to Khotan. They surmounted the first pass, Ulughat, at 10,000 feet where they had established a triangulation station in 1900. They then passed through eroded gorges and over boulder-strewn trails and reached the second pass, Yagan-Dawan. Stein mentions that only the 'stringent orders' from the Amban could enable him to obtain assistance from the suspicious mountain people. He reached Nissa on the 15th. Stein's long-held desire was to map the high Kara-lun and find the headwaters of the Yurung-kash. Stein faced continual reluctance of the local hill-people to admit to any knowledge of the mountains and passes to the south. Finally he practically forced the headman of Karanghu-tagh to admit that he had 'heard' of some locations. They continued and on the 17th reached the lead of the Nissa valley. All about the surprisingly wide valley were mountains over 20,000 feet with broad glaciers flowing into the valley. Yet even here, far into the mountains, Stein received another bag of mail, including more proof sheets of Ancient Khotan, from Khotan brought by a resourceful dak. On 19 August Stein and Ram Singh climbed another high ridge, over 1200 feet up (total elevation 15,000 feet), to establish a triangulation station and plane table. On the return to camp they had to use ropes to enable the yaks to swim across streams now swollen by melting snow on a hot afternoon.

Chapter XVI - On the Otrughul Glacier
Stein moved his camp on 19 August to the foot of the Otrughul Glacier. The photographs that illustrate this expedition into the high mountains are spectacular. Stein had to climb as far up this glacier just for the joy of doing so despite the strong reluctance of the Taghliks. It took him over 4 hours of hard climbing over steep ice sheets to reach an elevation of 16,000 feet. There he again set up his plane table and accomplished more triangulations. At this point in his narrative he notes that it was two years' later when he was able to see these same mountains from the south side (as we will see in the last several chapters). Stein then moved back down to Nissa. On 22 August he moved eastward toward Karanghu-tagh. They crossed the Pom-tagh Dawan at 11,500 feet into the Kash River valley. He had to use the goat-skin bladders he had thoughtfully prepared in India for just such an eventuality to cross the raging Kash River. Secured by a steel cable also prepared ahead of time Stein attempted to make a crossing with the makeshift raft. (fig 61). The raft proved impossible, so the next day the baggage was eventually moved across the torrent by slings attached to the steel cable.

Chapter XVII - In the Karanghu-tagh Mountains
Stein accomplished the crossing on the morning of 24 August. After seeing the baggage safely across the men followed using a rickety wooden bridge - that is by a single beam precariously fixed into crevasses on both sides. Then the party reached the mud huts of Karanghu-tagh village. Stein was determined to find the route over the Kun-lun via the Busat Valley and pass. It took much demanding before a few reluctant villagers could be almost forced to make this trip into the unknown. Then came rain. Stein continued to climb a glacier until the Taghlik porters were too terrified to continue. The rain continued, as did the terror. At one point several Taghlik's became frenzied and attacked their own head man for his efforts in Stein's behalf. On 28 August with the rain and fog preventing further surveying Stein started back down toward Karanghu-tagh. Again, the Taghlik yak drivers rebelled at staying even another night in the mountains. Stein stopped in the Pisha Valley for 4 days to complete his observations. Still the Karanghu-tagh men refused to provide further assistance or information about the mountain valleys. Stein sent the recalcitrant local Yuz-bashi under escort down to Khotan to the Amban's court. Then he sent Ram Singh on one route over mountains to Khotan and himself on 5 September followed another in order to expand the survey. Sure enough, on 8 September at Bizil he was entertained at another welcome 'Dastarkhan'.

Chapter XVIII - A Feast At Khotan
Stein returned to Akhun Beg's suburban garden, where he has stayed in 1900. But his host had departed on the holy pilgrimage to Mecca, an immensely arduous trip from Turkestan over the high mountains and across desolate plains. The usual route west was through Kashgar and across Russia to Constantinople and the return journey was then to India and across the Karakorum back to Khotan. This section is filled with Stein's detailed description of an elaborate banquet in his honor given by the Amban. The entire Chinese staff and all the Mohammedan dignitaries were in attendance. This chapter is a marvelous insight into Chinese and Turkestani culture and political life in 19th and early 20th century Central Asia.

Chapter XIX - By the Desert Edge of Khotan
Stein departed from Khotan on 15 September to revisit the Rawak ruin. He sent Ram Singh off to survey the Kun-lun foothills east of Khotan and south of Keriya. Again, Stein found he was specially honored by the Amban, Ch'e Ta-jen, who turned out the entire Chinese garrison in full ceremonial dress uniforms with musicians and officials on the road east. Stein stopped to visit with the Amban in a local reception hall. Then he was accompanied by a group of Begs to the administrative boundary at the Yurung-kash River. Stein turned northward. He was met at Jiya by Roze Akhun and ponies and supplies. He reached the Rawak stupa he following day. He found that the sand dunes had shifted eastward, opening some areas and covering others. He also found that unfortunately, after he had carefully covered the large stuccos (see Ancient Khotan) treasure hunters had uncovered and destroyed them. The stupa also was now buried more deeply in sand. Stein moved on and was led to two more small stupas. He spent 17 September surveying around Rawak and then moved on eastward to Kine-tokmak. There he found some remains of a temple. Nearby were the remains of several houses and orchards. Then Stein turned back southward toward Hanguya. They found Ak-terek, another ruined temple around which fragments of terra-cotta and pottery were found. These included seated Buddhas. By digging a trench in the sand dunes they uncovered a wall and then the floor of another temple in which they found more fragments. Many of these still held bits of gold flakes, which Stein believed were the result of Khotanese gilding of their sacred art. On the following day Stein brought more laborers to the site and succeeded in excavating a long length of the wall from which many small relievos depicting Buddhist themes were recovered. Work around the wide Ak-terek continued through 20 September. Stein then moved on to Lop town.

Chapter XX - The Shrines of Khadalik
On September 22nd Stein started for Chira where, at nightfall, he was met by Ibrahim Beg, his Daroga from 1900, who had located a suitable camp ground. With another local guide, Mullah Khwaja, Stein was led to another ruin near Domoko. On 24 September Stein found Khadalik and began excavation. There his workers soon uncovered remains of a Buddhist shrine containing bits of broken fresco and fragments of paper manuscripts written in Indian script. Soon more extensive documents in Indian Brahmi writing were revealed. Some were in Sanskrit and some in an ancient Khotanese language and by night he had over 100 more documents. With an added contingent of local laborers on the following day Stein managed to uncover the walls of the building itself to over 74 feet in length. By then the immense number of document fragments coming to light overwhelmed the meticulous Stein's efforts to mark and catalog each individually. He resorted to packing them into bags which eventually reached 230 in number. They worked under the hot sun for 12 hours a day with only a half hour break for lunch for three full days. In addition to manuscripts there were plenty of pieces of terra-cotta and frescos. But the remaining wood from the walls indicated that long ago someone had chopped them up in order to take as much precious-to-them wood away as possible. Stein found a large section of former wall under the sand. It was covered with stenciled images of the Buddha. Although he found no datable documents, Stein estimated the site was from the eighth century. On 27 September Stein shifted work to another ruin some 50 yards from the first one. There he found even more artistic panels and plaster figures. There also were Sanskrit Buddhist texts written on birch bark he estimated dated from the 4th century. Most important for Stein was discovery of a batch of Chinese coins of the T'ang era the most recent being dated 780-783 AD. Excavation continued to 1 October on more buildings containing a variety of domestic implements as well as more documents. Some of the documents were in Tibetan indicating that their invasion had reached this area.

Chapter XXI - Sites Around Domoko
While work at Khadalik continued Stein visited other potential sites within range. There was much broken debris on the ground. But he also found another Buddhist shrine. Stein found the remains of many villages all lost and abandoned but giving Stein ample evidence about erosion in the desert. He completed work at Khadalik and departed on 3 October and visited the main oasis of Domoko. There he had an example of the shifting of the cultivated areas to consider with relation to the shifting of the rivers and springs. The soil in these areas was fertile, but everything depended on obtaining water for irrigation as well as domestic use. So he studied the dams and canals and described their purposes and results fully. Near the Domoko dam Stein was shown another mound, which turned out to be a refuse dump containing rather unsavory material. Stein persisted, nevertheless, and was rewarded by a find of over 50 wooden documents in Chinese and/or Brahmi script. One conclusion Stein reached was that there has been a general decline in the water flowing north from the high mountains.

Chapter XXII - To Keriya and the Niya River
Stein's move to Keriya began on 6 October. Rather than take the well trod main road over which he had passed in 1901, he again moved through the desert to the northern side. Along the way he was again impressed by the new agricultural activity in areas reclaimed from the desert. At Keriya Stein was again treated to welcome lunch by the Amban, Ho Ta-lao-ye, who also provided much important assistance for the coming winter expedition. Despite the Amban's orders the local camel owners sought to force Stein into bad bargains for these animals, but in the end Stein was able to purchase excellent camels that served admirably for the following years. Stein also comments on the Chinese administration's political methods in appointing and frequently changing local Begs. An interesting financial transaction was the method by which Stein obtained 2550 Taels of silver cash from the Keriya Ya-men against a similar amount deposited by Mr. Macartney ( the British consul) in Kashgar to the Keriya account. This convenience saved Stein the trouble of carrying such a weight of silver for months from Kashgar and at the same time relieved the Keriya Amban from having to send dues to Kashgar.
On October 13th Stein departed again for Niya, a two day march. There Rai Ram Singh, the Surveyor, also arrived having completed his extensive survey across the mountains south of Keriya and Niya. Stein had no difficulty recruiting an ample number of laborers due to rembrance of his previous visit and Ibrahim Beg's return. Thus Stein was able to set out again into the desert on 16 October with a large work party and enough food and water for a month's work. He again stopped for a moment at Padshahim Iman Ja'far Sadik shrine on the 18th.

Chapter XXIII - At the Niya Site Ruins
On 20 October Stein left all possible extra loads with Chiang at Iman Ja'far and moved on with 50 laborers and 5 camel loads of water. Right off he was led to new ruins dating from the 3rd century. At another newly found building he quickly uncovered wooden tablets with writing in the ancient Indian Kharoshthi script. Then more ruins of houses appeared in the desert. Once at the new Niya ruin Stein sent Rai Ram Singh off to explore further to the north-east. With the new camp established Stein immediately got to work and in the very first building excavated found more and more wooden Kharoshthi documents. The variety of documents was similar to that described in his report from the first expedition. Some were in such excellent preservation that the clay seals were intact and these included Greek style representations of Eros and Hermes. There were also many domestic items such as chairs boot-last, eating tray, weaving instruments and mouse trap. In another characteristic refuse dump Stein found remarkable slips in Chinese that turned out to be notes originally accompanying gifts between members of local royalty. All in all this was a very large village whose many houses kept Stein's work team busy 10 - 11 hours a day. Almost every building revealed Kharoshthi documents of many different types plus a wide variety of domestic utensils and furniture. All around were the remains of orchards of fruit trees. Yet all around outside the former inhabited area was desolate sand dunes.

Chapter XXIV - Records of a Hidden Archive
Stein shifted his camp again to a location nearer his next excavations. The fact he had to shift his camp repeatedly around the Niya area to be near the next work sites indicates the very wide spread ruins over such a large area. This does not mean the entire area had been an urban conglomeration, but that the home of wealthy or senior individuals were spread out with gardens and orchards in between. Stein's next objective was building Nxxiv. It was another mine of significant documents and wood carvings. Stein presumed it contained someone's office with the remains of official files. From the addresses on many documents, he determined the former owner was named Cojhbo Sojaka. In one corner, under the floor, his experienced worker found a whole cache (some three dozen in all) of important Prakrit dialect language and Kharoshthi script documents on wood, unopened, with their seals and tie strings in place, a hidden archive, as Stein called it. Among the seals were images of Diodotos, Pallas Promoachos, Heracles, and Zeus. Stein reasoned that the existence of such an intact archive was evidence that the place had been abandoned hurriedly in an emergency, not as a result of a gradual loss of irrigation water or other climate change. Stein provides a detailed description of the physical structure of the documents and photographic illustrations. At the time of his writing their content had not yet been translated in London.
The former residence/office also contained various wooden utensils and pieces of furniture that had survived a thousand years of potential decay. Stein here repeats his estimate that the Niya site was a large agricultural community that flourished in the 3rd century and was abandoned when Chinese power over the Tarim Basin collapsed. But that during Chinese over all suzerainty the local administration was conducted in and Indian language and script. He repeats his comment that Hsuan-tsang wrote that local legend claimed that Khotan was largely populated by immigrants from Takshasila (Taxila) in north-west India.

Chapter XXV - Last Days at a Dead Oasis
Stein continued excavations of more nearby buildings on 25 and 26 October. Among other structural remains he found a carved ceiling beam of such beauty that he had it carefully hollowed out and cut into sections that could be carried to London. At a refuse heap he had excavated in 1901 he recovered pieces of leather armor which he had, himself, thrown away without knowing their significance. He shifted the camp 5 miles southward on 27 October to yet another section of the former community. Stein finished with the final clearing on 30 October. The nighttime temperature now reached 20 degrees below freezing. The ponies arrived and the whole crew returned to Iman Ja'far's Mazar, where Stein had to pay his Niya team and collect more supplies for the next leg. He also received another 3 bags of mail from Europe, brought by faithful Turdi. With 400 miles yet to traverse to Lop Nor Stein could not tarry at the Mazar beyond 31 October.

Chapter XXVI - To the Endere River
On 1 November Stein broke camp and separated the team again. He sent Surveyor Ram Singh back south past Niya to continue the survey of the Kun-lun eastward as far as possible toward Lop Nor. Stein wanted to move directly east past Endere to Charchan where the two routes would meet again. He crossed the Yar-tungaz River and visited once again with Abdul Karim Akhun at the latter's farm, where he also assembled a fresh team of laborers. On 4 November Stein moved on across high sand dunes to another group of ruined homes surrounded by a low rampart - 16 feet wide at base and 8 feet high. From the look of the remains Stein concluded that the village was relatively newer than Niya and also much less well built. He guessed it was an attempt to locate a agricultural community in the area that failed with the Endere River changed its course once more.

Chapter XXVII - From the Endere Ruins to Charchan
On 8 November Stein continued south east to reach the Endere stupa he had excavated in 1901. His objective was the complete excavations that were left on the previous visit for lack of time. He immediately set his team to work uncovering a buried dwelling and was rewarded by finding more Kharoshthi script on wood and leather. Stein was delighted with the discovery. From his exploration of the Endere fort in 1901 he had proven that it was occupied by a Chinese garrison and then by Tibetans in the 8th century. But he could not reconcile this with Hsuan-tsang's memoir in which the Chinese monk reported that when he passed this site in 645 AD there was nothing but remains of a then ancient and abandoned village. now Stein had his answer - Two ruined villages on the same site but hundreds of years apart in time. The Kharoshthi documents proved the area had indeed been inhabited by the Indo-Scythians from the Oxus (Tukhara) in the 3rd century., abandoned, and then reoccupied shortly after Hsuan-tsang's visit during the T'ang Chinese return to control of the Tarim in the 7th century. More Kharosththi documents in other buildings and the ruined stupa were uncovered on 9 November along with Han Dynasty copper coins. Soon Stein's workers found that the 7th century rampart of the Endere fort itself had been built over a deep layer of 3rd century remains. With his new labor team Stein was now able to clear all the remaining buried buildings within the fort. Mean while Stein found more ruins outside the fort, including a large enclosure with a stamped clay wall 30-35 feet thick and 540 feet by 340 feet in circumference and another, smaller fort with walls 8 feet thick. On 13 November Stein paid his Niya workers who delightedly set off on the 4 day trip home. Stein set out eastward on 15 November to Charchan, a 106 mile, 6 day journey. The caravan route lies along the border between the stone glacis from the mountains to the south meets the sand dunes of the desert to the north. Vegetation along the western part of this border is fed by springs that appear when mountain fed rivers flowing under the glacis appear. After the 4th day's travel the vegetation ceases and pure desert appears. On the 5th day there was a very welcome surprise. The Beg of Charchan brought a large delegation of locals including four Pathan merchants out to meet Stein, the famous 'Sahib' from India, with a full "Dastarkhan" of assorted dishes and beverages. Stein quizzed the Pathan merchants about travel conditions. They all reached Charchan by torch light.
Illustrations:

Plan of ruined fort at Endere
48 - Ruined building E iii, within Endere fort, seen from south-west
50 - Ruined stupa, Endere site, seen from south-west
51 - Ruined rampart near gateway, Endere fort seen from interior
103 - Interior of Ruined fort of Endere seen from east - In foreground walls of large dwelling, Eiii, excavated in 1901.
104 - Remains of ancient wall on erosion 'witness', Endere site
105 - Ruined tower with remains of wind-eroded dwelling in foreground - Endere site
106 - Ruin of ancient fortified post, near high tamarisk-covered sand cone - Endere site
Plate - Plan of ruined fort at Endere
Plate - plans for several ruined structures at Endere
Plate 20 - Plan for ruined fort of T'ang period at Endere - Stein first found this ruin during his first expedition in 1901 and returned during his second in 1906.
Plan and cross section of ruined stupa at Endere.
Plate - plan of ruined village - Bilel-konghan - near Endere River

Chapter XXVIII - Along the Charchan River
Stein stayed in Charchan for two days to give the men a brief rest after 5 weeks in the desert. The oasis was thriving and a good place to secure more camels and warm winter clothing. It was located at a convenient place halfway between Keriya and Lop-nor. Both Hsuan-tsang and Marco Polo mentioned Charchan. Stein had a tour through the old and new sections and wrote up his observations of recent progress as well as what historical references he found. But there was nothing worth archeological study. They departed on 23 November for Charklik on the caravan route adjacent to the Charchan Darya. Stein provides his usual detailed description of the terrain and flora. Stein simply had to cross the ice filled river to inspect a mound on the left bank and received a good soaking for his trouble and that of Ismail who was attempting to carry him across. This was nothing but a small square structure that might have been the base of a stupa. On a following day Stein was again led off the route to inspect another ruined group of structures that turned out to be a Moslem cemetery.

Chapter XXIX - At Vash-shahri and Charklik
Vash-shahri was and ancient site a day's march from Charklik. On some interest is that Charklik was in the Tao-t'ai-ship of Ak-su, the administrative center on the north side of the Taklamakan rather than that of Keriya despite the geography of the basin. And Stein's friend from 1900, P'an Ta-jen had moved north to become the Tao-t'ai. Thus it was the before reaching Yash-shahri already the local Beg rode out to meet the travelers. On the following day Stein stopped long enough to examine the ruins near the modern settlement. Among other finds were T'ang and Sung dynasty coins, indicating the settlement had been occupied at least to the Mongol period. Here Stein obtained ox-hide for camel 'shoes' and explains that camel's feet are damaged by trodding sharp stones and that the hide is then sewn right onto their live skin, a very painful process only accomplished by an expert camel-man such as Hassan Akhun. This accomplished on December 1st Stein continued on for two more days' travel over 50 miles to Charklik. Again they were met outside the town by the local Begs who set up another Dastarkhan. Stein related Charklik through history as the capital of the ancient kingdom of Na-fo-po also called Lou-lan or Shan-shan during the Han Dynasty from 77 BC, as noted in T'ang Annals. Stein found welcome rest in a large home of Tursun Bai. The following morning he made the usual, important visit to the Amban's (Liao - Ta -lao-ye) Ya-men. The Amban's commands would be essential for the raising of any local work force and collection of supplies.
At Charklik Stein planned to hire 50 local laborers and collect enough supplies for them for 5 weeks plus another month for his own group and enough camels to transport all this into the desert. Winter work alone made it possible to do all this because water could be carried in the form of ice. Recruiting the workers was extremely difficult despite the Amban's commands as the locals were understandably frightened about going out into the unknown desert and in the frigid winter no less. With his usual forethought Stein had sent word ahead to Mullah and Tokhta Akhun, two hardy Loplik hunters from Abdal known as guides for Hedin around Lop-nor. They promptly rode the 60 miles from Abdal and were able to provide some reassurance to the reluctant locals. Eventually Stein obtained his 50 workers, personally inspected to insure they were the strongest and healthiest men. He specifically included two carpenters, a blacksmith, a leather worker, and for each pair a laden donkey with supply of oats and flour and more.
Ever eager for archeology Stein found right in Charklik the remains of an ancient fortification wall with remaining ramparts as high ad 12 to 20 feet. Outside this area he found the remains of a large Buddhist stupa. Stein remained in Charklik for 3 days. On the last evening he welcomed the return of Rai Ram Singh from his successful trigonometric survey of the Kun-lun.

Chapter XXX - Start for the Lop Desert
On December 6th Stein set out for Abdal. Crossing the desert on 8 December Stein came across several more ruined Stupas and other buildings, beyond which Tokhta Akhun showed him the ruin of the Miran fort. The ruin boasted long walls of 250 feet with shorter walls on the other side. There were massive oblong towers at the corners and bastions along the walls at their centers. One tower still rose to a height of 43 feet. Along the inner side of the east wall Stein set his full force to work. They soon found buried in the sand a line of room full of rubbish. But this rubbish contained over 200 documents in Tibetan writing on paper and wood. It also contained a mix of other things such as armor, seals, silk, leather, and more. Stein guessed the fort was occupied in the 8th or 9th centuries. Tokhta Akhun showed Stein an even more ancient ruined temple as well. Stein was eager to continue work at this site, but realized that a delay would endanger failure later at Lou-lan. Thus he had to fill in the excavation to protect it and hope for an opportunity to return. With Rai Ram Singh sick with rheumatic fever Stein had to reduce his loads in order to provide a camel for the Surveyor's transport. This mean reducing the number of workers. On 10 December he had to pay off and send 15 men back to Charklik. Although Abdal was a miserable village two of its Begs showed up to welcome them far from the place. At Abdal Stein had a ferry built to take the caravan across the Tarim River. Stein established his 'depot' near Abdal and left there his silver and all the artifacts so far uncovered with Tila Bai as guard. The ponies too would return to Abdal rather than enter the most dangerous part of the desert. He had to leave Chiang-ssu-yeh and his servant there as well. All this reorganization too much time and personal effort by Stein, but he nevertheless managed to get underway again on 11 December. On reaching the Alam-khoja-kol lagoon they cut blocks of ice and filled huge wool bags brought along for the purpose. Eleven camels were designated to carry this ice, some 400 pounds for each. The caravan still had 30 donkeys to carry more ice as far as possible to create a depot and then return to Abdal. The camels each had a drink of 6 or 7 buckets full of water that hopefully would last them for several weeks. They would subsist on a supply of rape seed oil. Along the route Stein established another supply depot. The Lopnik workers were pleased to catch fish under the thin layer of ice. Another long march was accomplished on 14 December.

Chapter XXXI - Across an Eroded Dry Delta
On the 15th Stein unloaded the ice from the 30 donkeys to establish another depot and sent them back with two laborers. Then after resting 2 days the workers were to load the donkeys with more ice and supplies and bring them back to this depot. At the same time Stein would unload the camels at the archeology site - Lou-lan and have them return to this depot to bring the supply on. All this planning shows training in military logistics. He also wrote a last letter to be carried by stages estimated at 6 weeks to Kashgar and another 6 weeks on to India and then on to Europe by Easter. As usual Stein provides a vivid description of the terrain, now hard surface with mesas between ravines cut by the wind. Right on the surface they found Stone Age implements in abundance. The caravan left no track across this hard ground, so Stein was careful to appoint a team of workers under Naik Ram Singh to build visible marker stations along the route. Stein also set his plane table up to triangulate on both the Kuruk-tagh mountains to north and the Kun-lun to the south. Eventually they came onto fields strewn with medieval debris - pottery and the like, and even a bronze ring and copper coins and arrow heads. Lou-lan was near. Then the found it, on 17 December - the tower of a ruined stupa - to everyone's relief and even joy. The ground there was even more densely strewn with coins and arrow heads.
Stein ends this chapter with a grateful recognition to Sven Hedin for excellent mapping during his treks through Lou-lan and across the Taklamakan in 1900 and 1901. Stein used Hedin's map for guidance. He congratulated the explorer by noting that his own survey with plane table, theodolite and astronomical observations marked the location as being within a mile and a half in longitude and half a mile in latitude of Hedin's calculations.

Chapter XXXII - First Excavations at the Lop-nor Site.
The temperature in the morning of 18 December was zero degrees Fahrenheit. But urgent work awaited. Stein despatched 5 camels back south to bring the remaining ice and supplies from the depot. The other camels were sent no north to the springs at Altmish-bulak to rest and graze. This effort was entrusted to Tokhta Akhun, who had been there the previous year. Rai Ram Singh was sent west to reconnaissance for more ruins. Hassan Akhun was to bring some camels back on 23 December to be used to shift the camp. One is struck by Stein's prior planning and organizational skill. In the first building being excavated Stein found both Chinese and Kharoshthi documents. He also found textiles - silk and wool carpet, and small implements and assorted jewelry, and many Chinese copper coins from the Han Dynasty era. During the night Stein recorded the temperature had fallen to 13 degrees below zero, plus the frigid north-east wind continued to blast them all. He had a special small tent and thick rugs. The laborers gathered around big bonfires.

Chapter XXXIII - Survey of the Ancient Stations
On December 19th Stein started work on another ruin. There they found more Kharoshthi texts on rough wooden tablets. Again, this illustrated that the Chinese authorities allowed local administration to remain in the hands of Indian ruling families. As work progressed Stein located the site's rampart of stamped clay in places to a height of 8 feet. The distance between northern and southern walls was about a quarter mile. The eastern wall was completely gone, but of the western wall two mounds remained which had been bastions or towers. Otherwise the incessant wind from the east had eroded not only the walls but much of the ground on which they had stood. Stein's men could find no more ruins south of the enclosure, but there were several to the north.

Chapter XXXIV - Records From an Ancient Rubbish Heap
Stein loved to dig into rubbish heaps and generally found much worthwhile remains in them. There was such a heap near the center of the enclosed area. Beneath stable remains the laborers dug up over 120 Chinese records on wood and paper. Also there were many textile fragments and broken tools.
The documents show that the Lou-lan post was a fortified garrison holding Chinese troops guarding the important caravan route from Kan-su to the north side of the Tarim Basin. This route along the foothills of the T'ien-shan was the main road for both commercial and military traffic. Among the dated records so far found are 15 for years between 263 and 270 AD, the period of Emperor Wu-ti, and another one dated 330 AD. The documents include many military orders and inventories. Some documents point to a Chinese effort to support this garrison with military colonists who would both cultivate agriculture to supply food and serve tours of duty as guards or patrols. Special officers were appointed to supervise all agricultural affairs, but another officer was to supervise irrigation. There was also an inspector for the armory. While the officers were Chinese many at least of the troops were Yuch-chih - that is Indo-Scythian barbarians. This group were the ancient enemy of the Huns, who were the main danger to Chinese activities in the Tarim.

Chapter XXXV - Discovery of Art Remains
The camels with the reserves from the depot having arrived, Stein was able to shift the camp 8 miles to the west on December 23rd. While making a detailed site plan of the large stupa, Stein found a metal tape measure clearly forgotten on it by Sven Hedin. Two years later he had the pleasure of returning it to its owner at a scientific meeting. On the 24th they started digging at the new location, a Buddhist shrine. The most important 'finds' at this location were well carved wooden panels and ornamental pieces. This showed the influence of the Gandhara style of northwest India. Excellent examples of this art work are included in the illustrations. Toward evening - Christmas Eve - Stein was amazed to see his faithful Dak man - Turdi - approaching with a heavy mail bag. Muhammadju, Jasvant Singh and a Loplik hunter came with him. Stein had sent Turdi west from the Endere River on 15 November to deliver the mail to Badruddhin Khan in Khotan. That was accomplished in 12 days. Now, Stein was at least 500 miles further east but Turdi delivered 3 bags of mail from Kashgar. After a single night rest he had started east and covered the normal 30 day trip in 21 days to Abdal. From there he had to guess where Stein could be located. The Beg of Abdal had supplied the Loplik as a guide and protector. They had ridden north part way on donkeys and then moved on foot for days following the track. Obviously Stein spent the night in his tent reading the greetings from friends as far west as Oxford, England. But the following morning he as back to leading the excavation teams to more buildings. All these contained a variety of artifacts and documents. By 28 December Stein felt he had found all that was to be found. He was suffering again from malaria for which he continually took quinine. They were running low on ice. The main body went directly back over the same route to Abdal, while Stein as usual veered off onto a new track to see what he could find. He really wanted to go south east right through the Lop salt sea to trace the caravan route there to Tun-huang, but lacked sufficient animals and ice. That expedition would have to wait until his third journey into the Tarim.

Chapter XXXVI - Across the Desert to the Tarim
After the main body was started on 29 December Stein set out south westward with only Ramzan, Muhammadju, Naik Ram Singh, Ibrahim Beg, Tokhta Akhun, Turdi and 6 Charklik laborers. Again, they picked up various neolithic tools off the desert floor. Stein describes the varied terrain. With the temperature now at 13 below zero they faced night with scanty wood for fires. They continued on New Year's Day 1907 in bitter cold. The photo illustrations show the endless sand dunes. Then night temperature dropped to 48 below zero. Then they at last found the lagoons by the Tarim river

Chapter XXXVII - By the Tarim and Charchan Darya
They crossed the frozen lake now no longer in fear of running out of ice (water) and forage for the camels. On 5 January they moved up the Ilek river to Merdek ruin. This fort was covered with reeds. They found the rampart that measured about 14 feet broad at the top with its base buried in sand. It was about 132 feet in diameter. The upper level was made of bricks similar to those at Lou-lan. They found Han Dynasty coins in the enclosure. Stein then continued south on 7 January. They passed Loplik fishing huts and reached the Chinese postal station at Kurghan. On January 14th Stein witnessed a solar eclipse while near Charklik. Stein then had to return to Charklik in order to receive a reserve of silver. They reached this base by evening of 16 January.

Chapter XXXVIII - The Ruined Fort of Miran
Back at Charklik Stein was welcomed by Amban Liao Ta-lao-ye at the Yamen. He had to send Ibrahim Beg north to Kara-shahr to pick up the silver, a full month journey to and back. Ibrahim also was to telegraph from Kara-shahr to the Surveyor General in India Stein's urgent request to send out a replacement surveyor for the very sick Ram Singh. It took 9 months but the very experienced Rai Lal Singh arrived to meet Stein in Kan-su. All the duties of meeting the Amban and personally packing the artifacts for transport to Kashgar took every moment of Stein's time. On 22 January he was ready to move back east to Miran with fresh laborers and supplies. By evening of the 23rd they were reunited with Chiang-ssu-yeh at a camp by Miran. Again, the fort gave up its hidden treasures for Stein's enjoyment. These were largely Tibetan documents on wood and paper. In places the rubbish was 9 feet high. By the end of the effort Stein had obtained over 1000 Tibetan documents. Stein was appalled at the evidence that the Tibetan soldiers, ignoring all manner of dirt and worse, had simply lived on in the very same rooms in which they continued to pile refuse under their own feet. The refuse contained a wide variety of pieces of textiles and personal implement such as combs and fishing gear. But of particular interest is the abundant pieces of defensive armor. These were mostly oblong, lacquered (mostly in bright red and black) pieces of leather of various sizes. Each piece had holes through which leather strings were used to tie them together. Some pieces remained tied. Although the individual pieces were of many different sizes and had different decorations (such as rings, ellipses, bronze rivets, and hooks) , they showed uniformity in technical make and in the methods for the lacing.

Chapter XXXIX - Finds of Tibetan Records
As usual, Stein separated his description and analysis of the documents from his discussion of where and how they were found. The excavation proceeded under extreme weather conditions with a powerful, cold wind blowing from the north or north-east. The freezing temperature numbed his fingers as he cleaned and numbered each find and even froze the ink in his pen. Stein remarks on the extra discomfort this created by blowing the 'filth' into everyone's face. He particularly mentioned the case of two rooms from which the wind blew fine gravel. Stein felt he had to supervise closely to prevent loss of theft of valuable documents, an indication of his assessment of the work force. On the other hand he always 'rewarded' finders of special items on the spot with silver. By then Surveyor Ram Singh was so sick with rheumatism that he had to be evacuated to Abdal and Naik Ram Singh was also fevered. Stein's Kashmiri cook promptly took sick and his Ladaki assistant never did learn how to prepare edible European meals. All in all the work at Miran was among the most difficult.
Full translation and analysis of the Tibetan texts, which Stein assesses as among the earliest ever found, had to await the time of experts. Meanwhile Stein provided his thoughts. He believed that most of the documents were official records, office papers, reports and such addressed to the garrison commanders. Some are battle reports and calls for reinforcements. Some are about logistics issues. In total they show the significant role Tibet played in Chinese Turkestan. The fort itself was constructed and garrisoned by Tibetans during their period of ascendancy, that is from the closing decades of the 8th century to the later part of the 9th. He notes that not a single Chinese document was found. However, one document in Runic Turki was uncovered. It is a list names of Turki individuals, possibly a collective passport for a military unit. There was no evidence of the for having been used by either Uigurs or Moslems subsequent to the end of Tibetan control. For the Tibetans Miran occupied a strategic site on the direct routes from Tibet north but was not so important to powers controlling the Tarim from other directions for which Charklik was more convenient.

Chapter XL - Ancient Temples of Miran
Stein in this chapter shifts attention from the fort and its contents to the nearby Buddhist shrine which Stein dated to long before the Tibetan fort. He moved some of the labor there on 29 January with the temperature at 37 degrees below freezing and an icy wind blowing strongly. He measured the edifice at 46 by 36 feet at the base. There was one story 9 feet high and another with a base of 15 to 17 feet and remaining height of 11 feet. Two sides were very eroded, but on the others Stein uncovered from the sand at the base long sections of detailed relievos. But the remains were lost above about 4 feet from the ground level. The illustration shows a series of decorated columns with niches between that originally contained statues. Stein recognized the influence of Graeco-Buddhist artistic style. He found several huge heads that had fallen into the preserving sand. He then uncovered the bases and lower legs of six huge seated statues of the Buddha. Stein then excavated other large buildings around the shrine. He was able to pack only a few of the fragile pieces and then re-buried the rest. He turned to yet more buildings on 31 January. These included a Stupa with surrounding walls and corridors. In these Stein found frescos still freshly painted. The colored illustrations show classical angels. The corridor also contained silk banners on which was Kharoshthi script. The writing consisted of prayers to the health of individuals with Iranian or Indian names. Thus Stein dated the structure to the end of the 3rd century AD. Stein spent the night in his tent contemplating how he could remove these priceless art works.

Chapter XLI - A Dado of Angels
Stein continues - on February 1st he continued excavation of the shrine. The corridor contained windows and between these were lunettes in which the frescos appeared. A few remained on the walls and large or small pieces of others were piled on the floor. The painting was on a very thin layer of clay and the backing was stucco made of clay and straw only half an inch thick. Even touching these was apt to cause disintegration. But Stein was determined to bring them home. Moreover, the pieces had originally been parts of a large scene, so they had to be carefully numbered so this might eventually be reconstructed. Taking notes and making photographs in the numbing cold and in cramped quarters was a trying exercise. (Remember he is using a camera model circa 1906). As Stein worked, his laborers dragged trunks of Toghraks from the river to be sawn into planks for protective cases. They worked into the night with the light and warmth of big bonfires and the promises of extra pay. Stein's military engineer corporal, Naik Ram Singh, despite being sick, provided professional help in removing and crating the fragile frescos. First a layer of cotton wool and paper was pressed against the fresco surface. Then a piece of tin reinforced by iron bands was slipped down behind the stucco detaching it from the wall. Then the whole encased fresco was carefully tilted forward to rest on a well padded wood board which was then carried out of the ruin so a complete box could be constructed around it. At the same time Stein had to record in detail the exact description of each piece and its relationship to the other pieces. The results now in the British Museum are stunning. Stein went on to note the sad fate of a pigeon that was found in its nest under a heavy section of fresco that had fallen and crushed it. The fresco depicted the Buddha holding his hand in the standard 'protection' pose. To remove the dado's of the angels Stein used a steel saw to cut them loose from the wall to separate the stucco from the bricks. Making boxes for transport involved another solution to a problem, how to insure the case would be rigid and sturdy enough to prevent the mud images from disintegrating, yet also flexible enough to withstand the months' of journey by camel, yak, donkey and manpower over high mountains and across the sea. The cases were packed with sheep's wool and masses of reeds, the reeds being cut to length and carefully laid in two opposing directions. Two frescos were packed with surfaces facing each other in each box. Finally each box of two frescos was placed inside a more solid case made of Toghrak wood. The success of the whole project was due to a combination of Stein's imagination and perserverance and the lucky presence of the professional Indian and Charklik carpenters. Stein was not to witness the successful results of his hard labor until three years later at the British Museum. We can see some of the results in the fine illustrations provided.

Chapter XLII - The Frescoes of Miran
Having explained to the reader the extraordinary effort and skill involved in moving the frescos to London, Stein in this chapter expounds on the knowledge revealed by their professional examination at the museum. He remarks that his original notes and photographs would never have been adequate for this task. He writes that the remains of pictorial representations of Graeco-Buddhist art in Gandhara have been completely lost, only sculpture and architecture remains. He includes in this book some samples of this unique art work. He explains that in Plate V on the left we see Buddha standing and wearing a simple robe of dark red-brown prescribed for ascetics and saintly preachers. The halo and top knot make it clear that this is a Buddha. The right hand is raised on the "Abhaya' or protection pose. Behind and to the left of the Buddha are six Arhats or Buddhist saints, ranged in two rows and wearing robes. Their shaven heads show that they are monks. The white fan carried by one is a symbol of sovereign power. Further to the right is a colored mass that likely was a tree with red and white flowers. There is the upraised right arm of someone in the act of throwing flowers. Thus the scene is in a garden. But the topic is not so interesting as the artistic style and skill. The head of the Buddha is clearly Hellenistic with a slight Semitic touch of the face and with the Indian Buddhist convention of the top knot and pierced ears. The large eyes have an European look and the hand of the last Arhat on the right is in the classical Roman pose of a hand emerging at the neck from inside a toga. Similar classical style is common in the Graeco-Buddhist statuary of Gandhara, but this is the only example of such Western artistic style yet found. As to technique, the use of light and shade when depicting flesh comes from the West as well. Stein expounds similarly on other of the recovered frescos illustrated in the book.

Chapter XLIII - A Cycle of Festive Figures
In this chapter Stein moves on to the results of digging in another building which, when uncovered from the sand proved to be yet another stupa inside a temple. A wall in this building displayed a fresco of an angle and another of a man defending himself against a monster with body of a lion and head and wings of a bird - a classical griffin. The stupa measured 11 feet high and 12.5 feet in diameter. The circle around the stupa and within the outer wall was 7 feet wide. Thus the dome that originally covered the entire structure must have been 26.5 feet in diameter. The inside of the circular wall had been covered with frescos, of which many remained to intrigue Stein. Some he considered related to Roman era Mithras. There were many portraits with decidedly Roman or barbarian features, male and female. Stein describes each individual in detail and provides a photo illustration.

Chapter XLIV - Mural painting of Buddhist Legend
Stein continues with description of this unusual painting. There appeared a prince of Indian visage mounted and riding out of a palace. There was a chariot with four white horses. above was the image of a lovely woman. There were inscriptions in Kharoshthi script. An elephant was being led by another Indian prince. Comparing the lengthy frieze above with the dado below, Stein considered the former representative of Graeco-Buddhist style of India while the latter was the style of Roman Orient in Persia. Later learned scholars in London determined that the frieze represented the legend of the life of Prince Vessantara, a previous incarnation of the Buddha. It turned out that the script was the signature of the artist, reading "This fresco is the work of Tita who has received 3000 Bhammakas for it". Marvelous. Stein claims this is a Sanskrit rendering of the Roman, Titus. Stein claims the weather and cramped location plus his amateur photographic skill was not up to adequate recording of the full richness of the colored frescos despite his repeating the work under varied conditions of light. Nor was it possible to remove the fresco without destroying it. And he lacked further time, since the remaining distance to Tun-huang would require three weeks and the life sustaining ice was sure to melt soon. Reluctantly Stein filled in the excavations to protect them and started back to Abdal on 11 February.

Chapter XLV - The Start for Tun-huang
Stein worked hard at Abdal to pack everything being sent back to Kashgar or Khotan. He retained only 4 boxes of documents and sent books and unneeded equipment to Khotan. The antiques went to Kashgar under protection of Karim Akhun and Muhammadju. And Turdi, the dak man, went also carrying more mail. Stein assembled a month's worth of food for men and animals and sufficient transport for 13 men, 11 ponies, and 8 camels. The camels required nothing but the rape seed oil, but the ponies required lots of fodder. Stein then had to hire extra donkeys to transport fodder. But as logistical studies of the movements of Alexander the Great and other ancient campaigns have shown adding even donkeys also adds the requirement to carry their fodder and the food for their drivers. The only solution was to create relays of donkeys and send groups back to Charklik. Stein recognized that the clever owners would attempt to saddle him with the worst animals in hopes of their dying and resulting in exorbitant payments to the owners. To thwart this gambit Stein ordered up from Charklik many more beasts than he required and then selected from the lot only the best, amounting to 40 of them. Then he had to guard the selected number to insure that clever substitutions would not occur. What a mess. And the men, too, were apprehensive. Even in the bleak deserts and mountains into which Stein had taken them they were still in their 'own' Turkestan. But across this remaining desert they would be in 'heathen' China in totally unknown territory and circumstances. It was a relief when Ibrahim Beg returned from Kara-shahr with 1500 Sers of silver Chinese 'horse-shoes'. Ibrahim made the trip of 330 miles from Charklik to Kara-shar in 7 days but required twice that for the return trip because the Amban there had insisted on sending two Ya-men guards along for safety. I think I would have also, considering the fortune entrusted to Ibrahim and the dangers on the road. A supply of Chinese 'horseshoe bullion was essential since no merchants in Tun-huang would accept silver or even gold coin. When weighted carefully it turned out that the difference in official scales at Kara-shahr and Stein's 'new dominion' scale resulted in a loss of over 40 rupees in the transaction. But more trying financial transactions were yet to come. Even with all this time consuming administrative and supervisory work each day, Stein still managed to conduct linguistic study with representatives of representatives of small tribes. And he also conducted anthropometric study, recording the facial features of each ethnic group around Charklik and Abdan. On February 21 Stein was finally ready to brave the eastern desert. They made it to the well watered oasis at Donglik that first day. Another 20 miles was completed the following day. Since no water could be found for 2 days they carried ice. The following morning Stein sent off another large mail bag to Kashgar and Europe to be carried by Islam, the Dakchi from Khotan. He learned when back in London that this batch had reached England in 3.5 months. Each morning they set up the plane table and took bearings and recorded the topography. Stein also had the very same kind of cyclometer wheel as did Alexander the Great to measure daily distances. In the evening they melted ice for the men, ponies and donkeys. Even with care three donkeys had already died. On 24 February they covered even worse terrain, a salt-encrusted wasteland, a section of the dead salt sea.

Chapter XLVI - On Old Travellers' Tracks
Stein halted for 26 February at Koshe-langza to enable men and animals to rest. The location had plenty of vegetation for grazing and adequate water. The temperature was up to 23 degrees. Stein occupied himself with recording accounts and notes, after which he indulged in reading from his reference books on the history of China. (Quite a remarkable fellow). He focused on those records relating to the route to Tun-huang and the early Han Dynasty Annals about Chinese expansion into the Tarim. He recounts the record of the embassy of Chang Ch'ien (136-123 BC) from Emperor Wu-ti to the Great Yueh-chih (Indo-Scythians) on the Oxus seeking alliance against the Hsiung-nu (Huns). They knew of two routes, one to the north via Turfan to Kashgar and then over the mountains to Sogdiana; and the other along the southern desert edge through Lou-lan (or Shan-shan) and through Khotan and Yarkand. Stein's recitation continues through subsequent Han expeditions and on into the medieval accounts of Fa-hsien, Hsuan-tsang and Marco Polo. Stein found all these clearly confirmed during his expedition.

Chapter XLVII - The Last of the Dry Lop-nor
Stein resumed travel on 27 February following the lake shore. Night was spent at Panja, where again they found water. On following days Stein continued his survey using the plane table and also wrote detailed descriptions of terrain and flora. On 1 March they reached Kum-kuduk. By March 4th they reached Besh-toghrak, an oasis and good place for another rest stop.

Chapter XLVIII - A Strange Old Lake Bed
At Besh-toghrak on 5 March the daytime temperature already reached 72 degrees, but nights remained as low as 23 degrees. There Stein left the 8 weakest donkeys with an Abdal man, having 20 days' rations, to wait for the return caravan. Surveyor Ram Singh had recovered sufficiently to take theodolite observations for latitude, clinometric measurements for heights, and barometrical and aneroid measurements for elevations. The elevation of Besh-toghrak was 2620 feet above sea level. On 6 March they continued eastward. The terrain soon became a dry salt basin. And that evening they ran out of water.

Chapter XLIX - First Glimpse of an Ancient Frontier
Han Dynasty Wall
Towers - During this day, 8 March, Stein spotted the first mound, but it was too far to the north to spare time to visit. A second mound soon appeared right near the caravan track. It was a brick watch tower 15 feet square and 23 feet high. It was located on a defensible mesa with ravines all about. Stein found a few iron implements and other items made of wood and fabric. They reached the water and grazing ground of Toghrak-bulak by nightfall. Next day Stein found a third watch-tower (he numbered Tiii) over 20 feet tall and this one was next to the remains of a wall stretching east to another tower. And there were more towers to the south-west. A bit of digging turned up a few artifacts including a board with Chinese characters. Before quitting this opportune occasion Stein rode east along the wall to make reconnaissance of still more towers. After more miles he came to the large tower he numbered Tix, at least 25 feet high and 23 feet square at its base.

Chapter L - Ruins en Route to Tun-huang
As much as he longed to get to archeological work, Stein could not spare the time to stop. He had to get his small caravan into Tun-huang. So early on 9 March while the men and animals rested Stein hurried back to the towers and wall that he already suspected was the continuation of the Great Wall. When the caravan got underway they continued to follow the line of the wall until they reached a massive tower (numbered Txi). The line of wall continued east along a ridge above the Su-lo Ho basin. As they continued they passed more and more towers. Then they reached a much larger brick building fully 90 feet on a side with walls of stamped clay 15 feet thick and 30 feet high. Stein immediately climbed to the top to view the terrain all about. Moving on eastward along the line of the wall Stein at dusk came to an even larger brick building he quickly measured at 440 feet long with thick walls rising to 25 feet (Txviii). High towers rose at each corner. From this location Stein could see the line of the wall with many towers stretching both east and west.

Chapter LI - First Halt at Tun-huang
The caravan entered Tun-huang (old name - Sha-chou) on 12 March 1907. Stein found that the political and cultural situation in China proper was much different from that in the 'new dominion' - Chinese Turkestan. In China the population was of course Chinese and much less dominated by the local magistrates than in Turkestan. There the Chinese overlords could order out work parties and demand delivery of goods and camels ( with suitable payments) but nevertheless with reluctance by the owners. But no such orders could be obeyed or even attempted in Tun-huang. Even though Stein's arrival had been announced by P'an Ta-jen from Ak-su long in advance, there was no such welcoming party at Tun-huang as he had received at all the Ya-mens no matter how small in Turkestan. Stein and his party had to fend for themselves in hunting for quarters. He describes the situation and lack of official responses in detail. He soon learned that in Tun-huang the military commander was more powerful than the local civil governor. Eventually the Amban, Wang Ta-lao-ye; and the military commander, Lin Ta-jen, stirred themselves to action and became a helpful associates expressing considerable interest in Stein's archeological and historical objectives. As usual Stein was heavily engaged in administration, including paying and dismissing the donkey caravan and drivers for their return to Charklik. He sent the camels off to rest as well. He had to prepare his financial records and reports to the Indian Government Treasury and separately to the Survey of India's financial department.

Chapter LII - To the 'Caves of the Thousand Buddhas'
Stein had learned about these caves (Ch'ien-fo'tung) already from several European travelers. Stein was eager to see them for himself and managed to get time on 16 March for a first reconnaissance. He described the route and appearance of the locale. He was overwhelmed with the quantity and quality of the extensive art work which he assessed as T'ang era creations. In this chapter he describes his initial observations, recognition of Buddhas, Bodhisattvas, Dvarapalas, scenes from the life of the Buddha; and notes the Chinese derivation from the Graeco-Buddhist art of Gandahara. He knew, of course, that it would be impossible to remove and take any of these marvelous frescos or statues so settled for the opportunity to photograph them. Stein had already also heard that there was a hidden library of manuscripts somewhere at the caves. He immediately began discrete inquiries about this potential treasure. He obtained the information desired from a young monk who appeared eventually. At nightfall he returned to his tent with his mind full of ideas about how he might obtain some of the manuscripts.

Chapter LIII - A Difficult Start from Tun-huang
Stein had to turn from further reconnaissances in order to prepare another batch of correspondence for his remaining Dak man, Kurban Niaz, to take to Kashgar with the next merchant caravan on 21 March. He decided that the coming hot weather indicated that he better return immediately to excavate sites at the watch-towers and wall and defer work at the caves for later. Obtaining the required number of local workers and amount of supplies was much more difficult than in Turkestan. Conversation with Wang Ta-lao-ye at a formal banquet in Stein's honor enabled him to press his case sufficiently to energize the Amban who had not heard of the ancient wall only a few miles from his headquarters. He started north for the wall and was eventually joined by a few laborers with more camels. Along the way Stein investigated ruined, walled hamlets that had been destroyed during the great Tungan rebellion of mid 19th century. He was delighted to receive a telegram forwarded from the nearest station at An-hsi advising that more silver would be issued from the Ya-men in Su-chou.

Chapter LIV - By the Ancient Wall North of Tun-huang
On March 25th Stein reached a watch tower 28 feet square at the base and 18 feet high. He included much more detailed descriptions of the watch towers with their contents and the wall in the official report, Serindia. Continuing northward, Stein reached the channels of the Su-lo Ho near which he camped for the night. The following morning he returned to the tower found on the 25th and turned east to explore the line of towers in that direction. First visited was a tower (Txxv) 26 feet square and 20 feet high located on a clay ridge 16 feet above the surrounding terrain. On top was a parapet with brick wall 12 feet high. Stein continued east without seeing more towers. They met two reticent Tungan nomads, armed with old flintlocks, with flocks of sheep, cows, camels and ponies. The herdsmen agreed to show Stein more towers, so he followed them east to a high plateau from which he could count 10 towers in the distance. Stein made camp on the spot. On 27 March Stein left the camp at its location and went south-south east to the towers. The first one he came (Txxvi) to was 22 feet high and 20 feet square built of alternating layers of hard clay 4 inches thick and tamarisk branches. The ends of wooden posts within appeared above the top. The remains of the wall passed by north of the tower and contained a bastion as well. The next tower was a mile and an eighth to the east. Stein first searched the ground around Txxvi and found the desired rubbish heap about 4 yards away with the remains of a mud wall. Several Chinese texts on wood appeared followed by several cooper coins of Han era. The team then moved south west along the trace of the ancient wall. After 2 plus miles they reached tower Txxvii on a plateau 17 feet above the surrounding ground level. In the nearby refuse they quickly found three more wooden pieces, 9.5 by .5 inches with Chinese writing, and these contained dates. Three more wood board were then recovered. Back at camp that evening Stein and Chiang-ssu-yeh eagerly searched their Chinese dynastic tables in search of the name of a ruler that would establish the dates. Finally they found the sign of Emperor Chien Wu indicating a date of 50 AD. Stein not only had written proof of the date for the wall but also had the earliest written Chinese text yet known.

Chapter LV - Discovery of Han Records
On the following morning Stein sent a messenger to the Amban asking for more laborers. He could see by then that the extent of the work would be huge. He also recognized that with a labor force of addicted opium smokers he could not hold them for long and the more incentive pay of silver he gave to encourage there finding valuables the more they would want to rush off to use that silver to get another smoke. He had all available men working on the next tower adjacent to the last one (Txxviii). Although the tower was badly decayed it was the location of an extensive rubbish deposit. This contained a wealth of Chinese wood documents, over 70. Two of these bore dates equivalent to 75 AD. Most were military records and reports. And the poplar was the most prevalent type of wood used. In addition many other artifacts including coarse pottery were uncovered.
While Chaing and Naik Ram Singh supervised this work, Stein rushed to the next two towers about 1 and an eighth miles to the south-west. He shifted the camp and work force to these towers on 29 March to save time and energy from walking back and forth. The first of these, (Txxix) built on a small clay ridge, was over 20 feet high and made of solid layers of stamped clay. At the top there were remains of a brick parapet. Instead of stairs there were holes on one side for a ladder. The tower was on the north-west corner of an enclosure some 107 feet square. No documents were found, but instead many pieces of glazed pottery including an almost complete jar. Stein found another building some 50 yards away with thick walls of sun-dried bricks. From the debris inside Stein identified the structure as a small Buddhist shrine with the usual fragments of Graeco-Buddhist style art. He believed this shrine was built later than the watch towers. He went to examine the next tower to the west (Txxx) about 2 miles off. He found that the cart track to Hami passed between Txxx and Txxix. The only other tower visible to the west was Txxiv. On 30 March the temperature fell to 30 below zero and the bitter wind continued strong. Stein went back to Txxvii. (fig 164). There was a room, the office or residence of an officer, adjacent to the tower from which Stein recovered several dozen more Chinese records on wood of which one was dated equivalent to 35 AD. Later M. Chavannes translated one document which mentioned the Chinese unit name, Hsien-wei, and the location , Wan-sui). He also found a fire stick identical to those at Niya and Endere. That day he received a relief team of 8 laborers from Tun-huang and let the earlier group depart. Stein continued to explore the wall to the east along which he found 4 more towers at spacings from 3/4 to 1 1/8 miles. The wall was straight between towers and had a semicircular bastion as it passed north of the tower. The towers were not ON the wall and not designed for its defense. They were purely watch and signaling towers. They were all quite similar, 20 feet square at base and over 20 feet tall, constructed of stamped clay. A fifth tower, Txxxv, was eroded to a heap of clay. A half mile further the wall also disappeared into the sand dunes. Further on, however, it appeared again in sections including one of 256 yards remaining over 7 feet high. (fig 165). The wall was built of alternating layers of Tamarisk twigs and reed 6 inches thick and coarse clay and gravel 3 to 4 inches thick (fig 163) and from 8 feet at base to 7 feet at top in width. The reeds on top of the tamarisk branches provided a level base for the next layer of clay. The whole was so impregnated with salt that it had hardened like concrete. Stein commented on the apparent skill of the Chinese engineers who made do so well with such limited local resources. Seeing no further trace of the wall to the east, he turned north and found the bed of the Su-lo Ho after a 7 mile ride. The river where they camped was 50 yards wide and quite deep with ice flows.
On 1 April Stein sent his camp group back south-west while he retraced his steps along the wall to dig further at each tower. Not much was found apart from some bronze arrow heads and a few coins. On the return trip, 20 miles to the temporary camp at Shih-tsao, a blizzard raged, driving fine sand into every pore.

Chapter LVI - To the Nan-hu Oasis
Stein led his caravan from Shih-tsao oasis back to Tun-huang. He stopped in to look at an active Taoist temple with pagoda. He was met by a 'swarm' of school boys enjoying a break. He remained in the town for one day and then departed for Nan-hu. He rushed through the day trying to get another pile of letters off in the mail for Kashgar and also coping with visits from the two Ambans, Kin Ta-jen and Wang Ta-lao-ye. Their hurried visit and increased eagerness to assist was prompted by a telegraphic order received by them from the Viceroy in Lan-chou urging all possible assistance be supplied to the important visitor, Stein. Stein also spent judicious time displaying some of his 'finds' from the Han wall which the Chinese officials were eager to learn about. He describes the laborious process involved in weighing out the sums in silver as payments for each laborer and owner of animals. He records his amazement that in a country long in possession of paper money the merchants still refused to deal with coined silver or gold and insisted on payments in lumps of silver of various weights down to decimal fractions of an ounce. Moreover his scale was light in comparison with that of the local merchants resulting in a loss of 4% on each transaction. Between all this forced activity Stein also arranged to hire 12 new workers, collect a month's supply and additional camels. He even had to purchase at high prices the peculiar 'ketmans' hoes from the Muhammadans for his Chinese laborers.
He managed to depart by noon on April 5th to the southwest to investigate the presence of a ruin near Nan-hu. Right outside the new Tun-huang he found and measured the ruin of the T'ang era town at 1500 yards north to south and 650 yards east to west within the decayed walls. He remarked that the current town was 1100 yards square. He found the Tang Ho and all its irrigation canals flowing full. They camped by the river. Then they made the 30 more miles to Nun-hu on the 6th and reached it by nightfall. He found an oasis with plentiful flowing water comprised of 25-30 households. The oasis measured 2 miles broad and nearly the same in width. Stein remained in this pleasant oasis for 4 days conducting explorations of local ruins. Among these was a walled town 400 yards wide within a wall in places 18 feet high. Its age was confirmed by finds of a few Han Dynasty coins and more from the T'ang era. In addition they found a few bronze arrow-heads and other pieces. He concluded that the site had been abandoned toward the end of the T'ang Dynasty. Stein also investigated several other walled enclosures and towers. He and his team enjoyed the coming of spring at Nan-hu with temperatures now rising just above freezing.
Stein also describes the current conditions of the inhabitants and the agricultural economy of Nun-hu. A flood in 1893 had caused major changes in local life. He found many more ruins as well, but does not describe them in detail

Chapter LVII - Ancient Remains for the Future
On 11 April Stein departed Nan-hu toward the north in order to reach the western section of the Han limes. They were immediately faced with a powerful gale from the north-west and forced to camp at a hamlet named Shui-i and he describes its three decrepit buildings. The following morning Stein arose as usual at 4 AM but could not get his slack workers under way until 7 AM for the 35 miles remaining to the wall. Along the march he came across several other abandoned hamlets, in the midst of encroaching sand dunes, all deserted due to the terrible Tungan rebellion of 1866. Nan-hu itself had been sacked and the population murdered. Yet sufficient water still flowed north to provide irrigation if sufficient labor could be found. About 12 miles north of Shui-i Stein found the first watch tower still 19 feet high. After another 28 miles Stein had to call another halt for camp due to the exhausted condition of his Nan-hu laborers. On April 13 Stein got underway again, despite the fact that two of the Nan-hu hired hands had disappeared - no doubt in his opinion to get another opium smoke. Finally, after pushing through tamarisk jungle they reached watch tower Txii, found previously on the way to Tung-huang and the camp ground established at that time near a lake. The laborers built their bonfire too close to Stein's camp tents so that everything had to be moved to avoid the spreading fire.

Chapter LVIII - First Excavation Along the Western Limes
Stein got to work promptly but already was concerned that the lack of sufficient laborers and the rapidly approaching hot weather would prevent a thorough excavation of the numerous towers spread miles apart. He sent Ts-ao-Ta-loa-ye, the local officer, back to Tun-huang to recruit more workers. On April 14th Stein started excavation at tower Tviii. He noted that the footprints made in early March were still easily seen in the desert despite wind erosion. Tower Tix still stood 30 feet high on a base 20 feet square (Fig 170). On its top there had once been a room with a parapet of brick that had fallen. Access to the top was by rope or ladders. at tower Tviii Stein found evidence of the ruin of former quarters attached to the towers. There was a mound 48 feet in diameter and still 10 feet high. Despite the continual use of opium by his Chinese hirelings he managed to get this large amount of debris cleared away so that wood with Chinese writing eventually turned up. The first item eventually was determined to be a kind of pass for an individual to authorize his absence from duty. Shortly official records on wood appeared. One had a date corresponding to 8 AD. Much more by way of implements and tools was uncovered including arrow heads and pieces of cross bows and most significant a ruler marked out with decimal increments. There were pieces of silk and wool and pottery jars. One label indicated the contents of a bag had been 100 bronze arrows and the name of the military unit. Stein was delighted to see that the Chinese 'clerks ' had been just as active in their desire to keep occupied by filing reports as he had found the Indian military clerks of his day. He found the lid for the medicine case of the Hsien-ming company designed with the same type of seals used for the documents he uncovered at Niya in 1901. Later, documents using the same seal design and centuries older than those at Niya confirmed Stein's opinion that the method was originally Chinese and then adopted in Central Asia. On the tower itself Stein counted 4 or 5 layers of white plaster and several coats of white wash. The excavation of tower Tviii required two full days, hampered by a powerful gale from the east that drove the dust into everyone.

Chapter LIX - Reconnaissances Along the Ancient Wall
Leaving Chiang-ssu-yeh and Naik Ram Singh to supervise the now routing excavations, Stein rode to locate other towers and trace the route of the wall. Always thinking ahead. Stein wanted to lay out a course of excavation of each tower depending on its apparent need and to establish the locations for each camp to minimize the time to be spent in moving back and forth from camps to towers. The complex terrain of gravel-covered plateaus with wide depressions between complicated movement. But it indicated to Stein how effectively the Chinese engineers had sited each tower for maximum utility and minimum construction labor. The depressions were full of impassable marshes and bogs of reeds fed by the early season springs. The wall was designed to cover every passable route from the north but save the trouble of building where marshes and the Su-lo Ho would prevent passage. ( fig 176) In large sections the wall retained its height while in others it practically disappeared. The towers were easy to see from a distance since they had been sited on high, prominent terrain on purpose. Some were over 25 feet high as well. From the vantage of the top of a tower Stein was able to make out the brown outline of the Kuruk-tagh mountains to the north. Stein discusses at length in a nostalgic passage the manner in which the objects thrown away by the ancient garrisons were preserved under very thin layers or sand or masonry debris from the effects of erosion over 2000 years. His absolute delight at being in the solitude of the desert despite alternating sub-zero gales in winter and furnace hot sand storms in summer. He delights also in recovering the copious. 200 year, output of what he calls 'scribe-ridden' military organization. From his correspondence one can see that he abhorred the bureaucracy he felt saddled him with written impedimenta in India. He describes listening to Chiang read these records of attacks, battles, logistic matters, inspections, troop movements, details about rations and clothing and personal matters. The dates found on increasing numbers of records pointed to the 2nd century BC with the Chinese Emperor Wu-ti ordering the expansion into Central Asia and construction of the defensive wall against the Hun raiders. Other records were dated as far back as 100 BC. For Stein this corresponded to his knowledge from the Han Annals of attacks by the Hsikung-nu and Chinese defensive wall building by 110 BC. Defense of the narrow passage westward south of the Su-lo Ho and between the Kuruk-tagh and the Nan shan was critical for Chinese further operations west throughout the broad Tarim Basin. Stein ruminates on the passage of the men who occupied these lonely stations but that so much of their physical presence was preserved over 2000 years, despite the ravages of harsh weather. Stein also found in the hard crust a beaten path some 30 feet south of the wall. This was the impression left by countless patrols walking back and forth over the centuries during which the wall was occupied. Stein also was surprised to find strange mounds of reed fascines laid out in checker board fashion near some of the towers. He initially concluded that these were stacks of fascines prepared for emergency repairs. Only later did he come to recognize that more likely these were stacks of fuel prepared for use in fire signals.

Chapter LX -Discoveries by the 'Jade Gate'
Stein spent a month excavating every tower he could locate over a distance of some 100 miles. Each one produced its quota of interesting artifacts. At the start on April 17th he again shifted his camp to a location central to a number of the towers and near a fort he had passed on the way to Tun-huang in March. At tower Txii he found more wooden records in Chinese with two dated to the equivalent of 1 AD and 20 AD. That evening Chiang and Naik Ram Singh brought in a very unusual find. It was a group of folded papers on which was Western writing, which appeared to be Aramaic. What puzzled Stein the most was that the documents were paper when real paper was not created in China until 105 AD. Later analysis by experts confirmed that the documents were of an earlier type of rag paper. Analysis also confirmed that the language was an early form of Iranian dialect from Sogdiana.
Turning to the old fort near his camp - Txiv - Stein was puzzled because it was not near the wall, but was nevertheless a formidable building with thick stamped-clay walls. It was on the main road to Lop-nor. Stein found a tunnel on a mound near this tower. This led to an underground room and nearby was a platform. At both locations numerous documents were found. Some of these were dated to 48 - 45 BC and others to era of Wang-Mang -early AD. Stein concluded that this was the "Jade Gate" he sought - the control station for the wall and entry - exit point for travelers crossing the wall. It was also on a transverse wall that led south to Nun-hu. Comparison of the dating on documents retrieved from the various towers showed that those east of this transverse wall were in use into the middle 2nd century AD while towers to the west had none dated past the reign of Mang-Mang. From this Stein concluded that the wall to the west had been abandoned and the wall to the south had taken its place. Excavation continued on 21 April across the entire hill for three days even though mid-day temperatures already reached 90 degrees. Many records were recovered from refuse scattered in deposits all over the hill. Some documents were dated as early as 96-94 BC. At this point the two missing Nan-hu workers reappeared after spending days wandering in the desert. Stein concluded that without water for that interval they were saved only by the opium they smoked.
Further excavations revealed a Buddhist shrine from a much later date - with coins there dated into 713-742 AD and pieces of Buddhist scripture. Many more artifacts were recovered on this hill by the "Jade Gate" but Stein mentions them only briefly

Chapter LXI - The Great Magazine of the Limes
On 24 April Stein moved again, northward to be closer to the wall. There a decayed mound (Txva) at the location where the south-north wall would have intersected the east-west wall revealed over 160 documents dated from 65 to 137 AD. Then at other locations there were more documents back to 61 BC. Stein describes some of the more interesting of these. One was a chapter from a famous Chinese lexicographical work - the Chi-chiu-chang - that was composed between 48 and 3 BC. Another find was a batch of 11 bamboo slips from a medical note book. Many others were military orders and reports (including of AWOL) directed to commanders along the Yu-men Barrier. A piece of silk was found to provide the details of length, weight and price of a bale of silk and the location of its manufacture, Yen-ch'eng in Shan-tung Province. This revealed that the standard for Han Dynasty length was 2 feet 2 inches. There were many other bits of silk containing Chinese writing. Stein completed this excavation on 25 April. On the 26th he moved 5 miles to the east to Txviii, which he had also inspected when enroute to Tun-huang in March. This was a huge building on both the caravan route and the line of the wall. But extensive lakes to the north made it unnecessary to construct a wall there. The building was over 550 feet long and divided inside into three big halls each 139 feet by 48.5 feet The interior walls were 5.5 feet thick. The outer walls were 15 feet high above the ground but the interior walls remained to a height of 25 feet. There also was the remains of a wall at distance completely surrounding the building itself. On the north it was 15 yards from the building but on the south 106 yards from it. On the corners of this decayed wall there remained huge watch towers till over 20 feet high. Instead of normal windows the interior walls had triangular openings at ground level and at 15 feet up. After much searching Stein found some bamboo slips, with and without writing on them. Finally one date appeared - equivalent to 52 BC. Eventually by close study of the fragments Chiang concluded that one referred to a granary. Later study confirmed that this large building was a supply depot.
Stein also found two nearby towers -Txix and Txx - to the north east which provided defense for the magazine. When Rai Ram Singh rejoined the camp after completing his survey to the western end of the wall, he reported that the Su-lo Ho now flowed through depressions around Toghrak-bulak that had been dry when the caravan passed eastward in March. On 30 April Stein sent Rai Ram Singh to survey eastward while he shifted the camp westward to work on that end of the wall. He notes that he had to find a suitable camp site on a high elevation to escape the insect 'fiends' what swarmed in the jungle by the river and lakes. Lucky for them also, because the careless workers soon set fire to the underbrush that spread on three sides of their high camp site. But getting the camels down to water proved a lengthy and laborious project.

Chapter LXII - On the Western Flank of the Limes
On 1 May Stein was ready to excavate the towers along the western extension of the 'limes'. He found reaching each tower impeded by intervening marshes and bogs, plus the remnants of the wild fire his own workers had accidentally started. He determined that Emperor Wu-ti's engineers had very professionally built the wall and towers with maximum use of terrain and had extended it south-westward to rest its end on a great lake. At the end of a narrow plateau stretching west they built tower Tivb of bricks. It was still over 23 feet high and at the edge of a cliff over 120 feet above the depression. It was at an ideal location for observation to west and north. From this tower the wall continued another 1.5 miles west to another isolated clay plateau on which was tower Tiva. There the wall turned south for another mile. There was a ruined quarters next to tower Tivb in which more well preserved wooden records were found. Next Stein found the earthen outline of a wall that surrounded an area between towers Tiva and Tivb, which he concluded was likely the remains of an entrenched camp. There was yet a third tower at Tivc guarding this camp. On May 2nd Stein rode northwest outside the line of the wall and found several isolated watch towers. More isolated watch and signalling towers were found to the south-west at extended intervals with no wall between. They faced impassible terrain. Stein observed that sections of the wall that paralleled the direction of prevailing wind - that is from northeast - were well preserved but that sections that were at an angle which caused the wind to hit them sideways were nearly destroyed. Stein remarks about seeing wild camels and also the evidence of their activity and passage through the jungles and marshes. Now in the 21st century a very large area around Tun-huang is contained in an official Wild Camel Reservation. On May 4th Stein reached tower Tvib. Refuse near by yielded over 500 more Chinese records, including dates between 65 and 57 BC. Stein concluded that the tower had been abandoned close to 57 BC. The adjacent quarters had a narrow entrance with strong door secured by large beams.

Chapter LXIII - Records From an Ancient Watch-station
In this chapter Stein summarizes the evidence and his conclusions about the general conditions of service and life along this frontier fortification system. The garrisons were made up of 'companies' - called 'tui' attached to a particular section. At tower Tvib they mention a unit 'Ling-hu' (meaning, barbarian suppressing) evidently the local garrison. The records mention some 20 different 'tui' or companies, while the tower station was called at 'T'ing'. One record states the Imperial edict by which 'Ling-hu' was raised. Each company had a commander who was often named. One record gave the unit strength as 145 individuals. Besides the commander there were officers in charge of the tower and assistants. Among the duties were setting fire signals, patrolling the wall, and performing fatigue duties including wall and tower construction, preparation of fire material, making bricks, collecting water, transporting supplies, and cooking. A unit garrisoned several towers. Groups of towers were commanded from a center which in turn was under the command of the commandant at the "Jade Gate" who reported to the commander at the Tun-huang garrison. The files illustrate the daily administrative life - notifications of appointments, orders to observe discipline, checks on travelers crossing the wall, and the like. The unit commanders were called to headquarters for discussion of current issues. There were personnel reports including grants of leave and medical status. There were announcements of coming inspections by higher command. There were references to local, non-Chinese troops. There was an imperial edict to establish a military-agricultural colony. Personnel records indicate that most of the garrison soldiers were recruited (conscripted) from far away Shan-hsi and Honan and a few also from Kan-su, Ssu-ch'uan and from Tung-huang. Most of the garrisons were infantry, but there were also cavalry men and mounted messengers. Ration allotments indicate portions of .6 bushel of grain per day. Pay was determined by length of service. Two days on the border counted for 3 of normal service. NCO's received silver payments. There were watch dogs on the roster for rations. Weapons inventories list swords, cross bows, quivers and shield. Cross bows were designed in various strengths and the variety of weight of arrow heads show that the bows must have been of various strengths as well. Bows weakened by age and use were noted. The individual soldier had an allotment of 150 arrows of two types. Shields were made at a factory in Ho-nan. The soldiers were issued clothing, tunics and dresses of linen charged to the company commander. At the unit supply level were tents, reserves of bow strings, axes, hammers, and other tools. The main function of the unit was to garrison the tower and give the signal in case an enemy was sighted. Night signals were by fire and day signals by smoke. The records include time and date such signals were sent or received. Mounted messengers using relays of horses were also employed to transmit information.
Keeping people out was only one side of the activity. Keeping folks in was the other. It required special permission for anyone to venture outside the frontier barrier.
Among the texts also were fragments of literary works, calendars and personal correspondence.
By 8 May Stein let the recalcitrant local laborers go home. Having accomplished all he felt he could in the hot weather (up to 150 degrees), on 14 May Stein followed them back to Tun-huang.

Chapter LXIV - Return to the "Thousand Buddhas'
Stein was eager to get to work at the caves, but had to wait five days until the end of the pilgrimage season to avoid curious eyes. He admits to relishing the opportunity for some rest. But naturally also filled the days with work and correspondence. He used the time to visit the famous Crescent Lake and the 'sounding hill' next to it. On 21 May Stein set out for the caves. Stein remarks on his concern for the women whose crippled feet gave evidence for the ancient barbaric practice. He recognized that the caves were a quite active religious objective, having seen thousands of pilgrims recently visiting. So he would have to confine his study to photography with no thought of carrying away any frescos or statues. However, the hidden manuscripts would be another issue. Could he manage to entice their owner with parting of any clandestinely? He describes the surroundings in which he found a location for his camp near the caves. In this chapter he describes his photography and his initial efforts at seduction of the pious monk who guarded the caves.

Chapter LXV - First Opening of the Hidden Chapel
This is the account of Stein's successful psychological campaign to secure a look at the documents and what his initial observation revealed - with photo illustrations. In a word, he was overwhelmed.

Chapter LXVI - A Walled-up Library and its Treasures
The campaign continues and Stein gradually obtains more and more of the precious manuscripts to secretly take to England. In this chapter he describes his initial examination of manuscripts obtained, especially mentioning those written in different languages.

Chapter LXVII - Buddhist Pictures from the Hidden Caves
Stein notes that it was impossible at the time for him to examine the multitude of rolled manuscripts. And even at the date of writing this personal account a year and a half later at the British Museum he still has not penetrated into their richness. So he confines himself with some comments on other acquisitions, such as paintings on silk, linen or paper. He notes that the majority of the art dates from the T'ang era (7th to 9th centuries AD). These examples are from an earlier period than most Buddhist art in China or Japan. It shows influence from India and Central Asia. Excellent illustrations are included in the book. One Sein includes in color depicts the birth of Buddha and his mother, Queen Maya, and father, King Suddhodana. There are numerous scenes depicting Buddhist heaven. Many depict various Lokapalas (armed warrior protecting kings) and Bodhisatvas (would be future saints) and Dharmapalas (protecors of religion). Among the names familiar to students of Buddhist iconography are Avalokitesvara - (Chinese Kuan-yi), Manjusri on a lion, Samantabhadra on an elephant, Vajrapani with a thunderbolt, Maitreya wit a rosary, and Kshitibarbha. There also are Vaisravana (king of the north with halberd), Dhritarashtra (king of the east with bow and arrow), Virudhaka (king of the south with his club), and Virupaksha (king of the west with his sword).

Chapter LXVIII - Large Paintings and Other Art Relics
Stein identifies some of the Buddhist figures represented in the large paintings. The descriptions will be well understood by students of Buddhist iconography. Stein chose an image of Vaisravana on a cloud with his attendants and demons for the frontispiece. The demons include Mara's army with one shooting an arrow at Garuda. There is a crowded scene in which the Bodhisattva, Majusri on his lion confronts the Bodhisattva Samantabhadra on his elephant. Above them is a row of four Bodhisattvas representing Avalokitesvaraa (thevery compassionate).

Chapter LXIX - A Polyglot Temple Library
In this chapter Stein returns to description of the manuscripts with special emphasis on Buddhist doctrine. He relies on the work of many specialist scholars for identification of the large variety of documents. He writes that those "of exceptional linguistic and paleographical interest are rolls and Pothis on paper written in several Central-Asian varieties of Indian Brahmi script but in non-Indian languages." Some of these are yet 'unknown'. Two are direct translations of well known Sanskrit texts of Buddhist literature, thus hopefully enabling future deciphering of the unknown language. There are also Sogdian manuscripts and even a Manichaean document. One item is the earliest written example of Runic Turki yet known. And so it goes with further examples of remarkable finds.

Chapter LXX - Decorative Art at the 'Thousand Buddhas'
This chapter is a discussion of the carvings, frescos and statues in the many caves. Stein comments that he lacked time to study more than a few of these. There were over 500 caves of greatly varied size and content. At least he arranged for Ram Singh to make a detailed site plan by use of plane table. Stein comments that in the frescos and murals he sees more Chinese artistic influence. But the magnitude of the collection in so many caves is beyond the space available for much description. Depicted on the walls and in the statures are the same themes as in the silk banners - Buddha of course among Arhats , and Bodhisattvas and various types of attendants. Several include images of the donors or inscriptions.
One can now compare the photographic illustrations Stein made with current views including those available on the Internet courtesy of Google Maps. Stein bewailed the efforts at 'restoration' which had created horrid results. But he also noted that even during the medieval centuries that the caves were flourishing the quality of the art work seemed to decline over that extended period. His photographic efforts were made even more difficult by the lack of uniform lighting - and sufficient light was available only during certain hours. Nevertheless his illustrations in the book are excellent, even magnificent.
Stgein records his great pleasure at the unexpected arrival of 'his' Dak-man, Turdi, who had ridden over sparse mountains and through deserts on relays of ponies for a month and a half from Khotan to bring him a sack of mail. Stein completed what work the time allowed and prepared to depart on 13 June. He was doubly glad to get away because his protracted presence was generating some official inquiries from higher headquarters which the local Ambans were hard pressed to answer. The Ambans gave him a fine departure ceremony at the town temple. Of course little did they know what he was carrying away with him.

Chapter LXXI - At An-hsi the 'West-protecting'
An-hsi was the headquarters town for the department and appeared to be a convenient location for Stein to set his 'depot' of baggage and artifacts. He would have to pass through An-hsi on his way back from Kan-chou to Hami and the north side of the Tarim Basin. Meanwhile he could first ride south into the Nan-shan mountains for survey work during the summer. From Tun-huang he reached Kua-chou-kou on 16 June and An-hsi the following day. During the T'ang era of control An-hsi was the capital for Chinese government over the Tarim, but Stein found it to be a poor town within crumbling walls. The local authorities already arranged for Stein's quarters in an old temple and to store his possessions in the government Ya-men. For protection Ibrahim Beg would remain on watch. The magistrate, En-T'ai-tsen also signed officially on the inventory and each box was carefully sealed. But obtaining local transport for the trip into the mountains was more difficult. After 6 days of negotiations Stein was able to start south. In the meantime he found more Han Dynasty towers and the remains of the wall near the town. The medieval town was a ruin some miles south of the modern An-hsi. Stein found the walls of stamped clay to be 15 feet thick forming a square of 600 yards each side. The east and west walls were significantly breached by the prevailing wind. but the north and south walls remained largely intact. Inspection of the walls of modern An-hsi showed the same pattern, with the east wall being eroded. Stein made sure to visit the Telegraph office and thank the official for accurate and timely transmission of messages to and from Peking and Kashgar.
On 24 June Stein got underway by 2PM for Ch'iao-tzu. Riding south-east Stein and company found two more watch towers and a section of the Han wall. The route led through repeated ridged and narrow valleys, but gradually rose in elevation. The baggage carts had been sent ahead hours before Stein left An-hsi but by dark and near Ch'iao-tzu they were no where to be found. He reluctantly settled down for the night.

Chapter LXXII - The Ruins of 'Ch'iao-tzu
Up at usual before dawn Stein was again searching for his baggage train. Eventually it was found close by but behind a grove of trees. Then behind the trees he found the walls of Ch'iao-tzu. The partially abandoned town had high walls that stretched 530 yards east to west and 500 yards on the south wall and 620 yards on the north wall, and an inner fort as well. The old town was called So-yang-cheng. A few families remained resident in a section. Stein settled into a pleasant area for a comfortable stay. He found the town to be a typical Chinese walled enclosure. In addition to the walls there were numerous ruins of towers bastions and buildings. There were few artifacts remaining except for copper coins from the T'ang Dynasty on to the 12th century. A quarter mile east there was a large stupa that Stein dated to the Sung Dynasty. To the northwest Stein found a walled cemetery.

Chapter LXXIII - The "Valley of the Myriad Buddhas'
Stein sent off another large mail bag carried by Turdi west across the mountains to Abdal. Then he set out into the Nan-shan Mountains on 29 June. The next stop was the small walled village called T'a-shih. The next was Wang-fo-hsia, 'the Valley of the Myriad Buddhas'. Further up the narrowing valley Stein came to a defensive stone wall across from ridge to ridge. He spent the night camping at 6200 feet elevation in front of the wall of caves. Then he filled 2 days with photography and survey of this ancient Buddhist shrine. He wrote a full description in the book.

Chapter LXXIV - In the Mountains of the Westernmost Nan-shan
Stein continued into the mountains on 3 July passing through an outer range of 10,000 feet elevation into an intermediate valley with a range of very high glacier covered mountains some 30 miles away to the south. In this open space Stein found Shih-pao-cheng, a small fort. They camped near by at 7450 feet elevation and made their astronomical observations from the top of the fortress tower. They surveyed 1200 square miles during the following 6 days. Then turned south-west toward the Kashkar Pass and up a slope to 10,000 feet elevation. The entire area was surprisingly dry and barren. Turning back eastward they entered the Ta-kung-ch'a valley where they could find water. They moved up the river and then turned eastward at 12,000 feet elevation to reach a pass at 13,400 feet. Stein sent his camels off with Hassan Akhun for their summer vacation. On 11 July he rode over the barren slopes to the north-east and crossed the watershed between the great basins of Shih-pao-ch'eng and Ch'ang-ma. Nearby they sighted the main range over 20,000 feet high. They could see in the distance the Su-lo Ho where it breaks through the mountains. They continued down the river to reach 7000 feet elevation again at Ch'ang-ma. There Stein encountered again the locals' reluctance to go into high mountains. He was determined, so set out promptly on 13 July for Chia-yu-quan - the famous gate in the Great wall on the other side of the mountain range. Before nightfall they reached the Su-lo Ho and crossed by a bridge. But as often happened the local Chinese guides denied any knowledge of a route over the To-lai-shan mountains via Shui-ch'u-k'ou Pass to Chia-yu-kuan. He commented repeatedly about the reluctance, even fear, the locals had of both deserts and mountains. But he was adamant as always and determined the route. Then when the party was well up in the mountains and growing short of water the same guides revealed their real and extensive knowledge by showing the way to a hidden spring. As Stein wrote, "All their previous protestations of ignorance had been lies intended to save them from a troublesome track." They reached the shrine at Ch'ing-t'sao-an-tzu on 16 July from which they had a marvelous view of the corridor that forms the main route between Su-chou and An-hsi. Still Stein forged ahead to see the Tu-ta-fan Pass over the Nan-shan. This necessitated continuing in the mountains, across the Po-yang Ho. On July 18 Stein sent the baggage off and climbed with the Surveyor to the pass. They found the pass defended by fortifications garrisoned by Mongols. They reached the watershed in the pass at 12,380 feet elevation from which they had a marvelous view over mountains and valleys as far as the To-lai-shan Range. Stein recognized also that he was at the defining border between the arid expanse of Central Asia and the more humid climatic zone affected by the Pacific Ocean in Kan-su. They started back down the valley, but they were still at 8000 feet elevation and still had a view from high in the Nan-shan clear across the 12-15 mile width of the Kan-su to An-hsi corridor to the barren mountains on the northern side. And looking east they could make out the white "Great Wall" and its famous gate fortress at Chia-yu-kuan 20 miles away. At nightfall they reached their camp at the fortified hamlet, Ta-han-chung where Stein was met by a honor guard from the garrison drawn up in front of the gate.

Chapter LXXV - By the Gate of the 'Great Wall'
Stein records his enjoyment of the scene at Ta-han-chung at 7700 feet elevation on July 19th. Any place that reminded him of Kashmir was sure to be delightful. He enjoyed also study of the small fort with its high watch tower. The fort walls were crenallated but the quarters left a lot to be desired. The small garrison turned out promptly in dress uniform, scarlet cloaks, at his approach. From the village the day's journey down to valley floor and then east was trying in the extreme heat. Stein kept careful watch for more Han period towers and saw several possibilities in the distance. He was personally appreciative of the fact he was now on the great historic route between China proper and the 'New Dominions' and the West. And he was conscious of the role that China played and recorded in its Annals of its dealing with the nomadic peoples who also swept into Europe and India. He understood the emotional feelings of his Chinese secretary - translator who had passed through the "Gate" 17 years before when leaving his wife and child at their Hu-nan home for service in the desolate Tarim. The full extent of the massive gated fortress (Chia-yu-kuan - Barrier of the Pleasant Valley) and the "Great Wall" extending on either side to block the entire valley between the Nan-shan and Pei-shan came into view when they were 2 miles distant. Stein wondered immediately at the relationship between this medieval wall and the sections of the much earlier wall he could see to the north. Again at the main gate in the fortress Stein was met by a ceremonial military unit. Stein was even more taken by the unexpected view of the green gardens and meadows. No sooner at his camp site than Stein saw the Commander of the Gate, Shuang Ta-jen, coming in state dress for a formal reception, which the genial host insisted on providing at his Ya-men. Stein had to acquiesce despite desire for a bath. Even so Stein used the occasion to study the interior of the fortress, noting its battlements and towers with loopholes and high vaulted gates. He reveled at his arrival at 'The Western Gate of the Middle Kingdom".
On the 20th Stein returned to make a careful study of the fortress and adjoining defenses. The latter included detached defense towers and entrenchments built on commanding ground to the north. From the construction methods and signs of age Stein decided this was all a Ming Era barrier. He noted that the loopholes and crenallations were evidence of expected use of firearms. The basic designs, however, were similar to those at the Han wall. The towers were formed by a solid central core of stamped clay 34 feet square with a watch room on top. The same double line of footholes used by men to climb with aid of a rope appeared on one side. Within the 60 foot square enclosures around each tower there were the remains of quarters. The main wall was 11 feet thick at its base and was 16 feet high including the parapet. When he reached the first wall tower some 2 miles north east of the gate fortress, Stein found the junction of the older wall he had noted when approaching the fort on the previous day. He could see that this wall was much older, only 8-9 feet thick and 10-11 feet high. And the layers were much thicker, rather more similar to the wall by Tun-huang. The watch towers along this other wall also were identical to those near Tun-huang. This early wall continued for miles northward up onto the slopes of the mountain range. A short spur was built to defend a narrow ravine. Further along in the gorge was yet a second wall built of massive stamped clay and stones and still retaining a height of 11 feet. Yet this one faced in the opposite direction from the first one. Stein concluded that the Han Dynasty wall from Tun-huang east past Su-chou was built to defend the passage from the Huns to the north. But the Ming Dynasty wall that blocked access across the access route the first wall defended was designed to prevent travel and passage both ways. The first was to open commerce to the west. The second was to close commerce from the west. He referred to the account of the embassy sent by Shah Rukh in 1420 who noted the measures taken by the Chinese border guards. On the way back to camp Stein passed through colorful fields of opium poppies, the major produce of Su-chou - "fields of iniquity!" as he termed them.

Chapter LXXVI - At Su-chou and its "Spring of Wine"
On 22 July Stein moved along over the 19 miles to Su-chou. Even at the unusual early hour the Commander "Guardian of the Gate' came personally to which Stein well. And Stein was then accompanied by a ten-man cavalry detachment sent by the commanding general at Su-chou. They arrived at the high walls of the city by 3 PM. He found excellent quarters prepared in advance, which actually Stein didn't like or use. (One has to note the extensive lengths to which the Chinese officials everywhere went to provide for Stein within the limits of their capacities.) In this case he found a pleasant "spring of wine" much more to his liking. This was Stein's abode for 6 days. First order of business was to visit the Tao-t'ai who had received from Peking the financial account of 6000 Taels for Stein's use. Next, the commanding general, Ch'ai, Stein found to be an 'imposing' figure. But Stein found the magistrate, Chin T'ai-tsin, to be the most interesting conversant. A double round of official visits to and from these dignitaries individually occupied much time and there was also the joint dinner to attend. The venue for this occasion was chosen by the officials to be Stein's very old hall which they rapidly repaired to his bemusement. Then they gathered a large throng of attendants all dressed in appropriate finery. The dinner lasted from 4 PM to nightfall while Stein lost count of the number of courses offered. When Stein brought up the 'business' that had brought him to Su-chou, namely exploration to the headwaters of the Su-lo Ho, Su-cho Ho and Kan-chou Ho he was met with united opposition by the officials all of whom claimed that movement into the mountains was too dangerous and that no locals would ever be persuaded to accompany him. Just as when dealing with Indian government officialdom, Stein knew how to mount an offensive to get his way. He records his feeling of frustration at having to rely on local animal transport and owners. During this period of delay he found great comfort from visiting with Father Essems, a Belgian missionary newly arrived at Su-chou. Stein takes this occasion to remember the Jesuit Father Benedict Goez, who died in Su-chou in 1607. He also visited the local markets hoping to find suitable gifts to offer during official visits, but found only goods to 'shoddy' for the purpose. He went with Father Essems to explore the sections of the Han wall north of Su-chou. The telegraph brought news of uprisings back at Tun-huang over taxes. Finally his determination and persistence resulted in his obtaining a few ponies, under duress of their owners. Apparently corvee service for providing transport to officialdom was accepted when activities in the lowland was required, but the read dread of entering the nearby mountains was overpowering. Stein simply approached the general, magistrate and Tao-t'ai with the attitude that their personal assurances of rendering all required assistance would as a matter of course achieve the desired results. They could hardly bear the dishonor of failure. Stein describes with great amusement the dinner he held for the Chinese officials. Chinese style multiple course affairs were far beyond the experience of his simple Kashmirian cook, who had enough trouble trying to serve European fare for his employer.

Chapter LXXVII - Through the Richthoven Range to the Nan-shan.
Finally on 28 July the requisitioned ponies and mules arrived late with their drivers and Stein was able to get underway around 11 AM. Two days of travel through lovely cultivated ground and past farmers brought them past the fort at Kuei-yin-ssu (with massive clay walls 30 feet high and surrounding an enclosure 250 yards square), to the edge of the mountain foothills at Chin-fo-ssu (Shrine of the Golden Buddha). This was also defended by high walls with towers and battlements. Another group of town officials and soldiers in red awaited Stein at the gate. He again refused the offered accommodations in a temple in favor of pitching his tent in a fruit garden overlooked by another tall watch tower. He was already at 6300 feet elevation. As soon as the baggage was unloaded the hired Su-cho drivers disappeared with their animals. Stein was again stuck. Chin T'ai-tsin came out from Su-chou to round up sufficient transport. Stein spent the forced delay on correspondence to be sent eastward via Tien-tsin and the Trans Siberian Railroad to reach England in October. Eventually animals and drivers were again produced, but Stein had to reduce his baggage and even leave Naik Ram Singh and Ahmad back at Kan-chou. Stein also had to pay extra and buy 24 days of food supply for these men. And the Magistrate insisted on sending 5 soldiers along for protection. On 31 July they finally got underway into the hills. With characteristic forethought Stein comments that on passing open coal seams not yet exploited he believed that they would provide at some future time for a successful industrial development. Further on they passed another stone wall designed to close off traffic. They camped in a verdant meadow at 10,400 elevation. Some of the ponies promptly became ill from eating the poisonous grass and had to be cared for. A strong rain storm drove the pony-men to Stein's tent. The following day at 11,000 feet they passed another old watch tower and then another with a small fort in the Chio-po-chia Pass. Next they crossed the Hou-tzu pass at 11,350 feet. Then came the Chio-po-chia Pass at 12,600 feet with yet another fort with ruined walls of stone and clay. Stein describes all manner of Alpine flora he enjoyed viewing. They continued on 2 August over difficult terrain, spurs of the main range and then the Chin-tou-an-shen Pass at 13,000 feet. Stein comments that this was 'the first real pass'. Well, what were all those others then? He made clinometric measurements of the mighty range to his front and noted elevations of nearly 19,000 feet with a snow line at 15,000 feet. After completing surveying he crossed the Li-yuan-ta-fan Pass followed by the Hsi-ta-fan at 14,000 feet and through the Ma-so Ho valley. They camped at Ch-ing-shui-k'a-tzu over night and on August 3rd found the temperature down to 2 degrees below freezing. They crossed the Ch'iang-tzu-k'ou Pass at 14,600 feet. Then were now south of the Richthofen Range that rose to 16,000 feet elevation with one peak to 18,000. In the far distance Stein could see the gorge through which the Pei-ta Ho made its way to the north. And to the south there rose the To-lai-shan Mountains from which the Kan-chou River flowed east. That evening they came to a region of small gold mines and washing areas.

Chapter LXXVIII - Across the To-lai-shan Range
While the camp followers dried out and prepared for the next march, Stein climbed an outcrop to 14,000 feet to secure a panorama view. From camp they moved along the plateau to the north west before crossing the To-lai-shan. More rain dampened their evening camp and continued throughout the night and day necessitating a stay at camp. The military escort departed and was replaced by another team. (One wonders how they found Stein's party.) On the following day they were able to move west to the Pei-ta Ho and then up and over the Chu-lung-kuan pass at 13,600 feet elevation. They camped at 12,100 feet elevation in a broad valley. The only available fuel was yak dung. They crossed the To-lai-shan on 7 August. The next pass, Huo-ning-to, proved to be extremely difficult, over and between sheer cliffs and through boulder filled stream beds until they reached the water-shed at 15,000 feet. Stein again, with Ram Singh, climbed to the crest of the To-lai-shan at 15,500 feet to set up another plane table survey and make a panoramic photo. To the north the entire Richthoven Range spread its mighty peaks. To the south the Alexander III Range displayed its glacier-clad mountains. This range they had to cross to reach the Su-lo Ho. The next valley to the south was that of the Pei-ta Ho. At once the pony-men tried to defect. Stein managed to block them and force the group toward the Alexander III Range. But it took the full force of the armed military escort, commanded by NCO Ching Ta-lao-ye, to drag the pony-men across the several streams and get the column moving. Threatened with dire punishment back at the Su-chou Ya-men they were forced back into line. August 9th Stein decreed for a day of rest while he searched for a route over the Alexander III. He found a pass at 15,200 feet elevation. From there the Suess Range to the south became visible. On the Alexander III Range itself Stein noted the line of vegetation on its north face was at 13,800 feet but on the south face was at 14,400 feet, due to more moisture to the south. After a difficult descent they reached the Su-lo Ho at 12,400 feet. They camped by the river. That night Stein slept peacefully but the Chinese pony-men were beset by 'dragons'.

From the Su-lo Ho Sources to Kan-chou
Stein turned east to follow the river in its wide valley toward its source. On August 12th they camped at 13,200 feet. On the 13th they crossed the Su-lo Ho basin over low spurs and halted to camp at 13,500 feet. Crossing the watershed between Su-lo Ho and Pei-ta Ho at 14,600 feet Stein could see the entire basin of the Su-lo Ho behind him. They missed when trying to shoot a wild yak for dinner. Descending to the Pei-ta Ho they encountered strong rain again. The pony-men and military escort were running short of rations. They had been ordered to bring flour for 24 days but had exhausted it already at 15. Stein made emergency rations out of extra barley brought for the ponies. The Chinese refused to eat this until shown the example of Chiang and Stein himself. Stein remarked on the contrast among the Chinese between their constant fear from many real and imaginary risks and their helplessness at taking any measures to avoid or alleviate them. Snow fall prevented further movement until August 15th. Continuing up the Pei-ta Ho they headed for the Ta-t'ung River, a tributary of the Huang Ho. At 13,600 feet Stein was delighted to reach the watershed between Central Asia and the Pacific. But elation soon turned to misery as another downpour soaked everyone. Misery turned into despair for the hapless Chinese. It was by now urgent that Stein rescue them. After crossing one more pass at 14,600 by a glacier Stein could see the valley of the Kan-chou River in the distance. They made camp at 12,500 feet. On 18 August they moved down the Kan-chou River. The military escort men managed to shoot a wild ass. But the Kan-chou would breach the mountains through a narrow and impassable gorge. So Stein had to find another pass, which he did at 14,000 feet. Again moving down the Khazan-gol Valley they were much relieved to find a Mongol camp and then another. The Mongols guided Stein's party over several more ridges and through passes and past more Mongol camps, where they were expected. By night of the 23rd they reached the Li-yuan Valley and the following day came to Li-yuan oasis. Another rain-storm caused a further day's delay. Again the local commander had been alerted to expect them. On 26 August they moved toward Kan-chou. There a large group of officials were drawn up as an official greeting party. There also Naik Ram Singh had prepared much appreciated quarters. And there Stein had his heavy mail bag delivered after a 60 day ride from Kashgar. He reflected on the work accomplished - 24,000 square miles of survey of mountainous terrain between An-hsi and Kan-chou.

Chapter LXXX - From Kan-chou to the Tien-shan
On awakening Stein found that what the careful Naik had thought was a temple turned out to be a 'coffin club', filled with coffins awaiting the families for burial. Naturally all promptly fled for more desirable spaces. Stein omits discussion of events during his 6-day stay. He also writes that his narrative from this point will be more brief. He started back west on September 3rd. Another winter campaign was in the offing. He spent the first day on the road examining ruined Hei-shui-kou 10 miles to the north-west. He sent Rai Ram Singh along the foothills of the Richthofen Range on the south side of the corridor while himself keeping to the main road to Su-chou. He studied the sections of the wall that appeared in places north east of the high road. He reached Su-chou on September 13th. There he also found Wang Ta-lao-ye, who had been brought back from Tun-huang after the riot. He bid good bye and gave many thanks to the officials for their significant assistance. On the road they passed a group of the rioters - complainers - who likely would not see Tun-huang again. Later he learned that Wang had been found not guilty and had been given a different office. They departed Su-chou on 16 September, passed through Chia-yu-kuan, and in days reached Yu-men-hsien where they found remains of the Han wall again. There was a line of towers north of the town. Two days out of Yu-men he again found towers and sections of the wall at that point north of the Su-lo Ho and the walled village, Bulunjir. After riding another 12 miles Stein found the location where the Chinese engineers had taken advantage of a ridge line to transfer the line of the wall and towers from the north - right bank of the Su-lo Ho to the south - left bank to continue on west to An-hsi.
At An-hsi Stein found the very experienced Rai Lal Singh waiting to replace Rai Ram Singh, who had been ill since the previous winter at Lop-nor. The Surveyor General of India had quickly despatch Rai Lal Singh 9 months earlier upon Stein's request. Rai Ram Singh had recovered sufficiently to render excellent surveying work in the Nan-shan but clearly would not be able to continue through another winter. But Rai Ram Singh was to accomplish one more important surveying task by taking a route back to Khotan through the Altin-tagh from Tun-huang to Charklik rather than following the previously surveyed desert road. They stayed at An-hsi for 12 days while the two surveyors produced ink copies of the survey work already accomplished while Stein wrote his reports and requests for further opportunities. On October 3rd Rai Ram Singh left for the 3 month journey to India via Khoan, Sarikol and Gilgit. Rai Lal Singh was completing the survey of the Han wall and towers east of An-hsi. And Chiang-ssu-yeh returned from a secret mission to Tun-huang by which he obtained four more camel loads ( over 200 bundles) of precious manuscripts from the 'thousand caves'. They were brought to An-hsi by night travel led by Ibrahim Beg and Hassan Akhun.
Stein departed An-hsi on 8 October for the 11-day ride to Hami along the Pei-shan hills. (Maps and Google aerials show that this route is now the major highway,)

Chapter LXXXI - At the Hami Oasis
In October the weather was already turning colder. But Hami (also called Kumul) at 43 degrees latitude, sheltered by the T'ien-shan was still in autumn. As usual the word sent by Mr. Macartney and by P'an Ta-jen had alerted all the local officials. So they promptly descended on Stein with welcoming presents. Then Stein had to reciprocate, beginning with a visit to the Muhammadan chief or Wang, Mahsud Shah. Stein found his fort or palace courtyard full of armed retainers. Stein was greatly impressed by this worthy. He then proceeded to the walled town to visit the several Ya-mens as protocol required. He admired the well -kept walls. At the Ya-men of the garrison commander, Colonel Yang, who turned out to be an old friend of Chiang's. The civil Amban also was impressive as a scholar. Stein received a surprise visitor in the person of Cecil Clementi, Assistant Colonial Secretary for Hongkong, who was passing through from Russia and Kashgar to Hong Kong. Once again it turned out they had mutual friends in India and England. Stein was greatly relieved to be back in an area where financial transactions could be conducted in real coin and even paper banker's notes rather than the chopped up bits of silver required eastward.
On 24 October he undertook a quick survey northward to Karlik-tagh mountains, which reach 13,000 feet in places. He spent the night at Toruk, where he was greeted as expected late at night, in the head man's snug home. The following morning he was off to Ara-tam. There he was delighted to find luxurious apple, apricot and peach orchards full of autumn color. Ara-tam was the Wang's country estate but Stein's interest was in the rear of the garden beyond, where he found caves with Buddhist stucco images including the remains of a huge Buddha. Stein dated the shrine to the 9-12 centuries. Next day he returned to the cliff side caves. Two days excavations turned up very little as the site was badly eroded and damaged by fire. Stein returned to Hami and greeted Lal Singh back from surveying the mountains to the north. On November 2nd they started west from Hami to Turfan. At Togucha they found more Buddhist temples.

Chapter LXXXII - Glimpses of Turfan Ruins
After 6 more day's of riding Stein reached Pichan oasis on 10 November. Stein notes that Russians and Germans had excavated many sites in the Turfan area. Rather than try to find interesting material where they had already worked, he proposed to seek other locations. He decided to start exploration around Chong-Hissar in the south-eastern section of the basin. He sent the baggage on to Turfan town while he and the Surveyor went to Lukchun. He was also interested in the "Karez' irrigation method -that is by digging out tunnels instead of canals on the surface. The result would be excellent irrigation and thus greatly increased land values. Chong-Hussar was a fortified hamlet about 140 yards from east to west and 100 yards north to south. Within the rampart were a multitude of small chambers. The walls were up to 7 feet thick. Stein quickly found Uigur texts. During the three-day stay Stein sent Lal Singh south east to survey the area toward Singer while he surveyed Kichik Hassar (Little Castle). There he found Buddhist art work, frescos, statues, paintings on linen and wood. There were documents in Uigur, Chinese and Tibetan. On November 18th Lal Singh went to the main range, Chiang and Ibrahim Beg to Turfan and Stein rode about the district to look quickly at the many archeological sites. Among the locations were Toyuk (many caves) and Kara-khoja (Uigur capital with large temples, monasteries and a fortified palace), and Bezeklik (more caves with art work). After a brief visit Stein moved camp to Turfan town. There he found the bazaars very active with trade in European goods coming south from the Trans Siberian Railroad in exchange for cotton and produce sent north. He spent a week at Turfan but used the time to excavate at Yar-khoto, the former capital during T'ang era. The town was situated on a narrow plateau surrounded by deep ravines. The ruins were so densely built that it was difficult to get between them. The site was to large and potential labor to extensive for Stein to undertake systematic excavation.

Chapter LXXXIII - Kara-shahr and its Old Sites
Come December 1st and Stein was anxious to get to work in the desert. He sent Lal Singh to survey around Singer. He rode for 8 days to Kara-shahr. In this area he found Mongol and Tungan families settled. The town was located at a strategic point where the terrain was relatively low between the Kuruk-tagh and T'ien-shan. The local magistrate had been alerted by P'an Ta'jen of Stein's visit. He helped arrange for labor to excavate at 'Ming-oi' a few miles to the south-west. The site had escaped the efforts of the German explorers. Stein made camp there on 11 December. The extensive ruin containing over 100 shrines prompted Stein to immediately hire more workers from Korla. The ruined buildings had suffered from rain and snow, but even more from a great fire. Stein found no coins dated later than the 8th century, which prompted him to guess that the area had been burned during the initial Moslem conquest. Even so he recovered many interesting relics. He collected hundreds of detached heads and torsos. Some he attributed to influence of classical style from Gandhara - northwest India or the Kabul valley. Besides sculpture, Stein found many frescos still undamaged and vivid. These included scenes from the life of Buddha. Naik Ram Singh was able to cut them off from the bricks behind them.
There also was a large watch tower on a ridge half a mile north of the main ruin. Already the temperature dropped to 42 degrees below zero.

Chapter LXXXIV - From Khora to Kuchar
By Christmas Stein considered his effort at Ming-oi was completed so he could move on to Khora. Local Mongols informed him about unseen Buddhist stupas on a rock outcropping. As Lal Singh continued to expand the survey area, Stein found articles at this obscure Buddhist shrine. Then they passed through the defile back southward. They remained working at Korla until January 1st 1908. Stein was intrigued by rumors of a ghost city off in the desert. Always eager to find ruins or disprove rumors, Stein started on New Year's Day with the locals who claimed to have seen this mirage south west across the frozen Konche Darya. The result was some added knowledge of the area between the Inchike and Charchak Rivers but no 'ghost' town. Stein comments on the propensity of desert dwellers to imagine all sorts of 'old towns' and 'hidden treasure'. On 12 January Stein and Lal Singh again separated so the Surveyor could map the river to Shahyar while Stein went north-west to Bugur and Kuchar. At the latter town Stein observed the results of extensive exploration by Japanese, German, French and Russian teams. Kuchar lies next the T'ien-shan where two rivers flow south. In preparation for new exploration on the south side of the Tarim, Stein sent his convoy of heavy baggage south to Khotan via the route up the Khotan River.
He planned a more direct but dangerous route to strike due south from the Tarim River and find the location where the Keriya River ended in the midst of the desert sands. He admits to the personal satisfaction to be gained by accomplishing such a daring exploit. On 25 January Stein's caravan of 24 heavily laden camels got underway for Khotan with Chiang-ssu-yeh and Tila Bai in command. With his own smaller team he was preparing to conquer the desert. In addition to his own 7 camels he hired 8 more, all to be loaded with bags of ice and fodder and survey equipment. On 13 January he reached Shahyar, the edge of the desert. As always he was greeted outside the town by the delegation of local Begs and other important people. But, contrary to expectations, no one is Shahyar would admit to knowing anything about a route across the desert to the Keriya River.

Chapter LXXXV - In the 'Sea of Sand'
A sea of sand indeed it was, with steep sand dunes rising like storm generated water mountains. Stein admits that if he had known in Kuchar that there would be no guides in Shahyar he might have forgone the dangerous adventure. He notes that (his unacknowledged rival) Sven Hedin had crossed the desert but from the south, going down the Keriya to its end and then simply due north with assurance of hitting the Tarim River some place. But to go south from the Tarim one would have to navigate exactly over at least 150 miles to find the delta of the Keriya in the midst of unending high dunes. One could miss the river's end by only a mile right or left and wander on without seeing it. Soon the water would be exhausted.
Stein decided to risk this route because taking another would cause too much delay. He made extra precautions that each individual would carry sufficient food for 1.5 months. The additional eight laborers hired at Shahyar were ordered to do likewise. Everyone was clothed for the winter weather. Stein hired more camels as well for a total of 15 to carry ice and food for 20 men. Stein also took 4 ponies for later use, but he would walk along with everyone while in the desert.
Anyone familiar with the process by which a quota of individuals is filled by their bosses will recognize Stein's problem when the village head men tried to assign the worst of their villagers to his convoy. Stein was used to countering this ploy. But even the strongest men he selected were already psychologically more than a little reluctant to face the desert. The Mandarin sweetened the job by promising the selectees, who were on their knees praying for exemption, that he would exempt their families from corvees for a year.
The expedition began with crossing of the final 13 miles of cultivation and steppe. Then they crossed the frozen, wide Tarim River bed and camped at the last human outpost of huts occupied by shepherds. Stein realized it would be 300 miles to the next dwelling. January 29th found them moving through the Toghrank jungle and reeds border such a large river at the Tarim to a temporary shepherd camp. The following day they continued through jungle. Here also they loaded the precious ice from deep pools fed from underground drainage. Eight heavy bags were loaded, one per camel. And the camels were given their last drink - 6 to 8 buckets full for each. From that point on the navigation would be only by compass, heading due south to the Keriya. On the very first morning in the desert the hired camels escaped their drivers before dawn. Hassan Akhun and the drivers spent 3 hours rounding them up. Already the sand dunes were over 40-50 feet high. They made only 10 miles that day. They managed to dig a well for water for the ponies. On 2 February they marched on over dunes now up to 80 feet high. At the evening camp they dug another well and found some dead brush for a fire. Next day they crossed an unusual line of poplars and tamarisks. That evening they again found good water at only 3.5 feet below the ground level. The march on February 4th of 14 miles was relatively easy but no water was found by digging. Nor was water found the next day, although they passed many tracks of wild camels. Temperatures were now down to 28 degrees below zero. But there were still clumps of dead trees available for fire wood. The water supply was now down to three bags and two iron tanks of ice. The following day they saw a live tree by a depression and eagerly got to work digging. At a very deep level below clay they found enough water to refresh the ponies and fill a few small bags. The next sand ridge was over 300 feet high but it afforded a view far to the south. Stein was following Sven Hedin's map in reverse. That evening spirits soared when Hassan found an depression into which a well only 4 feet deep produced fresh water. Temperature was now minus 37.

Chapter LXXXVI - In a Dead Delta
On 7 February Stein shifted the direction to west by south along a dry river bed. But it was leading them astray. They were in the dry delta of the river, but where was the still active river? Continuing with increasing concern, Stein tried following another dry river bed. Lal Singh used the plane table for astronomical observation of their latitude. In it during the day he guessed at a possible location for a well and was proved right at 14 feet deep when water appeared, but at 40 feet below the adjacent dune. February 9th was relatively easy as them moved on the hard surface of the river bed. There were clearly many wild camels around them, so water must be close. But it was not and no well proved successful. Everyone, especially the Shahyar men, was again very worried on 10 February as the party moved on through, again, high sand dunes. While searching for possible water, the Naik found the footprints of two men. On February 11th Stein got the caravan underway to follow the track of the two hunters. With the heavy caravan following Stein took a few men ahead to follow the track of the men and the camels they were clearly hunting. He describes the effort in detail. But by nightfall they lost the track and had to settle for camp without finding water.

Chapter LXXXVII - Salt Marsh or Ice?
With only 3 unfilled bags and the two iron tanks of ice remaining, (six days worth at most) Stein decided to halt the caravan while he and Lal Singh searched for a day or more, one east and one west, in hopes of finding the river. The ponies were by now showing signs of coming death. But in the morning the Shahyar men were panicing and demanding to go back. Stein knew that there was no chance for that. Stein had to point out that if they left the remaining ice would be even more for those remaining but any who left would perish in the desert. Stein made sure that Nail Ram Singh and Jasvant Singh were well armed as he left them in charge of the camp and guards of the ice supply. He and Lal Singh went ahead with two men to carry the plane table and theodolite and use the cyclometer. He climbed a 300 foot dune to set up a survey station. From the top he saw thin lines of white to the south east, through his binoculars. Elated, he ordered the caravan to move toward the hoped for frozen river. Joy abounded when the men reached what was indeed a strongly flowing stream of fresh water under a sheet of ice some 200 yards wide. The camels had been without water for 13 days, the ponies for 3. But there was scant brush for grazing. Stein remarks that his treat was the first bath in days.

Chapter LXXXVIII - By the New Keriya River Bed
The trying 16-day journey called for a day of rest on 13 February. They had not seen a living creature since leaving the Tarim. Stein calculated that the river had shifted its course again since Sven Hedin had passed by and mapped it. Astronomical observation indicated they were still miles north of the ruin at Kara-dong. On February 14th Stein started the caravan south, up the river, in search of food for the camels and ponies. The newness of this river course had not yet given rise to vegetation along the banks. But there were many tracks of hares, foxes, boars and other animals, plus birds in plenty. On 16 February while continuing south up the river they finally came across living reeds for the ponies. Then they found sheep and in the distance a man. Being frightened at the apparition of a mob appearing out of the desert he tried to escape but was soon run down. Stein immediately asked where they were and from the answer recognized the location in relation to his exploration in 1901. The former river bed lay to the east.

Chapter LXXXIX - More Taklamakan Ruins
Stein moved the short distance and began work again at Kara-dong on 19 February. His exploration in 1901 had been cut short by sand storms. (See Ancient Khotan Chapter XIII) It was his expectation of being able to work at Kara-dong directly on finding the Keriya River that had prompted him to attempt the desert crossing directly. And he had made arrangements while in Kuchar for his experienced team from Khotan to come north to meet him, and they brought him a much appreciated mail bag as well. On 25 February they arrived at the former camp site (Kochkar-oghil). This enabled Stein to discharge the ineffective Shahyar men with ample extra pay to make their way back home via Khotan. Stein was always generous in paying men extra above the originally agreed amount when the conditions had been extra difficult or their work extra good. He led the revived team over to a site north of Domoko, where he recruited more labor, (See Ancient Khotan) that the 'searchers' had discovered 8 miles north of Khadalik. At this Buddhist shrine, abandoned in the late 8th century, Stein unearthed more Sanskrit manuscripts, painted panels, and tablets in ancient Khotanese. Two frescos are reproduced in the book. They depict the Indian goddess of smallpox, Hariti, the Graeco-Buddhist dread goddess-destroyer of children. Stein also found that contemporary irrigation was being extended so that the area he had explored around Khadalik in 1901 was now under cultivation. He saw the same new activity also at Domoko and Gulakhma. And irrigation with expanded cultivation was also bringing population and economic activity in a bazaar. Stein continued during March to expand his excavations to all the other nearby sites such as Ulug-Mazar and Kara-yantak. In early April he stopped at Khotan where he was again greeted by old friends and found his convoy from Kuchar had arrived safely. Akhun Beg had returned from his one and a half year Haji to Mecca via Samarkand, Istambul, Bombay, Kashmir and the Kara-korum Pass.
On 5 April Stein again set out for the desert. At nearby Kara-sai (on the Kara-kash River) he recovered more Buddhist art works in Plaster of Paris. He stayed over night with Islam Beg, his foreman from 1900, in a new house suitable to the official title of Mirab - canal boss - and then a promotion to Beg, both earned as a result of Stein's recommendations. In this area also Stein was pleased to see expanding irrigation and cultivation. Crossing the desert between, Stein searched around the Yurung-kash and found another Buddhist temple under high sand dunes. The laborious digging was rewarded by finds of decorated walls with large frescos and stucco relievos. Moisture from the river destroyed any wood that had been there. It also made the frescos so weak that they collapsed. Stein could only photograph them.
Finishing this diversion, Stein set out back north to cross the Taklamakan desert again, down the Khotan River to Ak-su. From Tawakkel north his convoy was guided by Kasim Akhun, a hunter who had worked with Stein in 1901. Stein was headed first for a rumored fort by the river. They reached the red cliff on which Mazar-tagh stood on 16 April. The fortress was 200 feet above the river bed on a long ridge. The fort was well preserved. Stein made extensive site plans and as usual dug into the refuse dumps. He found hundreds of Tibetan records on wood and paper amidst the 'unspeakable' other debris. He also recovered Chinese, Brahmi and Uigur documents. He dated Mazar-tagh to the Tibetan invasion of the 8th and 9th centuries. Chinese documents included a daily record of expenses compiled by Buddhist monks.

Chapter XC - From Ak-su to Yarkand
Stein left Mazar-tagh on 20 April proceeding down the Khotan River northward for 8 days to reach the confluence of the Khotan with the Tarim River. The weather was already becoming very hot. Kasim led the group through the maze of lakes and jungles formed by the Tarim. Remarkably on one night they were disturbed by a visit from a tiger whose track they found had been following them for miles. They crossed the Tarim by ferry on April 27. It took three more days to reach Ak-su. Stein noted that there was plenty of water but relatively little use made of it for agriculture. He laid this difference to the nature of the local population in comparison with the inhabitants of Khotan and Keriya - indolence versus energy. The following day Stein met his old friend from 1901 - the Tao-t'ai, P'an Ta-jen. This was the principle purpose for the long trek back north. Stein spent the next five days in visits with P'an Ta-jen. He showed examples of his relics and an advance copy of Ancient Khotan.
Stein also sent Rai Lal Singh on an independent survey across the lower T'ien-shan from Ak-su clear back to Kashgar. (An account of Lal Singh's career would no doubt be very interesting as well. Stein mentions that it included 24 years of work from Yemen to Mongolia. Another purpose for the visit was to submit a strong recommendation for Chiang-ssu-yeh, through P'an Ta-jen, to the Governor-General (Fu-t'ai) at Urumchi to secure a good official position for Chiang.
Stein then moved north into the mountains by the Taushkan Darya.
He reached Uch-Turfan on 8 May. There he studied the local Kirghiz. From there he moved south across the hills to Kelpin oasis. On 11 May a ride of 35 miles brought him to Kirghiz Mangush Beg's camp ground. There he learned about a strange stone image located high in the mountains that was a magnet for local worship. Never leaving such interesting places aside, Stein enticed Mangush Beg to lead him there through the Saghiz-kan Pass at 9000 feet elevation. Baggage was switched from the camels to Kirghiz ponies. On 13 May they found the image at Chal-koide enclosed by a rough stone wall. The image was a simple slab about 3 feet high on which was a carved flat relievo of a man with sword, and next to it a carved stone shaped like a stupa. Stein considered that it must date back to the early Buddhist era. The image was imagined as that of the wife of some ancient hero called Kaz-ata, whose image is imagined to appear on a mountain peak. What intrigued him most was that this ancient shrine of 'idolaters' had long since been taken over by pious Moslems who for centuries had been leaving votive offerings despite the objections from low-land Mullahs. He considered it another example of the phenomena of ancient cult shrines being converted by later entirely different worshipers as an object for veneration. During his rides Stein made careful note of the condition of vegetation and noted the increasing dryness of the climate. He learned that former springs were now dry and that life for the nomad herders depends on exact knowledge of the location of surviving springs forming natural cisterns. He continued down to the Kelpin oasis through gorges in rock walls made by ancient but now dry streams. In Kelpin he found a group of hamlets surviving due to local springs. There he found conditions the opposite from along the southern border of the desert. The available water supply was too limited to sustain the growing population, whereas around Keriya a shortage of labor prevented full exploitation of the available water. Stein found enough ancient ruins to keep him busy for several days despite rising temperature. But erosion had already destroyed most remains around an old fort apart from coins and pottery shards lying on the hard clay. Before the loss of water the place had been very much inhabited from Han era to the 8th century. At this point Stein returned to the ancient Chinese main route to Kashgar and its topical line of ruined watch-towers. Finding also some modern coins of 18th and 19th centuries intrigued Stein until he learned from local head men that there had been a battle in 1876 in which rebels were defeated and driven into the desert to perish. Subsequently the locals had collected all the valuables they could from the corpses. He found more Buddhist ruins at Tumshuk. He could see that irrigation from the Kashgar River had in former times extended much further via canals than was currently the case.
He managed rides of 30-40 miles each day to Maral-bashi and other ruins, but the intense heat precluded further archeological investigations.
Reluctantly he set out again for Khotan. He also reluctantly refrained from making a 5-day side trip west to Kashgar. Instead he followed the direct route of 130 miles to Yarkand and completed it in 5 days all the while employing his plane table to continue topographic surveying in the blazing heat. At Yarkand he had to sell his original group of camels that had braved the deserts for 2 years. The remaining trip back over the Kara-korum would need yaks and ponies. Interesting, these camels had already become so famous in the bazaars of the Terim that potential buyers flocked to see them. Thanks to the excellent professional care of Hassan Akhun the camels were in excellent condition. An Afghan trader paid the highest bid at 51 Taels per beast, thus netting for the Indian Government a 70% profit after the two years of usage. Stein saw the camels off with a personal treat of a large loaf of bread.

Chapter XCI - Preparations a Khotan
By 9 June Stein had returned from Yarkand to Khotan. Along the way, at Pialma, he conferred with Satip-aldi Beg and arranged for the transport he would need in September across the Kara-korum. At Khotan he was back to his favorite camp site at Nar-bagh. ( - ) Stein became very busy organizing the mass of artifacts he had sent from along the southern edge of the Takla in 1906-07 with the loads being delivered from his more recent explorations north of it. He comments that the acquisitions of tin plate to fashion his crates took the supply of the whole of Turkestan. The laborious task of sorting and packing took 6 full weeks. Teams of carpenters were busy cutting tree trunks and fashioning boards to turn into cases. Stein made sure to do all the actual wrapping and packing of each artifact himself. He invented a method using glue and cotton to strengthen the backs of the fragile frescos from Miran. He had to prepare them for the coming 8000 mile journey to London.
A tragedy occurred during this period. In March Stein sent his faithful engineer, Naik Ram Singh, back to Miran to photograph the art work that could not be moved. There the healthy Sikh suddenly developed glaucoma and became blind. Despite this he resolutely continued to accomplish his mission at Miran. He insisted also on doing all his own cooking in conformity with caste eating rules. Finally, Ibrahim Beg brought him back to Khotan. Stein hired a Hindu cook and sent the two to Yarkand for medical attention. There the Swedish Medical Mission director, Rev. Mr. Raquette, diagnosed the disease. Stein then sent his engineer over the Kara-korum to India via Ladak with assistants to insure his care. The eventual outcome was that Ram Singh was medically retired with an exceptionally large pension, but died in 1909 after witch the Indian Government authorized a continued pension for the widow and son.
As Stein labored with the packing, Chiang-ssu-yeh continued with cataloguing as much of the Chinese documents as time permitted. Meanwhile, Stein also was busy with extensive preparation for another exploit. He was determined to find the head waters of the Yurung-kash River that had been blocked from him as he went south from Khotan. His solution was to journey east along the foothills and then turn south around the end of the highest range, pass behind it, and then turn back west up the inner plateaus to find the river before then continuing west to intersect the caravan route to the Kara-korum. Along this lengthy route he would of course complete the topographical survey of many square miles of unexplored terrain. And of course Stein simply loved being in extremely high mountains. But the whole region lacked vegetation, so how to feed ponies during what would surely be at least a 40-day trek? Stein planned a final supply at Polur, the last inhabited hamlet in the Kun-lun - then to reach the Kara-kash valley where Satip-aldi Beg would bring a Kirghiz supply caravan and establish a depot. As always, Stein's planning was meticulous and far-seeing, but events turned out otherwise. No matter how many ponies and donkeys were utilized they could only carry fodder for themselves. Yet the party required also 7 saddle ponies for the men and 10 more for the camp and survey equipment. which would require another 17 ponies to carry their fodder, and so on and on, plus the pack animals would require more men as drivers. (This is a familiar problem analyzed by students of Alexander the Great's campaigns). Stein decided to use teams of donkeys in groups that would be sent back as their fodder loads became exhausted. There were many donkey men in Khotan. But Stein faced the same kind of problem that he found in Kan-chou. He would be paying exorbitant prices for animals he wanted to survive the entire tourney; but, having received more than the full total value for the animal as a rent, the owners would prefer to palm off their worst beasts and not see then again. Stein was deeply appreciative of the invaluable assistance Badruddin Khan, the Afghan Ak-sakal, provided in expertly handling this task. Lal Singh arrived on 20 July from his extensive survey project which had been hampered by damage to the level on the theodolite. Stein completed his archeological efforts by a return to Yoktan to purchase some of the new season's 'finds' of terra-cotta figures. On 1 August he sent off the main caravan of over 50 loaded camels to Sanju and then on to the upper Kara-kash where Stein expected to join it for the trek over the Kara-korum. Moving south-east he had to cross the flooded Yurung-kash by ferry boat, an undertaking that required hours to get his three boatloads safely across. He rode through Sampula to Kotaz Langer. There he said good-bye to Badruddin Khan and Chiang.

Chapter XCII - In the Gorges of Polur and Zailik
Stein rode for 5 days into the Kun-lun mountains to reach Polur. By 5 August he was at Hasha. At Imamlar he visited a famous shrine at 7300 feet elevation and with temperature falling to 37 degrees. He reached Polur on 8 August. There by prior order a depot of flour, fodder, and live sheep and all was already established. Extra transport was assembled as well. The idea for this effort was to move extra store high into the mountain without use of the main transport animals. There he entrusted his last mail bag to devoted Turdi, the dak man, to carry to Khotan. (How this mail carrier always managed to be in the right place for Stein is a wonder in itself.
The trek resumed on 12 August. Three days of struggle over boulders and rock ledges through narrow gorges brought them to 13,000 feet at Khan Langar. At the most dangerous places the loads were removed from the donkeys and carried around rocky obstacles by the Polur men. On 15 August they were in a broad valley and over the watershed at 16,500 feet. Stein and Lal Singh climbed a ridge.
Here is his comment. "I enjoyed a glorious view both of the towering range behind us and the great wall of snowy mountains which flanks the Yurung-kash River sources. It was a grand panorama, but was hard to photograph it with an icy gale from the north benumbing one's fingers."
The contrast in Stein's vocabulary, when he describes in delighted terms the sights and his emotions in the highest mountains with his descriptions of the bleak and folorn scenes in the desert, leave no question about his love of mountain living.
They then descended to 15,000 feet to camp near several lakes. The temperature was now 20 below zero at night but 130 degrees by mid-day. Stein had a survey map created in 1897 by which he could establish some reference points. Lal Singh proceeded to expand a triangulation from these. The Surveyor luckily met a group of hunters from Keriya from which Stein promptly enlisted (although he was reluctant) as guide one Pasa. He informed that there was a gold-digging operation two days' march away and routes around the Yurung-kash. Stein sent the Polur men and donkeys home and shifted a depot supervised by Ibrahim Beg south west, toward Ladak. He then on 18 August went west over a pass at 16,200 feet to reach the valley of Zailik that contained the active gold diggings at 13,600 feet elevation. There they found about 50 diggers. The cliffs were covered with holes and galleries from centuries of gold prospecting. The many graves attested to the final results for many miners impressed to work these meager deposits. Stein and Lal Singh climbed peaks and descended gorges in the area to establish their trigonometric and topographic survey. They measured the Kun-lun peaks at 20,000 feet around the eastern sources of the Yurung-kash. The 'inexpressibly grand view' to the south measured 'a magnificent range of snowy peaks' to 23,000 elevation. They also recognized peaks that had been triangulated long before from the other - south - side north of Ladak.
They climbed ridges above Zailik from which 'opened out panorama more impressive and grand than any I had so far seen. As my eyes ranged over that amazing maze of ice-crowned spurs and deep-cut valleys enclosed between the two great Kun-lun ranges and drained by the Yurung-kash, it was an inspiriting thought that the whole of this grim mountain world was unexplored ground, and that in all probability human gaze had never rested so long on it.'
They were lucky also in that Stein was able to hire the miners for temporary work carrying his equipment and baggage. He engaged miners in 10 minute reliefs to sit beneath the plane table to steady it against the icy gale. The miners were only able to work at Zailik for a short summer season, the annual output amounted to only 300 ounces of gold. The miners are mostly bound-slaves. Thus they were ready to earn a fabulous wage carrying Stein's equipment. The next leg over even more difficult terrain was impassable for ponies. Stein sent the animals and much of his luggage back to Ulugh-kol depot and followed Pasa's guidance with the miners as carriers toward the Yung-kash glacier.

Chapter XCIII - To the Yurung-kash Glacier Sources
The next trek began on 25 August up a steep valley southward and across a ridge and over a pass at 17,700 feet between Zailik and the Yurung-kash valley. Stein again comments about the 'glorious panorama'. The following day they descended to the desolate Yurung-kash valley at 13,000 feet elevation. The next day they again climbed through another 17,000 foot pass. Stein and Lal Singh climbed another peak to establish a triangulation station with a 'magnificent panorama'. This one was at 18,612 elevation. They were able to see a huge massif, covered in snow that fed the eastern sources of the Yurung-kash. They measured the snow line at 17,500 feet on the north side and 19,500 feet on the south side of the range. On 20 August they continued over more spurs to the valley of Tuge-tash at 15,000 feet. The following day they reached the valley. Further up they reached huge glaciers. Not to be deterred, Stein used ropes to enable the donkeys to cross raging streams of glacier water while the men carried the loads.
Stein describes the dangerous path vividly. For instance, "On our left across the gorge there was a succession of perfectly wall-like spurs with deep chasms between; but in front of us there showed at last long rounded ridges of detrius. As we ascended towards these the frowning spur above us assumed fantastic shapes of towers and huge battlemented walls."
At 15,500 feet after a nine-hour struggle to move four and a half miles they crossed the mouth of a huge glacier that rose above them to 21,000 feet. They camped in a narrow gorge. On 1 September they continued to the north-east and at 15,600 feet elevation rejoined the ponies and supplies that had been sent ahead by a different route. This was the basin formed by the several feeders of the Yurung-kash. On September 2nd Stein and Lal Singh climbed another peak to 17,400 feet to photograph another panorama. They could see peaks as high as 23,000 feet and as far as 60 miles away.
"It is impossible to describe here all the features which gave overwhelming grandeur to this panorama, and which no photograph can adequately render."
Stein wanted to linger longer in his mountain paradise but the shortage of fodder made movement toward the Karakorum urgent. So on 3 September he started back to Ulugh-kol and his depot. The route crossed a pass at 16,000 feet.

Chapter XCIV - Across Tibetan Plateaus
At Ulugh-kol Stein reorganized his party as planned. He sent the now surplus donkeys and their attendants back to Polur. He sent the Zailik miners back as well with generous pay and food. He retained Pasa and another hunter to guide his party on. The remaining group started out on 4 September westward toward the Kara-kash via the Polur-Ladak route and across a basin at 17,000 feet with the headwaters of the Keriya River. For 5 days they followed a route previously surveyed by Captain Deasy. Unfortunately during the first night Pasa and his associate took what Stein calls 'French leave' and disappeared.Stein forgave his real fright at continuing into the unknown mountains and sent a reward to him via Badruddin Khan. The group continued to the Baba Hatim Pass at 17,600 feet led to the Keriya River. Rain and snow all the way greatly increased the difficulty. They continued south for two more days to another basin at 17,200 feet where the Keriya River begins at the base of large glaciers. Continued snow storms made further movement up the Keriya River difficult. The snow turned to mud making travel on 7 September along the river extremely difficult for the laden donkeys and ponies.
Nevertheless Stein writes: 'Whenever the sun broke through between the snow-storms my eyes could revel in the glorious glacier array to the west, with wonderful bluish tints in its shadows.'
That night they lost the first pony to exhaustion. Much extra labor for the men was required to pull the animals out of continual bogs. Gaining another ridge at 18,000 feet at least they bypassed more bogs. In this passage they left the Keriya watershed and entered that of Lake Lighten. That night they found some vegetation for the animals. The temperature was 17 degrees below freezing. On 9 September they left the Ladak route and turned south to Lake Lighten. At this they passed again into unknown territory. That night they again found some sparse grass for the animals. They followed the lake shore west for 20 miles at an elevation of 16,100 feet. Stein again waxes ecstatic in his description of the scenes unfolding as they moved along in this valley surrounded by giant mountain ranges. With supplies for man and beast becoming dangerously low they had to rush past all this and head for the Kara-kash. At least in the next valley they found wild yaks and the coarse grass that attracted them. Grazing was interrupted by another snow storm. On 11 September they passed the western end of the lake when an impassable cliff again forced them to seek a pass higher up. This they found at 17,700 feet. But the frightened drivers lagged on purpose causing Stein to back-track and remain another night in the previous valley. On the 12th then forced their way through over another spur and again found wild asses grazing in a secluded valley. They trudged on through snow storms and gales of ice. When the sky cleared they saw a 23,400 foot mountain to the north. Stein again: "quasi-arctic grandeur", but they could not afford even a brief stop to do a triangulation. September 13th found them again beset by ice and snow storms during another 20 mile march. That night two donkeys could not continue and had to be shot before the morning start. The terrain became more open but that meant the mountain range that sent water into the plateau was further to the north. The plateau became even dryer resulting in danger for their water supply as well as for fodder. Fortunately that evening they found water by digging a well to 3 feet depth. They were still at 15,500 feet elevation. By the next morning another pony died. On 15 September they sighted a salt lake some 16 miles long. Now the terrain was depressing even for Stein. At camp another well produced enough water.

Chapter XCV - On an Old Mountain Track
Now at this salt lake Stein had a sketch made by an explorer back in 1865. This gave him a course to follow to the Kara-kash. Despite the dire state of the animals and supplies there was no choice but to try to find this route. On the 16th they crossed another saddle at 16,800 feet. They faced another huge, desolate basin without indications of water or vegetation. Stein found some water. But the following morning had the very distressful event of the demise of his faithful Badakhshi pony that he had ridden for 2 years. September 17th was another depressing day with little fodder remaining and the stream they followed was too salty. Stein spotted two man-made cairns protruding from the sand. These were clearly a route marker of some sort - indeed the trail not used in 40 years. Stein remembered from history that this was the route Haji Habibullah had attempted to open from Khotan when contending with the power of Yakub Beg in 1865. This meant water, at least and fuel. They made camp at 15,700 feet elevation. Two more donkeys had to be shot the next morning. The path over the next pass at 16, 500 feet was well marked - they were now on a known trail. Each cairn had a store of firewood and a few relics from the 1860's. They followed the train down and then up across another saddle at 16,000 feet. In the next valley they found tracks of yaks and asses and TWO men. Further on then found the Haji Langer built long ago by Haji Habibullah that Satip-aldi had described to Stein back in Khotan. They and the animals were saved. But Stein thought most about the loss of Badakhshi.
Stein set out on 19 September with his refreshed crew on down the valley to reach Abdl-Ghafur-tam, where he hoped Satip-aldi's men would be waiting. After passing the branch of the Kara-kash flowing from a side valley they reached another small meadow with grass. During the evening Satip-aldi arrived in person with his yaks and followers and news that all further plans had been accomplished. Tila Bai was safely at Suget Karaul with the huge and precious caravan and Captain Oliver had prepared the yak transport on the other side of the Karakorum.

Chapter XCVI - The Search for the Yangi Dawan
The 20th was a day of rest and feasting on the supplies that the Kirgiz brought in on their yaks. But Stein and Lal Singh spent much time with their plane table. Stein was still determined to find the Yangi Dawan and the rest of Haji Babibullah's route. On the 21st Stein and Lal Singh set out with yaks up a side valley and by stone walls built over 40 years before to clear a path along steep mountain walls. They made their nightly camp at 16,700 feet. On the 22nd Stein assaulted the glacier with a team of Lal Singh, Musa (Lal's helper) and 4 hardy Kirgiz who had visited the area previously. By 8 AM they reached the limit of the Kirgiz's endeavor. They left most of the yaks and an attendant at that point. Stein then tied everyone together with ropes, mountaineer style, for crossing the glacier. With yaks making a path as far as possible they proceeded upward. After 7 hours of struggle through the snow they reached the crest at 3 PM, where Stein could set up the plane table. They were now at nearly 20,000 feet elevation. Stein included the panorama photo he made at this point. Again, he waxed poetic noting ' it as a picture of Alpine majesty such as I had nowhere beheld so close in the Kun-lun'. Peaks around them reached 21,000 and 23,000 feet elevation. Stein indicates he was 'elated' but not for long. He was so absorbed with surveying work and enjoyment of both his success and the scenery that he forgot what was happening to himself. By 4 PM and time to start down Stein's boots were soaked through and then frozen. He noticed no pain in his feet initially. It was dark when they reached the waiting Kirghiz and yaks but could not stop for fear of nightfall on the mountain. Descent in the darkness was slow. Pain began to mount in his feet. They reached camp and fire but it was too late. Stein had frost-bite in both feet. The servants worked hard to save the feet and did manage to help with the left foot. But the right foot was already too far gone at least in the toes.

Chapter XCVII - From the Kun-lun to London
This chapter comes as an anti-climax, as the title may indicate. On September 23rd Stein was suffering. He could not move but knew that he must reach medical help very fast indeed. He rested at the mountain camp one day and then had to suffer more strapped on a camel as they made it to the main camp. At Abdul-Ghafur-tam they met Ibrahim Beg with the ponies. He made a litter to be carried between two ponies. They reached Portash on the 27th where Stein could see the large caravan was assembled. There he had to spend two days calculating payments and settling accounts with Satip-aldi Beg and the Khotan men. He had to make the arrangements for the camels to carry the cases of antiques to the Karakorum and for yaks to carry them on across the Sasser Glacier. He left Rai Lal Singh in charge of the convoy. Word was rushed ahead to alert the Ladak villagers to assemble yaks. Another messenger was dispatched to seek medical help from the mission in Leh. Stein started south on 20 September. He passed lines of skeletons indicating the perils of the Karakorum which he crossed at 18,687 feet elevation on 3 October. For the next stage, impossible for a pony litter, he was met by teams of Tibetan laborers sent north from Leh for the purpose of carrying him. They crossed the glacier at the Sasser Pass on 7 October. On 8 October Reverend Schmitt from the Moravian Mission at Leh met Stein on the trail above the Kardong Pass. Four more days passed before they made it to Leh on October 12th - 300 miles from the site of the accident. The toes of the right foot were amputated. Recovery at the hospital in Leh required 3 weeks rest. On November 1st Stein bade Reverend Schmitt good by. He crossed the Zoiji-la Pass at 11,000 feet on 10 November and reached Srinagar on the 13th. Stein stayed there with Captain Oliver and Captain Macpherson. On December 1st Stein continued onto Lahore and then to Calcutta. At the headquarters of the Survey of India Stein saw Rai Sahibs Lal Singh and Ram Singh and the Rajput noble Mian Jasvant Singh. Rai Lal Singh was promoted with the title of Rai Bahadur. A special award of a gold watch was dispatched to Kashgar for Chiang-ssu-yeh.
On December 26th Stein left Bombay by steamer for England. He was granted 'detached duty' in London to organize the results of his memorable expedition.