286 
worshipped there owing to the popular belief that in ancient 
Buddhist times they had saved the land by destroying the 
leather of the harness and armour of some hostile army. The 
locality indicated by the Chinese pilgrim corresponds exactly to 
Kaptar-Mazar, and Dr. Stein has no doubts that the pigeons of 
the Muhammadan legend have taken the place of the rats of 
the legend as related by Hiuen-Tsiang. 
During the whole of his journey, Dr. Stein paid the greatest 
attention to historical topography. Everywhere he tried to 
trace and identify ancient sites mentioned by Hiuen-Tsiang and 
other Chinese travellers. Thus—to mention only some of the 
more important results—Paloyo, the Dard designation of the 
people of Baltistan, was identified with the term Po-liu, as used in 
the Chinese Annals and in the narratives of the Chinese pilgrims. 
Sir Henry Yule’s identification of Sarikol with the K’ie-p’an-to 
territory of Hiuen-Tsiang was fully confirmed by Dr. Stein’s 
investigations. On'his march to Khotan he was able to identify 
the small oasis of Moji with the town of Po-Kia-i, where a 
famous Buddha statue brought from Kashmir was worshipped 
in the times of Hiuen-Tsiang. Following the road once used 
by the Chinese pilgrim, he traced other ancient sites near the 
oasis of Zangiya, and close to the frontier of the Khotan district. 
Two identifications, previously made by M. Grenard, were fully 
borne out by the evidence found by Dr. Stein—that of the 
Kohmari ridge and cave with the ancient Gosrnga mountain 
and the cave where the popular legend of Hiuen-Tsiang’s time 
supposed a Buddhist saint to reside ‘* plunged in ecstasy and 
awaiting the coming of Maitreya Buddha”; and that of the 
village of Yotkan with the ancient capital of Khotan. Among 
the many proofs for the latter identification, the most convincing 
was that, from this starting point, Dr. Stein was able to identify 
the positions of the most important Buddhist shrines visited by 
Hiuen-Tsiang, the places of which are generally occupied now 
by Muhammadan Ziarats. Thus, the small hamlet of Somiya 
was found to correspond exactly to the Buddhist convent 
described by the Chinese pilgrim under the name of Sa-mo-joh. 
Finally, we may mention that on his march from the Karadong 
ruins to Keriya, Dr. Stein identified the position of the town of 
Pi-mo, described by Hiuen-Tsiang, in the neighbourhood of 
Lachim-Ata Mazar. 
But it is chiefly for his archaeological discoveries and his 
manuscript finds that Dr. Stein's journey of exploration will 
ever be memorable. We cannot enter here into details about 
the many interesting Buddhist monuments examined by Dr. 
Stein in the course of his travels in Gilgit, Hunza, Sarikol 
and Kashgar, and the antiquities collected by him on the 
Yarkand-Khotan route, in Khotan town and in the village of 
Y6tkan. Also for the excavations made at the Endere site, at 
Karadong and at Ak-sipil we must refer the reader to Dr. 
Stein’s ‘‘ Preliminary Report.” But a word or two must be 
said about the most important results of the excavations carried 
on among the ruins of Dandan-Uiliq, the Niya River site and 
of Rawak. 
No less than fourteen detached temples and dwelling-houses 
were excavated in Dandan-Uiliq. First of all two temple cellas 
were brought to light, richly decorated with wall paintings and 
stucco images. The interior of the larger cella was occupied by 
a colossal stucco statue, probably representing a Buddha. Each 
of the four corners of the same cella was occupied by a draped 
stucco figure standing on a lotus-shaped pedestal. The cella 
walls were decorated, inside with frescoes showing figures of 
Buddhas or Buddhist saints, and outside with fresco bands con- 
taining small representations of saints, seated in an attitude of 
meditation. In style of composition and the drawing of figures, 
these wall decorations are similar to the later of the Ajanta 
frescoes. But as we passess only very few specimens of old 
Indian painting, the study of the Dandan-Uiliq frescoes will 
prove of particular interest. For the same reason, the small 
painted tablets which Dr. Stein discovered on excavating the 
temple cellas are of importance. They were probably votive 
offerings from worshippers who had come to visit the shrines in 
ancient times. A figure represented on one of these tablets 
shows the head of a rat—which is interesting in view of the 
legend of sacred rats mentioned above. Near the excavated 
buildings Dr. Stein generally found groups of shrivelled and 
bleached trunks of poplar and fruit trees, the remains of ancient 
orchards or avenues. Also traces of old irrigation channels 
were recognisable in the sand. 
Of the manuscripts excavated at Dandan-Uiliq, the most 
important are some oblong leaves of paper inscribed with old 
NO. 1707, VOL. 66] 
NATURE 
| JuLy 17, 1902 
Indian Brahmi characters (z.e. the alphabet which is written 
from left to right, and used in the edicts of King Asoka, and 
similar epigraphic documents), and belonging to five different 
manuscripts, three of which are in Sanskrit and contain Buddhist 
texts. From their palzographic peculiarities Dr. Stein con- 
cludes that they cannot be later than the seventh, and may 
belong even to the sixth or fifth, century. Moreover, there 
were found single leaves of thin, coarse paper, inscribed with 
cursive Indian characters, but showing a non-Indian language, 
and some Chinese documents of similar material and appear- 
ance. Two of the latter bear dates, according to which they 
must have been written between 763-805 A.D. Dr. Stein thinks 
that these dates indicate about the time when the dwellings were 
abandoned. The evidence of numerous coins found in the 
course of excavations supports this dating of the Dandan-Uiliq 
ruins. 
Among the most interesting discoveries in the ruins at the 
Niya River site, there are remains of two large dwelling- 
houses, excavated by Dr. Stein. In one of them some speci- 
mens of household furniture, illustrating the industrial arts of 
the period, were found, amongst others a wooden chair with 
ornamental wood carving, the decorative motives of which 
closely resemble those of the relievo sculptures of the Buddhist 
monasteries of Yusufzai and Swat (the ancient Gandhara). In 
one room, the stuccoed walls of which showed a carefully 
executed fresco decoration, the pieces of a coloured rug—an 
interesting specimen of ancient textile industry—were brought to 
light. Again, in another of the excavated houses there were 
found the lees and arm-rests of a wooden chair, representing 
lions and human-headed monsters, and still retaining traces of 
colour, and also the broken end of a kind of guitar, resembling 
the popular ‘‘ Rahab” of modern Turkestan. 
But most important of all are the manuscripts unearthed at 
the Niya River site. More than 500 wooden tablets inscribed 
with ancient Kharoshthi characters (¢.e. the alphabet written 
from right to left, and known chiefly from Indo-Scythian and 
Indo-Greek coins, found in the north-west of India) were found 
among the ruins of this site. Most of them are wedge-shaped, 
from 7 to 15 inches long, and arranged in pairs; and some of 
them still retained their string and clay sealing intact, thus 
illustrating the ingenious manner adopted for the fastening and 
sealing of these documents. Other tablets were oblong, some 
of considerable length (up to 30 inches), resembling the Indian 
palm-leaf manuscripts. An ancient pen, made of tamarisk 
wood, with a bone knob, was found, and gives us some idea of 
clerical work in this remote period. A considerable number of 
these tablets were found in an ancient rubbish heap, and there 
were also some narrow pieces of wood inscribed with Chinese 
characters. The same rubbish heap yielded another very rare, 
and in a Buddhist country particularly surprising, writing 
material, namely, about two dozen documents written in Indian 
Kharoshthi characters on leather. A thorough examination of 
all these documents as to their contents will take much more 
time than Dr. Stein was able to bestow on them during his 
short deputation. But he could make out that most of them 
were written in an old Prakrit dialect with an admixture of 
Sanskrit terms, and the wedge-shaped tablets seem to contain 
correspondence, records of agreement, bonds, memoranda and 
the like, while religious texts, votive records, &c., will probably 
be found to form the contents of the longer tablets. As to the 
date of these documents, palzeographical evidence proves them 
to belong to the first centuries of our era. For the writing re- 
sembles closely that on the inscriptions of the Indo-Scythian 
kings who ruled over the Punjab and the Kabul region during 
the first two centuries, and the Kharoshthi alphabet soon 
ceased to be used after that period. These wooden tablets 
must, therefore, be considered at present as the oldest Indian 
manuscripts extant. The use of wood as writing material is 
also a proof of considerable antiquity. From the fourth century 
onward, the use of paper as writing material is attested for 
Turkestan. Yet not the smallest scrap of paper was discovered 
in the ruins of the Niya River site. Numismatic finds, as well 
as the influence of classical art shown on some of the clay seals 
attached to the tablets, confirm this dating. 
The last excavations were those made at Rawak, where Dr. 
Stein found an imposing Stipa surrounded by a court forming a 
quadrangle 164 feet long and 143 feet broad. Both inside and 
outside, the walls of this Stipa court were decorated with rows of 
colossal statues in stucco, representing Buddhas or Bodhisattvas, 
and between them at frequent intervals with smaller relievos 
