that he lost his life in consequence of an enemy having addressed to him a secret political letter written in the "smaller Cathayan character," with a view to encompassing his destruction. But as Nüchen is officially stated to be based on Cathayan, it is not impossible that Salican knew both. In 1887 Dr. Hirth (now in Munich) discovered a Chinese key to the Nüchen script; but whether the Berlin authorities, in whose possession I believe it now is, have utilized it in order to elucidate all the points raised by M. Devéria (Revue de l'Extrême Orient, 1882) I do not know. I learn, however, from M. Chavannes that M. Grube has ascertained from a study of that key that another inscription, known as that of Yen-t'ai, is undoubtedly Nüchen. This much is quite certain: the Annamese, Tanguts, Cathayans, and Nüchens all constructed for themselves syllabaries formed by the comparatively simple process of grouping together in an incongruous. way the strokes or halves of Chinese characters. With Annamese this is quite easy, for the language consists of monosyllabic and tonal roots like Chinese; but, as at least two of the other three languages mentioned are agglutinative, purely phonetic signs had to be devised. for prefixes and terminations, as well as ideographs for root-meanings or roots. No doubt in connection with or in continuation of the same inventions it was that, for sesquipedalian languages like Corean and Japanese, systems of a more purely syllabic, not to say alphabetical nature, were evolved from the same mutilated Chinese materials, eked out with ideas derived from Sanskrit or Pali priests, who wandered all over China at that time. The story of Nüchen script, as I gather it from Nüchen. history, is as follows: In the year 1119 was issued to the public the Nüchen form of script invented by Wanyen Hiyin, and in 1125 one Yelu was ordered up to the capital to teach it. In 1138 the Emperor Khara himself invented a new form of Nüchen, called the "small" (or short-hand) character; it was ordained that Cathayans, Nüchens, and Chinese should each use their own writing,

THIRD SERIES. VOL. TY.

U

and that the Bohai people were to count for this purpose as Chinese. In 1145 the first official use of these small characters was made, and in 1183 a considerable number of Chinese classical books, histories, etc., were published in Nüchen; a little later all hereditary mingans and meuks had to be able to read Nüchen before succeeding to their commands; in 1188 a Nüchen college was established; it is distinctly stated, however, that the Emperor Matako was the only one of the Nüchen princes who ever became a really competent scholar in his own language. In 1191 the Cathayan written character was abolished, but Chinese and Nüchen law-clerks still accompanied each circuit judge. In 1194 Yelu Kushen's memory was honoured with a temple at the upper capital because (like the semi-mythical Ts'ang Kieh, it is at the same time stated, who was similarly honoured for inventing Chinese) he "first made the Nüchen script."

Now Yelu, Yelu Kushen, and Wanyen Hi-yin, are manifestly one and the same man, for in Wanyen's biography it is stated that his "old name" was Kushen. He was the son of one Hwantu, and his great-grandfathèr had been an intimate and fellow-villager of the fifth Nüchen chief, who, as we have seen, was a Cathayan tiyin, or "governor," and would therefore be a man of some ideas and instruction. Yelü was the surname of the Cathayan royal house, and it would be quite customary to “present the royal surname " to a prominent man, who, being of the royal Nüchen Wanyen tribe, as his name shows, would naturally revert to his own surname when the Nüchens overturned the Cathayans. Yelu scarcely differs in sound from Yelü. Wanyen Hiyin was present at the taking of Peking, and his name is mentioned later on in connection with Kara-Cathayan and Tangut plots. It is said in his biography that Khara was jealous of him, and that, having been degraded in 1139, he was "allowed to commit suicide' by the same Khara. Khara is stated to have been jealous because he had no son of his own; it might have been literary jealousy, too.

There were a great many shiftings about of populations during the early part of the Nüchen dynasty. Useful Chinese, such as artisans and scribes, were moved up to the upper capital on a wholesale scale, and Nüchens from the unproductive lands of the Altchuk Valley were sent westward to cultivate parts of the old Cathayan metropolitan circuit-what is now the modern Korchin Mongol reserve; the Salican above mentioned was one of the few who successfully resisted this forced emigration from his ancestral river. There are no details of population until 1183, when it was found that there were 615,624 households of 6,158,636 souls (one quarter slaves), under 202 mingans and 1,878 meuks, cultivating 1,690,380 k'ing—say 26,000,000 acres—of land, and owning 285,000 cattle. In 1190 there were 6,939,000 households of 45,447,900 souls in the whole empire (apparently inclusive of mingans and meuks), which then extended to the river Hwai, and included Shen Si, Shan Si, Shan Tung, Ho Nan, and even part of Kiang Su. By 1195 these figures had gone up to 7,223,400 and 48,490,400 respectively. This is a very high figure indeed for so limited an area; but even in Nüchen times the Mongol wars had considerably reduced this, and in 1274, under Kublai Khan, there was only one third or a quarter of that population in the same area. The Nüchens had half a dozen different classes of householders: the "proper" were genuine Nüchens; the "mixed" were Cathayans and Chinese. The other classes are not clearly defined, but they point to a probable discrimination between soldiers, scholars, colonists, occupiers of tents, slaves, etc. In 1193 there were 11,495 officials in the Nüchen Empire, 6,790 being Chinese, and 4,705 Nüchens; no Kitans.

Mr. Pozdneyev is in error when he says that the Mongol history makes no mention of Northern Manchuria. During the Mongol Dynasty (1234-1368) scarcely a year passes without some mention of the Nüchens, who are throughout in most cases mentioned with the "Water Tatas;" these appear to have been the descendants of those Black

waters north of the Hun-t'ung (Dr. Bretschneider thinks perhaps the "Su Moals" or "Water Mongols" of Rubruquis are meant). In 1283 both branches of this Tungusic race were placed under the provincial government of Liao Tung-Hai Si. Those Nüchens born in the north-west and ignorant of Chinese were treated as Mongols, the rest as Chinese-so far as holding office went. They took an important part in the disastrous invasion of Japan under Hung San-K'iu (Marco Polo's Von-sani-cin). The modern Mukden was part of Nayen's appanage Kublai had to march in person against this Prince, as correctly stated by Marco Polo, whose “Barscol" may possibly be Bars-koto on the Kerulun, and perhaps the kotun city of Cathayan history, which was certainly situated about there. In 1697 the Manchus mention a place called Pa-r-s-ku-r, near Hami; but Nayen's appanage hardly went so far west. There were "dog-posts" on the Sungari-Amur roads under an official called the tuctakhasun; but nothing is said of such in Nüchen times, though the Nüchens often brought dogs to Cathay as tribute. The dogs in Mongol times were fed on fish. After 1330 the word Nüchen does not occur, but a tribe called the " Ushe Wild Men" are mentioned with the Water Tartars as being in joint revolt. Both in Cathayan and Nüchen history this tribe is frequently mentioned under the same name, or as "Uje" or "Uzhe," and their habitat seems to have been west of modern Ninguta. In 1355 a decachiliarch was placed over the "Wushe" Wild Men, with residence at Harfen. This is probably the Harpin of to-day, one of the railway-stations on the Russo-KirinTsitsihar line, where it is joined by the Vladivostock branch. The word also seems to occur in Nüchen history in the form Holipin-te, said to be on the north border.

During the Ming Dynasty, 1368-1644, there does not seem to be a single instance where either Nüchen or Water Tartars are even so much as named as individual peoples. But in 1387 a decachiliarch of the Uche, Nüchen, and

Kilimi was established, with residence west of K'ai-yüan city. In this Government was a Kin-shui Ho, or "Goldwater River," running north into the Sung-hwa Kiangmanifestly the Altchuk running into the Sungari. The word Kilimi also occurs in Nüchen history as the name of a tribe to their extreme east. Mr. Pozdneyev speaks of an ancient fortification near Mergen called Urkho-Kherim by the natives, and this might have something to do with the Kilimi in question: possibly the Cherim Mongols of to-day: but the direction given is rather wrong. During the Ming

founder's long reign even Peking was the appanage of his son, and scarcely yet formed part of the organized empire -à fortiori places north of it. That son conducted several campaigns into the "Uliangha" country, which was divided into three military circuits, possibly coextensive with the decachiliarch's three tribes. As the natives of those military districts were allowed to come and sell horses at K'ai-yüan and Kwang-ning, which places still exist under those names, and as a portion of those natives are called "Hai-si," it is evident that some, at least, were Nüchens, and lived in Central Manchuria or Kirin. Moreover, in 1486, natives of the same three districts were allowed to take refuge in Liao Tung from revolters, so that probably all three were well north of the Liao Valley. During the next 150 years the same tribes made frequent raids upon Kwang-ning and Liao Tung, and in 1606, after one such raid, we are suddenly told that "all the Kalka Mongols joined the Manchus ;" so that it is evident there must have been some connection between these natives and the Manchus. As the Manchus edited the Ming history, of course they would not dwell too much therein upon their own humble origin. As to the word Uliangha, the Mongols in the dynasty previous to the Ming always use it in the sense of a Mongol tribe; for instance, the great conqueror Subudai above mentioned, who, besides beating the Nüchens, assisted in the conquest of Russia, was, as also, of course, his son Ulianghadai (=man of Uliangha),