nor after that was there more than a single short border war with the Kitan power. As to Japan, in 925 alleged tribute was sent; but Japanese history puts a very different complexion upon the matter, and says that the Mikado refused to recognise the envoy sent to Japan "by Manchuria” (ie., by the state conquered in 925 by the Kitans) because he was not duly accredited. In 1091 a certain Chêng Yuan

conducted a large tribute mission from Japan to Cathay. Japanese history says nothing of it, but exactly the same man is mentioned in the Sung history as being a Japanese author a century earlier; still, the names of two of his fellowenvoys have the ring of truth about them, and whilst there may be some confusion, it is improbable that there is misrepresentation.

The Cathayans had extensive relations with Tangut (ie., the kingdom of the Ordos or Yellow River Loop region), and also with various Tibetan tribes of the Kokonor region. The diplomatic relations with the three or four petty Ouigour States were regular; but Cathayan influence. never extended to modern Turfan, Hami, Urumtsi, Cobdo, Kashgar, or Yarkand; unless, indeed, the Tsupu represent the Hami-Cobdo region. The Turks came in 928 and 991; but these must have been mere remnant tribes, for the Ouigours (themselves mere remnants) had long supplanted the true Turks, who were now in full career west. Kirghiz came in 931, 952, 976; but that means very little, as the Kirghiz never became a serious power, and had long disappeared from Chinese ken; moreover, there is evidence that in one of the above cases the words "Turk" and "Kirghiz" were used by the historians to indicate what was really one and the same mission. From 1093 to 1099 there were brushes with the Basmäls, a tribe of Turks who are known to have lived in 700 somewhere about Urumtsi, and are also mentioned in the Turkish inscriptions.

The

Thus it will be seen that China proper, except in the neighbourhood of Peking and the Great Wall, remained practically untouched by Cathay, as also did Corea, Tibet,

and Turkestan. Cathay was in its essence a purely MongolManchu power, the Mongol element being dominant. The Nüchêns were much the same power with another face to it, the Manchu element being dominant; but the area of China proper encroached upon by the Nüchêns was larger. The present Manchu power is once more the same, or was

but the Mongol element has now been completely emasculated by drink and religion, whilst the dominant Manchu element has, in politically absorbing the whole of China proper, been itself socially absorbed into Chinese life, and has emasculated itself by opium and women.

The Cathayan military organization was very like that of the Boers. Every male between fifteen and fifty was liable to serve, and every man provided his own horse and kit. The rallying-place where armies were generated was Dalai Nor (Lat. 43° 20′, Long. 117°). The campaigns were almost always conducted in the autumn-winter season, and it was the invariable practice of the Cathayans to ravage the enemy's country. A gold-fish tally was the sole means of authorizing the movement of provincial troops: bird tallies had other uses; and all troops were mounted, whilst an excellent system of scouting and cavalry picketing secured the main army from surprise whilst on the march. In the five metropolitan provinces-i.e., in the empire viewed from a regulation or directly administrative point of view-there were at the best period 1,640,000 soldiers. Besides these, there were the troops of the sixty vassal states, only a few of which were actually ever called upon to supply armies. Thus the Shih-wei regularly supplied standing regiments, and also sent horses; the Hi (apparently the Tata-Bi of the Turkish inscriptions) were incorporated bodily; the Nüchêns had to fight against Corea, and at the same time were held responsible for keeping the other eastern Tunguses in order; they also supplied horses. The Coreans, on the other hand, were expected to join against

The

the Nüchêns. The Ouigours
The Ouigours supplied remounts.
Tanguts were ordered to assist against the Basmäls; the
Tibetans to assist against the Tanguts; and so on with the
Tsupu, Tukuhun, and other mixed or doubtful tribes. All
the Cathayan military titles are given, but I will not en-
cumber these pages with unintelligible names. I will merely
indicate, amongst others, the well-known Turkish and
Ouigour title of muiluk (identified by Dr. Marquart with
what he calls the "Old-Turkish" word buiruk); it appears
in the following disguises, miluk, meilao, molin, moli, meili,
meiling, muilêng, from A.D. 600 down to our own day; also
the word ta-la-kan (tarkhan), which also has a pedigree of
1,200 years. The title sz-kin (djigin) is specifically stated
to be Turkish. The Cathayan t'ik-yin I take also to be
the Turkish and Ouigour tegin, which appears so often in
the inscriptions. The rest of the purely Cathayan titles
must lie by for the present until we know more. If we ask
from what population so large a number of soldiers was
drawn, we must answer that the Cathayans have left us no
military records similar to those of the Nüchêns; besides,
they never owned the four or five Chinese provinces which
easily brought the population of the Nüchên Empire up to
50,000,000. It is evident that 1,640,000 men between the
ages of fifteen and fifty must mean at least 3,200,000 males,
as many females, and twice as many children—say, a
minimum population of 20,000,000, of which the Chinese
would certainly form two-thirds. The outlying tribes of
the areas subject to Cathay have never, even to this day,
numbered more than about a million souls, nor does all
Siberia now contain 5,000,000. Nor is there any record
left of a financial and administrative capacity such as
distinguishes the Nüchêns and Manchus, and even to
a certain extent the Mongols. The Cathayans, in short,
were the Vandals of China, and have left not even a wrack
behind them; their very name is unknown to popular
tradition in the places they once occupied; nor is there a
single incident recorded of any Cathayan which suggests

the least nobility of character, with the single exception of Yelüh Tashih.

In religion the Cathayans seem to have been shamanistic, but no mention is made of priests. Later, Buddhist bonzes gained some influence. The old Kitans worshipped the sun, but not the moon, and faced east at all sacrifices. There were sacrifices to mountains, prayers for rain, burnt sacrifices of gratitude, and (for the royal families alone) twelve-yearly lustrations called "being born again." White horses and gray oxen were sacrificed to Heaven and Earth, and later also to ancestors; sometimes gray horses and red oxen. Geese and deer are mentioned as being sacrificed to Heaven alone, and the sun-worship at Tailin (a vague term variously supposed to be a willow coppice and a place-name) is specifically and manifestly that of the ancient Sien-pi; moreover, the burnt rams, facing east, the "Red Mountain," the marriage customs, etc., all combine to show that the Cathayans and Shih-wei (ie., as I believe the Mongols) were different phases of one and the same Tungusic souche. Human sacrifices, akin to suttee, of concubines, slaves, favourites, or prisoners with the distinguished dead, were not infrequent, but perhaps not common. A curious ceremony of "leading a ram" (or possibly goat) is twice mentioned in connection with the formal surrender of kings in mourning attire; also the presenting to the surrendered as personal names of the names of the conqueror's war chargers.

There is not the same specific evidence of Cathayans marrying stepmothers and sisters-in-law as there is in Turkish history. In 940 marriages with deceased wives' sisters were abolished, which looks as though a radical change in ideas had taken place. In the same year Cathayans holding Chinese office were allowed to marry Chinese with Chinese forms. In 1017 women of a certain rank were forbidden to re-marry; and in 1019 the "three superior tents" (probably akin to the Manchu "three superior banners") were forbidden to marry with the "meaner

tents," and all marriages in superior tents had to receive imperial sanction. In 1094 border people were prohibited from marrying with foreigners.

*

It is not easy to make much out of the Kitan words. given to us on Kien-lung's authority in Chinese dress. But two are very suggestive: Ta-hu-li, "cultivator," and Sholu-n, "mountain-peak." I take these words to be the etymological origin of the Daour and Solun Tunguses of to-day. There is also a word A-mu-t'a-ha, "hunter," which may be the Manchu word butkha, "hunter," now applied to the town of "Putcha," written Butkha by the Russians. T'ê-li-n is Cathayan for "lake,” and tê-lin is Nüchên for "sea"; they both suggest connection with the Manchu mederin, "the The swan wo-lu-to is said to mean "intimate ordo" -i.e., the royal or higher caste. The other ordo are thus enumerated: Kwo-a-lien, "grazers"; ye-lu-wan, “ prosperous"; p'u-su-wan, "start grandeur"; to-li-pên, "conquering"; kien-mu, "bequeath"; ku-wên, "prince" (possibly a misprint for "jewel "); nüku, "gold"; wo-tu-wan, procreate"; a-sz, "broad"; alu-wan, "assist"; tê-shih-tê-pên (or ch'ih-shih-tê-pên), "filial"; na-po, "shooting-box," or "travelling palace"; cho-wa, "falcons"; tê-li-kien, "Empress"; nou-wo (mo), “earth (mother)"; sa-la, "wine cup." The word i-r occurs in such a way that it must mean "day," and this suggests a connection with the Corean il, day," which, however, is a strictly regular form of the ancient Chinese yit, "a day," now pronounced ji in Peking. Wan is evidently a grammatical inflection or agglutination.

The next question is, Where are the Cathayans now? In 1161-63 there was a serious rebellion, which caused the Nüchêns to abolish the semi-independent Kitan chiliarchs and centurions, to confiscate all their horses, and to distribute the Cathayans over the Nüchên divisions. In 1169 another Cathayan revolt was quelled, and the Cathayans who did not revolt were removed to the "Ukuli-Shile "