The word has generally been interpreted bluish-grey, but in the passage just quoted Fr.-Michel explains it by brillans. However, the evidence for noirs here seems strongest. Rashiduddin says that when Kublai was born Chinghiz expressed surprise at the child's being so brown, as its father and all his other sons were fair. Indeed, we are told that the descendants of Yesugai (the father of Chinghiz) were in general distinguished by blue eyes and reddish hair. (Michel's Joinville, p. 324: D'Ohsson, II. 475; Erdmann, 252.)

NOTE 2. According to Hammer's authority (Rashid?) Kublai had seven wives; Gaubil's Chinese sources assign him five, with the title of empress (Hoang-heu). Of these the best beloved was the beautiful Jamui Khátún (Lady or Empress Jamui, illustrating what the text says of the manner of styling these ladies), who bore him four sons and five daughters. Rashiduddin adds that she was called Ķún Ķú, or the great consort, evidently the term Hoang-heu. (Gen. Tables in Hammer's Ilkhans; Gaubil, 223; Erdmann, 200.)

NOTE 3.- Ungrat, the reading of the Crusca, seems to be that to which the others point, and I doubt not that it represents the great Mongol tribe of ĶUNGURAT, which gave more wives than any other to the princes of the house of Chinghiz; a conclusion in which I find I have been anticipated by Demailla or his editor (IX. 426). To this tribe belonged Burteh Fujin, the favourite wife of Chinghiz himself, and mother of his four heirs; to the same tribe belonged the two wives of Chagatai, two of Hulaku's seven wives, one of Mangu Kaan's, two at least of Kublai's including the beloved Jamui Khatun, one at least of Abaka's, two of Ahmed Tigudar's, two of Arghun's, and two of Ghazan's.

The seat of the Kungurats was near the Great Wall. Their name is still applied to one of the tribes of the Uzbeks of Western Turkestan, whose body appears to have been made up of fractions of many of the Turk and Mongol tribes. Kungurat is also the name of a town near the Sea of Aral, perhaps borrowed from the Uzbek clan.

The conversion of Kungurat into Ungrat is due, I suppose, to that Mongol tendency to soften gutturals which has been before noticed. (Erdm. 199-200; Hammer, passim; Burnes, III. 143, 225.)

The Ramusian version adds here these curious and apparently genuine particulars :--

"The Great Kaan sends his commissioners to the Province to select four or five hundred, or whatever number may be ordered, of the most beautiful young women, according to the scale of beauty enjoined upon them. And they set a value upon the comparative beauty of the damsels in this way. The commissioners on arriving assemble all the girls of the province, in presence of appraisers appointed for the purpose. These carefully survey the points of each girl in succession, as (for example) her hair, her complexion, eyebrows, mouth, lips, and the proportion of all her limbs. They will then set down some as estimated

1

at 16 carats, some at 17, 18, 20, or more or less, according to the sum of the beauties or defects of each. And whatever standard the Great Kaan may have fixed for those that are to be brought to him, whether it be 20 carats or 21, the commissioners select the required number from those who have attained that standard, and bring them to him. And when they reach his presence he has them appraised anew by other parties, and has a selection made of 30 or 40 of those who then get the highest valuation."

Marsden and Murray miss the meaning of this curious statement in a surprising manner, supposing the carat to represent some absolute · value, 4 grains of gold according to the former, whence the damsel of 20 carats was estimated at 135. 4d.! This is sad nonsense; but Marsden would not have made the mistake had he not been fortunate enough to live before the introduction of Competitive Examinations. This Kungurat business was in fact a competitive examination in beauty; total marks attainable 24 ; no candidate to pass who did not get 20 or Carat expresses n÷24, not any absolute value.

21.

Apart from the mode of valuation, it appears that a like system of selection was continued by the Ming, and that some such selection from the daughters of the Manchu nobles has been maintained till recent times. Herodotus tells that the like custom prevailed among the Adyrmachidae, the Libyan tribe next Egypt. Old Eden too relates it of the "Princes of Moscovia." (Middle Km. I. 318; Herod. IV. 168, Rawl.; Notes on Russia, Hak. Soc. II. 253.)

CHAPTER IX.

CONCERNING THE GREAT KAAN'S SONS.

THE Emperor hath by those four wives of his twenty-two male children; the eldest of whom was called CHINKIN for the love of the good Chinghis Kaan, the first Lord of the Tartars. And this Chinkin, as the Eldest Son of the Kaan, was to have reigned after his father's death; but, as it came to pass, he died. He left a son behind him, however, whose name is TEMUR, and he is to be the Great Kaan and Emperor after the death of his Grandfather, as is but right; he being the child of the Great Kaan's eldest son. And this Temur is an able and brave man, as he hath already proven on many occasions."

VOL I.

Y

The Great Kaan hath also twenty-five other sons by his concubines; and these are good and valiant soldiers, and each of them is a great chief. I tell you moreover that of his children by his four lawful wives there are seven who are kings of vast realms or provinces, and govern them well; being all able and gallant men, as might be expected. For the Great Kaan their sire is, I tell you, the wisest and most accomplished man, the greatest Captain, the best to govern men and rule an Empire, as well as the most valiant, that ever has existed among all the Tribes of Tartars."

NOTE 1.- Kublai had a son older than CHIMKIN or CHINGKIM, to whom Hammer's Genealogical Table gives the name of Jurji, and attributes a son called Ananda. The Chinese authorities of Gaubil and Pauthier call him Turchi or Torchi, i.e. Dorje, "Noble Stone," the Tibetan name of a sacred Buddhist emblem in the form of a dumb-bell, representing the Vajra or Thunderbolt. Probably Dorjé died early, as in the passage we shall quote from Wassáf also Chingkim is styled the Eldest Son Marco is probably wrong in connecting the name of the latter with that of Chinghiz. Schmidt says that he does not know what Chingkim means.

Chingkim died in the 12th moon of 1284-5, aged 43. He had received a Chinese education, and the Chinese Annals ascribe to him all the virtues which so often pertain in history to heirs apparent who have not reigned.

"When Kublai approached his 70th year," says Wassáf, "he desired to raise his eldest son Chimkin to the position of his representative and declared successor, during his own lifetime; so he took counsel with the chiefs, in view to giving the Prince a share of his authority and a place on the Imperial Throne. The chiefs, who are the Pillars of Majesty and Props of the Empire, represented that His Majesty's proposal to invest his Son, during his own lifetime, with Imperial authority, was not in accordance with the precedents and Institutes (Yasa) of the World-conquering Padshah Chinghiz Khan; but still they would consent to execute a solemn document securing the Kaanship to Chimkin, and pledging themselves to lifelong obedience and allegiance to him. It was however the Divine Fiat that the intended successor should predecease him who bestowed the nomination. . . . . The dignitaries of the Empire then united their voices in favour of TEIMUR the son of Chimkin."

Teimur, according to the same authority, was the third son of Chimkin; but the eldest, Kambala, squinted; the second, Tarmah (pro

perly Tarmabala for Dharma-phala, a Buddhist Sanscrit name) was rickety in constitution; and on the death of the old Kaar. (1294) Teimur was unanimously named to the Throne, after some opposition from Kambala, which was put down by the decided bearing of the great soldier Bayan. (Schmidt, p. 399; Demailla, IX. 424; Gaubil, 203; Wassáf, 46.)

NOTE 2.-The Chinese Annals, according to Pauthier and Gaubil, give only ten sons to Kublai, at least by his legitimate wives; Hammer's Table gives twelve. It is very probable that xxii. was an early clerical error in the texts of Polo for xii. Dodeci indeed occurs in one MS. (that cited by Baldello as Magliab. II.), though not one of much weight.

Of these legitimate sons Polo mentions, in different parts of his work, five by name. The following is the list from Hammer and D'Ohsson, with the Chinese forms from Pauthier in parentheses. The seven whose names are in capitals had the title of Wang or King" of particular territories, as M. Pauthier has shown from the Chinese Annals, thus confirming Marco's accuracy on that point.

I. Jurji or Dorjé (Torchi). II. CHIMKIN or CHINGKIM (Yutsung, King of Yen, i.e. Old Peking). III. MANGALAI (Mankola, "King of the Pacified West"), mentioned by Polo (infrà, chap. xli.) as King of Kenjanfu or Shensi. IV. NUMUGAN (Numukan, "Pacifying King of the North") mentioned by Polo (Bk. IV. chap. ii.) as with King George joint leader of the Kaan's Army against Kaidu. V. Kuridai (not in Chinese List). VI. HUKAJI (Hukochi, “King of Yunnan ") mentioned by Polo (infrà, chap. xlix.) as King of Carajan. VII. AGHRUKJI or UĶURUJI (Gaoluchi, "King of Siping" or Tibet). VIII. Abaji (Gaiyachi?). IX. KUKJU or GEUKJU (Khokhochu, "King of Ning" or Tangut). X. Kutuktemur (Hutulu Temurh). XI. TUKAN (Thohoan, "King of Chinnan "). His command lay on the Tungking frontier, where he came to great grief in 1288, in consequence of which he was disgraced (see Cathay, p. 272). XII. Temkan (not in Chinese List). Gaubil's Chinese List omits Hutulu Temurh, and introduces a prince called Gantanpouhoa as 4th son.

M. Pauthier lays great stress on Polo's intimate knowledge of the Imperial affairs (p. 263) because he knew the name of the Hereditary Prince to be Teimur; this being, he says, the private name which could not be known until after the owner's death, except by those in the most confidential intimacy. The public only then discovered that, like the Irishman's dog, his real name was Turk though he had always been called Toby! But M. Pauthier's learning has misled him. At least the secret must have been very badly kept, for it was known in Teimur's lifetime not only to Marco, but to Rashiduddin in Persia, and to Hayton in Armenia; to say nothing of the circumstance that the name Temur Khaghan is also used during that Emperor's life by Oljaitu Khan of Persia in writing to the King of France a letter which M. Pauthier has himself republished and commented upon (see his book, p. 780).

CHAPTER X.

CONCERNING THE PALACE OF THE GREAT KAAN.

You must know that for three months of the year, to wit December, January, and February, the Great Kaan resides in the capital city of Cathay, which is called CAMBALUC, and which is at the north-eastern extremity of the country. In that city stands his great Palace, and now I will tell you what it is like.

that is to

It is enclosed all round by a great wall forming a square, each side of which is a mile in length; say, the whole compass thereof is four miles. This you may depend on; it is also very thick, and a good ten paces in height, whitewashed and loop-holed all round. At each angle of the wall there is a very fine and rich palace in which the war-harness of the Emperor is kept, such as bows and quivers,' saddles and bridles, and bowstrings, and everything needful for an army. Also midway between every two of these Corner Palaces there is another of the like; so that taking the whole compass of the enclosure you find eight vast palaces stored with the Great Lord's harness of war. And you must understand that each Palace is assigned to only one kind of article; thus one is stored with bows, a second with saddles, a third with bridles, and so on in succession right round.2

The great wall has five gates on its southern face, the middle one being the great gate which is never opened on any occasion except when the Great Kaan himself goes forth or enters. Close on either side of this great gate is a smaller one by which all other people pass; and then towards each angle is another great gate, also open to people in general; so that on that side there are five gates in all.3

Inside of this wall there is a second, enclosing a space that is somewhat greater in length than in breadth. This enclosure also has eight palaces corresponding to those of