war. On the occasion of Chêng Ho's second visit in 1415, Sukanla claimed a share in the presents, and attacked the Chinese, who, supported by the legitimist troops, defeated him, and drove him to Nan-puh-li state (Lambri).

In 1434-35 the King sent his brother (? brothers) Ha-li-chi-han (? and) Ha-ni-chê-han to Court-it is not clear whether there were one or two men; but Ali Jehan is manifestly one name intended, and this one died " 'greatly regretted" at Peking. The King, being now very old, abdicated in favour of another son, A-puh - sai - yih - ti (evidently Abu Saïd), and China in due course confirmed the arrangement. Between that date and the arrival of the Portuguese there was only one mission, and that apparently a "bogus" one, in 1486: the rest of the Chinese information seems to be mere hearsay. They tell a long story about a wily slave having persuaded his master the commander-in-chief to assassinate the King, after which the slave in turn assassinated his master, and changed the name of the state to A-ch'i (Atjeh, or Acheen).

As to the Chinese yarn about the fisherman, it is curious to compare it with the Malay legend about Mara Silu, a fisherman, being converted to Islam, adopting the name Malik-al-Sálih, and assigning Samudra to one son Dháhir, with Pasei to the other son Mansúr: unfortunately, the dates will not suit at all. The slave dynasty of Acheen may or may not be the power now ruling, which is strong enough, any way, to defy all the efforts of the Dutch. Kollewijn says that, when the first Portuguese landed in Sumatra (1506), Acheen was the leading state.

As to the other states in Sumatra, we have seen that in 1309 envoys were sent to Pa-sih; but, apart from the fact that there is nothing to show where it was, this Pa-sih, even if in Sumatra, could not then have been of much importance it might just as well be Pasig in Luzon, or Passir in Borneo, for the first Mussulman state of Pasei in Sumatra was scarcely yet formed.

The Ming Records say that the Frank adventurers Pedro, "Sushili," etc., after plundering Malacca and other states, consoled themselves for the repulse of the mission to Peking by sailing with five ships to attack Pa-si; and finally it winds up the account of the Franks (Portuguese) by saying that "they swept the seas in such a way that neither Malacca, Pasi, nor Luzon could attempt to cope with them." But the ancient Chinese map of Sumatra, discovered by Mr. Phillips, does not mark Pasi at all, which is further evidence that its existence as a state was short.

Nevertheless, Marco Polo says Basman owned the supremacy of the Great Khan, and as he uses the Northern Chinese word Manzi, or man-tsz, to signify the southern half of China, we are pretty safe in assuming that Basman simply means Pasei-man, or the "barbarians of Pasei," and that it ceased to have an independent existence about 1530. The word man is frequently thus tacked on to the name of a foreign country. Colonel Yule says that Malacca, Pasei, and Majapahit (Java) were (about a century before that date) the three chief cities of the Archipelago, a statement almost textually confirmed, as above, by the Ming Records, but for a later date. The "Encyclopædia Britannica

be

wrong in saying that Sumatra was Pedir's dependency in 1506, and that Pedir and Pasei were the only two states of Sultan rank: as we have seen, the Chinese give Sultan rank to Sumatra from 1383, and the Portuguese found Acheen the most powerful state in 1506. Nothing is recorded of Sumatra by the Chinese subsequent to the period of its climax under Jokandar Muda (1607-36), nor of Pasei, Pedir, Johore, Pahang, Quedah, Perak, Aru, Padang, and other states which the "Encyclopædia " says were vassal to Sumatra at this time. The Chinese mention the Siamese as trying to use political influence in Sumatra affairs in 1406-8; but at this time Pahang was, in the eyes of China, independent, and Johore did not yet exist.

I can only find one certain mention of Marco Polo's

Fansur in all Chinese history. In 1418 the King of Java sent back to China some soldiers belonging to the suite of a Chinese envoy who had been wrecked at, or had drifted to, Pan-tsu-rh, whence they had been ransomed by a friendly Java man, who brought them back to Java.

A state called Kuli-Pantsu (the word Kuli elsewhere meaning "Calicut") is stated to have sent tribute between 1403 and 1424, but there is nothing further said by which this state can be identified.

Mr. Phillips' Chinese map, which he believes to be as old as the year 1399 at least, marks Pan-tsu on the northwest coast of Sumatra, and uses the two first characters of Pan-tsu-rh, and the two last of Kuli-Pansuh to write it.

I notice on a modern English map a large island called Panchor off the east coast, opposite Malacca and Singapore ; but whether the Chinese Calicut-Fansur and plain Fansur of the records are, both or either of them, the same place with Panchor, or with the Fansur marked on the Chinese map, I cannot say. Colonel Yule seems a trifle overzealous in twisting bárús (camphor) round to be the same word as pansur (camphor). The word now pronounced polüh (having retrospectively and provably the etymological power barut) is as old as the first Chinese knowledge of the Archipelago, and is used in reference to the best "dragon-brain camphor brought by traders from Java, Sumatra, and other Archipelagan states. It is true two Chinese authorities say the said camphor comes from polüh state (almost the same word). That is a question I shall postpone for discussion under the head "Borneo." Borneo." But Pantsu is a stray word, 1,000 years younger than Barut, with which it cannot possibly have any etymological connection.

Nan-wu-li (south dialect, Lam-bu-li) is marked on the old Chinese map as being at the extreme north-western point of the island it is first mentioned, as already explained, in 1284, 1286, and 1294, the name being on each occasion spelt differently, so far as the eye is concerned, but always so as to produce the same sound. Nothing further is said.

In 1405 the eunuch representing the new Chinese dynasty seems to have sent a lieutenant thither, and to have gone himself in 1408. In 1411 the King sent an envoy, who came along with the envoys of Caïl (India) and Kelantan (Malay) there was tribute again in 1416, and there is an end to it—no details given.

But, strange to say, another state called Nan-pʻuh-li sends tribute in 1412, and again in 1415, 1416, 1418, 1419, 1421, and 1423. Nothing whatever is said about any Chinese envoy ever going there; but in 1429, when the eunuch went on his last voyage, Nan-p'uh-li got a share of the imperial presents.

The explanation of all this is that, when the Lambri mission reached Peking, the Pekingese would have to spell the word in a way to suit their own dialect; but it is curious that, when Chêng Ho was ordered to take the Nan-wu-li envoy back in 1416, nothing was said by his scribes about the Nan-p'uh-li envoy of the same year being the same man or a different man. However, it is absolutely provable, from the extracts translated, that Nan-p'uh-li is Lambri; and as we have already seen that Nan-wu-li is also marked in Lambri's position, it follows that both states are one.

The Ming Records are clear as to Nan-p'uh-li's position: it is three whole days' sail west of [the port of] Sumêntala, which was nine days west of Malacca. If in 1415 the eunuch pursued the rebel Sukanla into Lambri, it must have been adjoining Samudra; but here again it is strange the eunuch, who had himself been in Nan-wu-li in 1408, left no record of its being the Nan-p'uh-li which he approached as a general by land in 1415. It was evidently a very petty state, for we are told "the King and inhabitants are all Mussulmans, only some 1,000 families; little grain produced; fish and shrimps the chief food. King Mohammed Shah sent an envoy with the Samudra envoy in 1412; to the end of the Emperor's reign (1424) they continued to send tribute the King's son, Shah Jehan, also sent an envoy."

Here follows a curious addition: "In the sea to the north-west of them there is a lofty mountain (or island) called Hat Mountain, west of which again is the great sea called Na-muh-li Ocean: ocean ships coming from the west use this as a mark: close by, the water is shallow and produces coral-trees, the highest over 3 feet." The Chinese map certainly points to Hat Island being close to Sumatra, and this is the view taken by Mr. Phillips and Mr. Groeneveldt. Still, it is interesting to notice the significance of a "Sombrero" [i.e., Hat] Channel in English maps amongst the South Nicobars, which in the Chinese map are as near Hat Island as the latter is to Sumatra. Colonel Yule also quotes Rashiduddin, who speaks of "the very large island of Lámúri, lying beyond Ceylon and adjoining the country of Sumutra." On the other hand, Friar Odoric crosses from India to the Lamori country, and thence to Sumoltra in the same island.

Hence, though there is no question about Lambri, it appears both from Chinese and Western accounts that, unless all parties are mistaken, west of Lambri there was a something else, either sea or land, having a name uncommonly like Lambri.

The Chinese histories do not mention Lide, nor does the map give any place which could possibly be mistaken for it; but east of and adjoining Lambri the Ming Records say there was a state subordinate to Sumoltra called Li-fah, lying to the west of Nagur and Sumoltra. This is the exact position of Lide according to De Barros' enumeration of the petty states he visited, and I accept the view of Messrs. Groeneveldt and Phillips (disapproved by Dr. Bretschneider) that some editions (mine for one) print a stroke too much, thus turning tai into fah. As a rule I look very much askance at alleged "misprints" when made use of to explain inconsistencies; but this is one of the cases where the character by its own ambiguity positively invites misprint. Moreover, in a private work called the Ying-yai, the form