NOTE 2.—­Valentyn appears to be repeating a native tradition when he says:  “In old times the island had, as they loosely say, a good 400 miles (i.e.  Dutch, say 1600 miles) of compass, but at the north end the sea has from time to time carried away a large part of it.” (Ceylon, in vol. v., p. 18.) Curious particulars touching the exaggerated ideas of the ancients, inherited by the Arabs, as to the dimensions of Ceylon, will be found in Tennent’s Ceylon, ch. i.  The Chinese pilgrim Hiuen Tsang has the same tale.  According to him, the circuit was 7000 li, or 1400 miles.  We see from Marco’s curious notice of the old charts (G.T. “selonc qe se treuve en la mapemondi des mariner de cel mer”) that travellers had begun to find that the dimensions were exaggerated.  The real circuit is under 700 miles!

On the ground that all the derivations of the name SAILAN or CEYLON from the old Sinhala, Serendib, and what not, seem forced, Van der Tuuk has suggested that the name may have been originally Javanese, being formed (he says) according to the rules of that language from Sela, “a precious stone,” so that Pulo Selan would be the “Island of Gems.” [Professor Schlegel says (Geog.  Notes, I. p. 19, note) that “it seems better to think of the Sanskrit sila, ‘a stone or rock,’ or saila, ‘a mountain,’ which agree with the Chinese interpretation.”—­H.C.] The Island was really called anciently Ratnadvipa, “the Island of Gems” (Mem. de H.Y., II. 125, and Harivansa, I. 403); and it is termed by an Arab Historian of the 9th century Jazirat al Yakut, “The Isle of Rubies.” [The (Chinese) characters ya-ku-pao-shih are in some accounts of Ceylon used to express Yakut. (Ma-Huan, transl. by Phillips, p. 213.)—­H.C.] As a matter of fact, we derive originally from the Malays nearly all the forms we have adopted for names of countries reached by sea to the east of the Bay of Bengal, e.g. Awa, Barma, Paigu, Siyam, China, Japun, Kochi (Cochin China), Champa, Kamboja, Maluka (properly a place in the Island of Ceram), Suluk, Burnei, Tanasari, Martavan, etc.  That accidents in the history of marine affairs in those seas should have led to the adoption of the Malay and Javanese names in the case of Ceylon also is at least conceivable.  But Dr. Caldwell has pointed out to me that the Pali form of Sinhala was Sihalan, and that this must have been colloquially shortened to Silan, for it appears in old Tamul inscriptions as Ilam.[1] Hence there is nothing really strained in the derivation of Sailan from Sinhala.  Tennent (Ceylon, I. 549) and Crawford (Malay Dict. p. 171) ascribe the name Selan, Zeilan, to the Portuguese, but this is quite unfounded, as our author sufficiently testifies.  The name Sailan