A chinese mason will give preternatural length to the arms of a Kuan-yin so as to include the pedestal in the proportions of the design. And this eye for composition has made coiffure and coronet such integral parts of these two heads that without them the faces would lose half their distinction.

Readers of my The Land of the Great Image will remember how the Portuguese friar, Brother Manrique, came in 1630 over the rainy mountains to Mahāmuni in Arakan, the most celebrated Buddhist shrine at that time in Further India. The King of Arakan, Thiri-thu-dhamma, was visiting the shrine, having travelled by house-boat from his capital, Mrauk-u, the Monkey's Egg, which lay on the Eiver of Jems some 40 miles distant. The friar was received in audience at the foot of the sacred hill, Sirigutta, upon the top of which stood the Mahāmuni, one of the portrait images of the Buddha supposed to have been cast in his lifetime, another being located in Tavatimsa, a paradisal region inhabited by the displaced Hindu gods, who, having been converted to the truth, desired to possess in their territory an image of the Blessed One to worship.

The three inscriptions shown in Plates X and XI were copied by Mr. Harold Ingrams in the course of a journey from Bir 'Ali (? the ancient Cana or Kana) on the south coast of Arabia to Shabwa and Al 'Abr during the last ten days of April, 1939. At an early stage of that journey he passed through the great wall of Wadi Bana, in whose gateway stands the celebrated record known as the “Obne inscription” (R.E.S. 2687) and first discovered by Adolph von Wrede in 1843. Beyond this point Mr. Ingrams noted or collected a number of graffiti, pictures, and “inscribed stones”, none of which are at present available for examination. But it was not till he reached the pass of 'Aqaba Futura (3,500 feet above sea-level and about a week's journey from the wall) at the head of Wadi 'Anna, flowing westward across the plateau to Shabwa and eventually into Wadi Hadhramaut, that he found his first considerable inscription, No. 1 (Plate X). The second inscription, No. 2 (Plate XI), was found on the final pass, 'Aqabat 'Aqaiba (2,750 feet), leading down from the plateau to the Shabwa plain, while the third of this group, No. 3 (Plate XI), was found near the desert well of Al 'Abr, 56 miles north of Shabwa.

There are two rulers in Malaya who claim descent from a Bichitram (? = Vicitram), reputed kinsman of the Sri Maharajas of Sri Vijaya, the Buddhist empire (fl. a.d. 750–1350) that extended over Sumatra and Northern Malaya and for a while Java. The name Bichitram is whispered into the ear of every Perak Sultan at his enthronement as that of the ancestor of the Perak (and old Malacca) dynasty. And Bichitram, according to the Sejarah Melayu, was brother of the first king of Palembang (= Sri Vijaya) and Singapore, and was himself ancestor of the Minangkabau line, from which the Yang di-pertuan of Negri Sembilan claims descent.

In an excursus to his article on “The Early Sasanians” in BSOAS., xi, 42 sqq., S. H. Taqizadeh has given a full and illuminating discussion of the Manichæan two-day fasts which preceded the feast of the Bema. All hitherto known about these fasts is contained in a passage in the K. al-Fihrist, 333, 28–334, 1, which runs as follows:—

As regards the fasting—

(1) when the sun is in Sagittarius and the moon is full, they fast two days without break,

(2) then when the New Light appears, they fast two days without break,

(3) after this, they fast two days when the moon is full (and the sun is) in Capricornus,

(4) then when the New Light appears and the sun is in Aquarius and eight days have passed of the (lunar) month, they fast for thirty days, but break the fast each day at nightfall.

The date of the composition of Jāmī's “Chain of Gold” is usually given, for instance, by E. G. Browne in his Literary History of Persia, iii, p. 516, as a.h. 890. This is certainly the date of the second of its three daftar, for the poet states the fact plainly in its penultimate bayt, and the addition of 890 in figures in a few MSS., notably the famous Leningrad autograph, Rosen No. 80, adds nothing to this evidence. In very few MSS., however, to judge from the many catalogues which I have searched, is there any other indication of the date of any part of the poem, but I have found descriptions of two, which contain matter conflicting with the extension to the first daftar of the date of the second. The more important is No. 185 in vol. ii of the Catalogue of the Oriental Public Library at Bankipore (Patna, Bihar), published in 1910. This volume is missing from the two Cambridge copies of the Catalogue, but Dr. L. D. Barnett has kindly transcribed the entry for me from the copy in the Library of the School of Oriental and African Studies. It runs as follows: “No. 185. Silsilat al-dhahab, daftar 1, and minor lyrical poems, supposed to be in the author's own hand. MS. not dated but contains note and chronograms on birth of Jāmī's son, Shauwal 9, 882.” The Secretary and Librarian of the Oriental Public Library, Khan Saheb K. Hassan, has since called my attention to an article by Mr. M. Mahfuz-ul Haq in Islamic Culture, i, 1927, pp. 608 ff., entitled “Jami and his Autographs”.

Sometime in 1921 I was sitting in the shade of a mango tree on the top of a bluff overlooking the mangrove-filled creek of Chake-Chake in Pemba, called by the Arabs the Green Island. I was the guest of the Sheikh of the Mauli tribe, Sheikh Salim bin Khalif, at his clove plantation of Kaole. There were a lot of Maulis, who came from the Wadi al Ma'awil in 'Uman, in positions of authority in Pemba, and with me sat Sheikh Abdulla Mbaruk al Mauli, the Mudir of Chake-Chake. We were watching the little ship Khalifa far out at her anchorage. Presently one of my boatboys climbed up the hill and handed me a large bag of mail brought from the ship. In it was a parcel. I opened it and took out the two fat volumes of Zanzibar, City, Island, and Coast. That was my introduction to the works of Sir Richard Burton.