The island is a little paradise of fruit-trees, palms and corn, in the middle of which is a village of some thirty houses built in the heaped-up ruins of the castle. From among the houses springs a tall and beautiful minaret, octagonal in plan (Fig. 56). Its height is broken by eight rows of niches, each face of the octagon bearing in alternate storeys a double and single niche, all terminating in the cusped arch which is employed at Raḳḳah. Some of the niches are pierced with windows to light the winding stair. The tower rises yet another two storeys, but the upper part is of narrower diameter, and the windows and niches are covered with plain round arches. At the northern end of the island the walls and round bastions of the fortress stand in part, but they are not very ancient. Ibn Khurdâdhbeh, who is the first of the Mohammadan geographers to mention ’Ânah, says only that it is a small town on an island;[63] in Abu’l Fidâ’s time it was still confined to the island;[64] Rauwolff (1564) notices the town on the island and the town on the right bank;[65] Yâḳût (1225) speaks of the castle, but the walls which I saw cannot be as old as his day. The minaret may belong to a different period, and de Beylié places it in the earliest centuries of Islâm.[66] I think that there was probably a fortress on the island long before the first written record which has come down to us, but I was close upon a generation too late to see the remains of it. From two informants in ’Ânah I heard that there had been big stone slabs at the northern end of the island “with figures of men upon them and a writing like nails,” but they had fallen into the water within the memory of the older inhabitants and had been washed away or covered by the stream. This tale of cuneiform inscriptions would not in itself be worth much, but while I was examining the minaret, a villager brought me a fragment of stone covered with carving in relief which was unmistakably Assyrian. I asked him whence it came, and he replied that it had formed part of a big stone picture which had fallen into the river. I bought from him a broken bowl inscribed with Jewish incantations of the well-known type.[67] The island was once connected with both banks by bridges. There are some traces of the section that led across to the Jezîreh, and many piers of the Shâmîyeh bridge stand in the river. Though these piers no longer serve the purpose for which they were intended, they are still put to use, for the inhabitants of the island spread nets between them, and the fish swimming down with the current are entangled in the meshes and so caught (Fig. 52). We pulled up one of the nets as we passed, and it produced two large fish which I bought for a few pence. It is curious that the Bedouin neglect the ample supply of food with which the river would furnish them; in spite of frequent inquiries we had never found fish in their tents.