212 ALONG THE PERSIAN GULF COAST [Chap. VII, Sec.
and at the kilns of Tiz, which, owing to its close resemblance to wares from Samarra and Brahmanâbâd, can with Mr. Hobson be safely dated as from the
ninth and tenth centuries.8 The same type of colouring appears in bands, but
without incised patterns, in pieces like 122. A peculiar kind of design in brown and greyish-green (118, 130, 132) under light creamy glaze appears on stone
ware of superior body, which in one piece (113 ) has a crackled surface and might possibly be Chinese. Painted stripes in reddish-brown are seen in some unglazed fragments (119) .
Decorated pottery unpainted and- unglazed is represented by ribbed pieces like 73 ( Pl. XXVI ) and by others where the geometrical patterns, whether raised
or sunk, were obviously produced from moulds. In them, as well as in some in-
cised pieces, the workmanship is distinctly less careful than in the corresponding ware from the sites of Old Hormuz. There still remain to be noticed the plenti-
ful fragments of porcelain, most of them plain ware in white or shades of grey.
Specimens of decorated porcelain, like 117, 121, 134, 159, are clearly Chinese, but these still await approximate dating. A few porcelain fragments of coarse
design look much later and may be of Persian manufacture. This is likely to be the case also with the small and neatly executed fragment 147, showing a design manifestly Persian. Among the glass fragments varying in colours from white to black the tubular bead 190 (Pl. X) deserves mention for its inlaid decoration with groups of circlets.
I cannot conclude this account of the observations made at Sirâf without briefly referring to two facts which, though negative, yet deserve attention. One
is that our examination of the surface remains of the site, extensive as these are,
did not bring to my notice any indication of Sirâf having served as a port or having been occupied by a settled population in pre-Muhammadan times. The
other is the absence of any special advantages at the site itself which would have
recommended it for selection as an emporium for maritime trade. The unproductive character of the ground near it, the extreme limitation of the avail-
able building space, and the inadequate protection afforded by its open roadstead had, as the accounts of the Arab geographers show, impressed those who visited Sirâf while it was still a much-frequented port, quite as much as they must impress us now. This striking incongruity between the local features of Sirâf and the important part it had played in the trade of the Gulf region during early Islamic times provided a special inducement for me to gain some acquaintance with the tracts which form the hinterland of this barren coast, and with the routes which passed through them and once connected Sir-if with the old economic centres of Persia.
8 See above, pp. 85 sq., 91 sq., and Mr. Hobson's Appendix A, pp. 244 sq.