National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Sec. iii]   REMAINS OF THE PAINTED FRIEZES OF M. HI   505

At the feet of the teacher, on the right, there appears a smaller figure with hands folded and raised in adoration. Its dress is similar in arrangement, and shows the same draping after classical models. But the princely character of this worshipper is distinctly indicated by a curious white conical hat or turban, adorned with red rings above and two lunette-shaped dark red flaps below. We shall meet with this peculiar head-dress again in the paintings of the rotunda M. v, where it certainly marks royal personages. Neither in the sculptures of Gandhara nor in the painted or plastic remains of later Buddhist shrines of Eastern Turkestan have I been able to trace it, and its origin remains at present quite uncertain. Could it possibly be a reminiscence of princely insignia used for a time in one of the more westerly territories, such as Bactria or Sogdiana, through which this Central-Asian adaptation of Graeco-Buddhist art must have passed into the Tarim Basin ? However this may be, we can have no doubt as to the origin of the iconographic convention which made the painter represent this adoring figure disproportionately small, in spite of its princely rank. It is the same convention which causes the Gandhara sculptors very frequently to give greater stature to Gautama where, in his Bodhisattva state or as Buddha, he appears among minor personages in the same sacred scene. It was equally familiar to late Hellenistic art when representing emperors or the central figure of Christian story.' Of another attendant figure on the left, also of smaller size, the fragment only preserves parts of the left arm and knee. The representation of two tanks, or possibly railed terraces, in the foreground conveys the impression that a scene in a palace or royal garden was intended.

The examination of the larger fresco pieces has allowed us to discuss in some detail the chief features characteristic of the painted frieze. Hence, for the smaller fragments, brief notes on points of special interest will suffice. The most curious among these fragments is, perhaps, M. III. 0019 (Plate xLIV), which shows the heads and busts of two small female figures evidently in a pose of worship. Their faces are painted in careful chiaroscuro, and bear the same individualized and animated expression which we shall presently have to note as making the `angel' heads of the dado so attractive. The large almond-shaped eyes vaguely recall a Persian type of beauty such as the head of the princess in the fine Dandin-oilik panel D. x. 4 unmistakably displays.' The fact that the hair is dressed practically in the same way on these heads as on those of the princess and her attendants in the painted panel, with long tresses hanging behind the ears and wavy ringlets in front, suggests that we have here a fashion not of local and contemporary character, but introduced from outside and maintained by artistic tradition. The same arrangement of the hair, with very slight modifications, appears also on the girls' heads which decorate the dado of M. v (Figs. 138-

140, 143).

Another interesting fragment, M. III. 009-10 (Plate xLV), presents us with the bust of a figure, probably female, dressed in elaborate garments of a curious quatrocento look, and wearing flowers on the breast. The head has suffered badly ; but in other fragments, M. III. 0033-34 (Plate xLV), we have replicas of it which enable us to realize better the peculiarity of its type. As all of them approach life size, we must conclude that the figures to which these fragments belonged formed part of a higher frieze which is likely to have extended into the vaulted portion of the walls. The fragmentary male head, M. in. 008 (Plate xLV), represents quite a different type and deserves mention on account of its rapid but very effective painting, which clearly reveals its methods as described in the list. The presence of life-size figures in what must necessarily have been upper parts of the wall decoration is proved by other fragments also, such as M. III. 0035 (Plate xLV), which shows a remarkably well-painted hand ; M. III. 0039-40, 0052, 0063. The pieces M. III.

$ Cf. Foucher, L'arl du Gandhdra, i. p. 603 ; Grümvedcl-   9 See Ancient Kho/an, i. pp. 259 sq., 300 ; ii. Pl. LXIII.

Burgess, Buddhist Art in India, p. 138.

1974

3T

Figure of royal

worshiprer.

Heads of female figures.

Minor fragments of upper friezes.