Sec. ii] THE STUPA CELLA M. ni AND ITS WALL-PAINTINGS 5oi
conclusive. Among all the pieces of painted plaster from the friezes which came to light in the débris of the circular passage and which are recorded in the Descriptive List below,10 there is not one indicating a division of the wall-paintings into distinct panels or fields by decorative architectural motifs or otherwise. This is in full agreement with the surviving part of the frieze in the southern hemicycle of the cella walls of M. v. There a series of independent but consecutive scenes unfolds itself combined into one unbroken composition to illustrate the Jâtaka legend of King Vessantara.
The point is one of considerable interest and deserving of special notice for two reasons. The first is that this continuity of composition in the wall-paintings is in striking contrast with the treatment of similar legendary subjects in the plastic art of Gandhdra. There we find that the relievo representations of scenes which form part of the same legendary cycles are, at least in the vast majority of cases, divided into separate panels or compartments, even though they are there also ranged into regular friezes.rt We are not concerned here with the origin and explanation of this peculiarity in the design of the sculptured friezes of Gandhdra. Unfortunately the total absence of pictorial remains of Graeco-Buddhist art makes it impossible for us to ascertain whether it applied there also to wall-paintings. But it is certainly noteworthy—and that is the second reason for my calling attention to the point—that the combination of several scenes into one continuous field, which is well known to later classical art, according to a very competent authority originated in the Hellenistic Near East.12 Even to the system which is traceable there of explaining the different scenes depicted in this fashion by means of short painted inscriptions the frieze of M. v, as we shall see, offers an exact parallel.
Comparison with the frieze of M. v is also helpful with regard to the second point of interest I have mentioned above. That in any case the lower of the friezes in M. III is likely to have contained scenes taken from a Buddhist legend is indicated in general by the character of the large piece, M. in. 003 (Plate XLII), to be discussed presently, and that of the panel M. in. 002 (Plate XLIII). This shows figures drawn to the same scale, and may therefore be assumed to have belonged to the same frieze. Now it is significant that among the fragments recovered there are several which certainly belonged to replicas of figures appearing in these larger pieces. Thus we have the rows of Buddhist monks representing disciples in M. iii. 003 repeated with exactly the same treatment in M. ni. oo5 (Plate xLlv). The head of the princely worshipper, seen on the right of M. ni. 002 ( Plate XLIII), is found repeated with exactly the same type, treatment, and head-dress no less than
five times (M. 006, 0031-32, 0037, 0056 ; Plates XLIV, XLV). Of the head seen in the interesting
representation of a well-dressed personage, M. III. 009-10 (Plate XLV), we find similarly obvious replicas in the two fragments, M. III. 0033-34 (Plate XLV). It is clear that in the representation of scenes belonging to the same story there must arise constant necessity to introduce certain principal actors again and again, and to indicate their identity by close reproduction of type, treatment, etc. We find this necessity strikingly illustrated by the Vessantara Jâtaka frieze of M. v, and hence the presence of replicas also in the frieze of M. in, demonstrated by the fragments just quoted, may well be accepted as an indirect proof that the subject was here, too, a story from the life of Buddha in his last or an earlier incarnation.
10 See below, pp. 539 sqq.
" Cf. Foucher, L'art du Gandhdra, i. pp. 182, 266 sqq., where references will be found to the abundant relievo friezes illustrating the above observation. For rare instances of a continuous composition, see ibid. i. p. 603. That this arrangement of relievo scenes in compartments, as it were
in metopes ', was not primarily due to technical necessities arising from the size of the stone materials to be worked, is obvious from the friezes which decorate the bases of very small Stûpas and are carved on single slabs; see e.g. ibid., Figs. 70-72.
12 Cf. J. Strzygowski, Orient oder Rom, p. 39.