Sec. iv] FIRST FINDS OF ANCIENT MANUSCRIPTS 263
and with more elaborate mouldings. The photograph shows that it rested on a low circular base decorated with lotus-petals. The whole front portion of the pedestal and base was found to have been destroyed, no doubt in the course of some ` treasure-seeking ' operations, and of the statue which once had stood on it not even fragments survived. It was probably due to the same reason that no painted panels were found here. Near the north wall, however, some small reliefs in plaster of Paris were found, showing that the upper portions of the walls had once borne decorations similar to those of D. i and D. n. One of these reliefs (D. xII. i), seen in Plate LV 1, and originally coloured, represents a flying female figure, probably a Gandharvi, holding in her outstretched hands a festoon cloth. The other pieces are replicas of the Gandharvi figure D. xir. 5, illustrated in the same plate, which is shown here as rising from a large-petalled lotus. We have met with this motive already among the stucco decorations from D. i.
Of the frescoes which had covered the cella walls naturally very little remained. Each wall-face, except the one to the north, seems to have borne three life-size paintings of Buddhas or Bodhisattvas surrounded by aureoles and standing on open lotuses. The photograph reproduced in Plate V shows what remains of this mural decoration at the western end of the south wall. The triangular space left between the lowest portions of two aureoles is filled by a well-draped female figure with her head turned upwards and holding in her left a lyre. She is shown standing on a lotus, and may thus be meant for a divine attendant. The small figures, male and female, seen at the bottom of the wall, are typical representations of worshippers. Very little remained of the colouring of these _frescoes apart from the outlines of the figures, usually drawn in a kind of terra-cotta colour over the greyish-white distemper.
The clearing of the south-east corner revealed, close to the floor, manuscript remains of considerable extent, but unfortunately decayed into an almost unrecognizable state. The first indication of such remains was furnished by the fragment of a wooden vessel, D. mi. 6, about 7â by 3 in., to which were adhering, both inside and outside, layers of completely rotten paper. Writing of upright Gupta characters, arranged apparently in four lines, can just be distinguished on the blackened surface. A few inches below this I came upon a consolidated mass of ` sand,' rate loess dust, extending over a space of about i t by 6 in. and firmly adhering to the floor. In it were embedded relatively thick layers of almost completely perished manuscript leaves. The latter, together with the overlying brittle crust of ` sand,' broke at the slightest touch. It proved equally impossible to disengage the leaves or to remove the whole layer. I succeeded in slicing the latter off the floor but only in fragments, and the separation of the paper-flakes proved an impossible task even in the British Museum. The few characters distinguishable on the surface show bold Gupta writing similar to that seen in Plate CVI I. Embedded in the layer were two thin wooden boards, about 92 in. long and 4 in. wide (see Plate CVI), which had undoubtedly served as covers for one of the Pothis that had supplied these perished leaves. Flakes of paper still adhere to them in layers. A third board, completely decayed, which lay at the bottom touching the floor, had evidently belonged to a second Path'.
The entire decay of these manuscripts can be explained only by the damp to which they must have been exposed for some time while lying on the floor of the shrine. The observation is of interest as showing that even after the abandonment of the settlement there followed a period when water must have reached it, at least periodically, in quantities sufficient to spread moisture to the basement of buildings. If the manuscripts had come to lie above a layer of the drift-sand which invaded the building, or if all supply of water had ceased immediately after their deposition on the floor, their state of preservation would have probably differed little from that of the manuscripts found in D. III.
| || || || || || |