about 5 feet 1 inch, the surface of the wall available for the painted dado could accommodate a total of thirty lunettes with the corresponding number of portraits within them. If there ever was an entrance or a window reaching down low on the west side, where I found the passage wall completely broken, the total number of lunettes and portraits would have been limited to twenty-eight. Out of these, as we shall see, twenty could still be traced, though some only in very imperfect outlines.
The raised portions of the wreath on either side of the supporting figures had a fastening ribbon, usually double and of greenish-buff colour. Below and above each ribbon the wreath was decorated with triplets of small balls representing flowers, alternately pink and white. Ribbons in exactly the same position are also found on festoons of Gandhdra relievos,1 and examples of flowers fixed to the wreaths would probably be traceable on closer examination. At the bottom of the drooping curve of the festoon and partly covering it there was always painted a curious oval object resembling a wide turned-down jar. It can be seen in part in Figs. 134, 135, but had generally suffered badly through effacement. Its colours, where they could still be made out, varied much. There was solid black edged with pink dots resembling small flowers ; crimson ; white ; bright red ; slaty grey, and in one case pink filled with small seed-like dots in black. The true interpretation is furnished by a comparison with the festoons of the Gandhdra relievo, where we very often see the drooping portions weighed down by big fruits suggesting gourds or melons.2 The background of the dado was of a pale buff colour. Within the lunettes, on either side of the head and between it and the festoon, was painted a rosette, six- or seven-petalled, of the conventional type common in late Hellenistic art (see M. v. vi in Plate XLIV). The colours of these rosettes were always green on one side and pink or white on the other.
The youthful supporters of the festoon were always standing figures of three-fourths length. Among them two distinct types were represented, both unmistakably classical in origin and design, and treated with an animated freedom and a variety of individual expression which clearly bear the impress of Hellenistic art. To one type belonged the wingless anmorini, of which we see characteristic specimens in Figs. 134, 138, 140. The other was represented by youthful draped figures wearing the Phrygian cap, and, in spite of a certain girlish cast of face, unmistakably recalling the figure of the young Mithras, whose worship had spread from the Iranian marches of the Hellenistic East to the most distant provinces of the Roman empire. Specimens of this type are seen in Figs. z 35, 136, 137, 143, and one particularly good, though poorly preserved, head is found in the fresco panel M. v. x (Plate xLlv). It is of interest to note that among the wingless aniorini which carry the festoons of the Gandhara relievos draped figures of children are not altogether unknown.3 Another point of contact with the arrangement of these sculptured panels is the way in which the festoon-carrying figures of the dado usually faced each other in pairs.4 Their function necessarily implied a certain limitation of the poses. Yet there was an obvious effort made to introduce a pleasing variety, not merely in facial features and the expression of the eyes, but also in the position of the arms and hands. Thus some ftulti were shown with both arms passed behind the festoon and their fingers resting on its edges (Fig. 138) ; others carried the festoon only on one shoulder, usually the left (Figs. 134, 135, 40, 143), while one or two among them had a small appropriate object, such as a grape or Patera (M. v. x, Plate xLIV), placed in their right hand. It was, no doubt, also due to the same love of variety that the amorini and Phrygian-capped youths were either introduced alternately or else, when placed in pairs, always made to face each other.
This happy association of wingless Erotes and childlike Mithras figures seemed symbolic of the
1 Cf. e.g. Foucher, L'ail du Gandhdra, i. Fig. 118. 4 Cf. Figs. 135-36, 138-40; Foucher, L'ail du Gan-
2 See below, Pl. ML.I; Foucher, loc. cil., i. Figs. 117, 118. dhdra, i. p. 24o.
3 Cf. the panel on left of PI. XI..1.