494 THE RUINS OF AK-SIPIL AND RAWAK [Chap. XIV
sand cover which along this wall was only from 5 to 7 ft. in height, and several of the torsos collapsed after excavation. As seen in Plate X IV. a, b, c, the extant portions of these figures closely resemble the colossal Buddha images which were found along the same wall-face north of the gate and are somewhat better preserved (R. xxix, xxx, xxxi, xxxiii, xxxiv, xxxvi, xxxviii, xl, xli ; see Plates XIV. d, XV. a, b, c). Judging from the latter, we may safely assume that the posture of all of them was the ` Abhayamudrâ'. From the colossal Buddhas on the south-west wall this series seems to have differed only in minor points, such as the bold outward sweep given to the drapery folds below, on the left proper (see e. g. R. xx, xxiii, xxiv, &c.), and the pose of the left hand, which, as shown in R. xxxvi, xl, and xli (see Plate XV. b, c), was probably a little more detached from the drapery of the robe. The height of the statues up to the navel appears to have been about 5 ft. A number of smaller reliefs found by their side probably represent later additions, perhaps donations of individual worshippers. R. xxii is a small male figure, found headless, surmounted by a plaque forming a nimbus after the fashion of that behind R. ii. R. xxiv. a (see Plate XIV. b) is a richly draped figure, also headless, measuring about 3 ft. to the neck, and probably representing a Bodhisattva. The folds of the robe below are arranged in wave lines, while remains of ornaments appear hanging from the neck and around the arms.
As seen in Plate X IV. b, small broken Buddha figures, evidently replicas of those in the vesicas of R. xii, xiii, were found placed between or near the feet of several of the colossal images, as noted above in connexion with similar deposits near R. xi. In addition to these, three small plaques representing seated Buddhas were found near R. xxv. Two of them were replicas of the corresponding appliqué reliefs in the nimbus of R. ii and R. iv. The third (R. xxv. 3), measuring about 7 by 5 in., is a replica of R. 2 (see Plate LXXXII), a small but well-executed relief, found in the débris of the south corner, showing Buddha seated in the ` Dhyânamudrâ' attitude with a small tree, evidently meant for a Ficus indica, behind the right shoulder. These two pieces are of particular interest as showing how closely the treatment of the drapery also in seated Buddha figures agrees with that of the corresponding Gandhara models. The good proportions of the features recall the best type of Gandhara work, as is easily realized on comparing this small plaque with Gandhara reliefs representing the same subject 13.
The gate was flanked on either side on the inner wall-face by a pair of standing figures, worked almost in the round and about life-size, R. xxvii, xxviii (Plate XIV. c, d). These deserve special attention as the only images on the walls of the Vihâra court which do not present themselves in the orthodox garb of Buddhas or Bodhisattvas. Though the upper portions of their bodies had broken away and were recovered only in fragments, there can be no doubt as to their being intended to represent Dvarapâlas or ` Guardians of the Gate '. It seems very probable that the four guardians at the Rawak Vihâra gate, like those sculptured at the entrance of many a Buddhist shrine in India, were meant for Yaksas, and in particular for the four Yaksa chiefs (Kubera, Virûdhaka, Virtipâ,ksa, and Dhrtarâstra) who as Lokapâlas or ` World protectors ' play so conspicuous a part In, the Buddhist mythology of India, as well as of Tibet and the Far East 14. Buddhist iconography invariably gave to them, as to Yaksas generally, a wholly human and quasi-secular form, as we noted when identifying with Kubera the armed figure represented by the side of the main image in the small temple cella of D. n at Dandân-Uilig13. It fully accords herewith that the figures at the Rawak gate exhibit