PART II of *Ajanta* was published in 1933, and although the material for Part III was collected and sent to the press shortly afterwards, yet the making of blocks and printing of plates, which require a great deal of care, took considerable time, and meanwhile broke out the war, which caused further hindrance in the progress of the work. I am, however, very grateful to His Exalted Highness the Nizam’s Government for financing the production of this Part, and also to the authorities of the Oxford University Press, particularly to Dr. John Johnson, C.B.E., the Printer, who managed to keep the work going under very trying circumstances. The production of the book during war-time is therefore a great achievement of the organizing capacity of the Oxford University Press on the one hand, and the generous policy and patronage of learning of His Exalted Highness the Nizam’s Government on the other.

v3.c11.12

As this Part deals with some caves in which the original frescoes have deteriorated badly and become very dim, but artistic detail can still be copied and utilized for study, I have arranged for the inclusion of the ‘outlines’ of these subjects, which number as many as twenty-three and are shown in nine plates. Further, to avoid misunderstanding regarding the faithfulness of the ‘outlines’ I have also included the photographs of all these dim subjects, having taken them direct from the rock-walls. By this plan, although the number of plates has increased, yet it has served a double purpose: first the photographic survey of the frescoes of Ajanta has been made as complete as possible, and secondly the student of art will now be able to compare the ‘outlines’ with the photographs in regard to the salient features of the drawings. Apart from the reproductions of the inscriptions which are included in the Text volume and whose number is considerable, the Plates portfolio of this Part comprises five key-plans, fifty-two monochrome and seventeen colour-plates, the total number thus exceeding that of the plates contained in Parts I and II. The monochrome plates and the Text volume have been printed by the Oxford University Press, while the making of the blocks of the colour-plates and their printing have been done by Messrs. Henry Stone & Son of Banbury and London, and I have great pleasure in stating that the Directors and the staff of both these presses have not only observed the utmost care in executing the difficult task entrusted to them but have shown a genuine personal interest in making the efforts of the Archaeological Department of Hyderabad successful as regards the publication of this book.

v3.c11.13

This Part also contains a large number of inscriptions, ranging in date from the second century B.C. to the fifth or sixth centuries A.D., and I have to express my deep gratitude to Dr. N. P. Chakravarti for kindly editing all these records in spite of his heavy duties under the Government of India, first as Government Epigraphist and afterwards as Deputy Director-General of Archaeology in India.

v3.c11.14

Lastly, I should thank heartily my esteemed friends Mr. C. E. A. W. Oldham, C.S.I., Hon. Vice-President, Royal Asiatic Society of Great Britain and Ireland, and Dr. L. D. Barnett, Librarian and Lecturer in the School of Oriental Studies, University of London, for their kind and willing help in the matter of the transliteration of proper

v3.c11.15

names, the correction of proofs, and the furnishing of scholarly information from Buddhist literature and other books which are beyond the pale of study of an archaeologist like myself.

v3.c11.16

G. YAZDANI

v3.c11.17

ORANGE GROVE, KHAIRATABAD, HYDERABAD, DECCAN Dated 19th May 1942

v3.c12.1

The Conversion of Nanda (?) . . . . . . 52 *Mānushi* Buddhas . . . . . . . 55 Flying figures (*Apsarases*?) and the Buddha in the Teaching Attitude . 56 The Buddha in the Teaching Attitude . . . . . 57 The Elephant Procession: Visit of Ajātaśatru (Ajātasattu) to the Buddha (?) . . . . . . . 58 The Buddha preaching to the Congregation . . . . 60 Scenes from the Life of the Buddha: Sujātā offering Food to the Buddha, the Offering of Trapusha (Tapussa) and Bhallika (Bhalluka), the Visit of the King of Magadha, and the Buddha in the Streets of Rājagṛiha . . . . . . . 61 Scenes from the Life of the Buddha: the First Meditation and the Four Signs . . . . . . . 63 A Palace Scene: the Conception (?) . . . . . 65 Scenes from the Early Life of the Buddha: the Visit of the Sage Asita, the Young Prince at School, and the Practice of Archery . 68 A Rājā Distributing Alms: *Not identified* . . . . 70 A Palace Scene: *Not identified* . . . . . 73 Indra and *Apsarases* . . . . . . . 75 The *Mānushi* Buddhas and Pairs of *Yakshas* and *Yakshiṇīs* . . 77 *Apsarases* and *Gandharvas* adoring the Buddha . . . 78 The Story of the Furious Elephant Nālagiri . . . . 80

v3.c12.2

APPENDIX: A Note on the Painted Inscriptions in Caves VI, IX, X, XII, XVI, and XVII by N. P. Chakravarti . . 85

v3.c12.3

THIS Part takes back the history of Ajanta frescoes to the second century B.C., for an inscription painted in Cave X on the left wall, behind the third pillar from the entrance, has been assigned by scholars to the middle of that century, while another inscription carved on the façade of the same cave is considered to be still earlier.¹ Further, from the affinity between certain paintings of this cave and those of Cave IX,² it appears that the early frescoes of both these caves are almost contemporary, and represent a style in which the artist depicts the story in a truthful manner without much decorative embellishment. The style is, however, fairly developed, the drawing being firm and accurate and showing due regard to the three dimensions. We also notice a balanced judgement in the grouping of figures and a cultured taste in the choice of poses (Plate XXIV). The colours used are only a few, red-ochre, terre-verte, lamp-black, and white of lime, which have been used pure or mixed to produce the desired effect in the scheme (Plate XXVI).³ The figures are not altogether devoid of expression, and they show considerable religious feeling; but this feature of the Ajanta paintings is more prominent in the frescoes of the right wall of Cave X; in the representations of the Śyāma Jātaka and the Shaḍ-Danta Jātaka. They are of course some five hundred to six hundred years later than their prototypes on the left wall,⁴ but during this period it appears that the artists of Ajanta had made very great progress in imparting life and giving definite characters to the dramatis personae of their productions. The grief of the blind parents of Śyāma at his tragic death by an arrow shot by the Rājā of Benares (Plate XXIX b) and the fainting of the Queen at the sight of the tusks in the Shaḍ-Danta Jātaka (Plate XXX c) are so vivid that it is difficult to find their match in the contemporary paintings of other countries.

v3.c12.4

¹ In the Text volume of Part I (p. 2) I assigned Cave X on the authority of Fergusson (Cave Temples of India, pp. 285-7) to the second century A.D., but recently the inscriptions of Cave X have been carefully studied by Prof. Lüders, who on palaeographic grounds is definitely of opinion that the painted inscription belongs to the middle of the second century B.C., while the inscription of the façade is still earlier, va, pu, gha, and kha, showing Aśokan forms.

v3.c12.5

² Cf. Plates XV-XVIII. The pillars of the front gallery of this cave are identical in form with the pillars of the veranda of the vihāra at the right end of the caves at Bhaja which bears an inscription of the first century B.C. There is also considerable resemblance between the ornaments and the dress of the Nāga figures of Cave IX and the figures of rājās shown in the sculptures of the vihāra at Bhaja.

v3.c12.6

³ For further information regarding the pigments used at Ajanta see the Annual Report of the Archaeological Department, Hyderabad, for 1936-7, Appendix A, pp. 25-30.

v3.c12.7

⁴ There are several inscriptions which on palaeographic grounds may be assigned to the third and fourth centuries.

v3.c12.8

In the fifth century A.D. the art of Ajanta attained to a still higher level, and the frescoes of Caves XVI and XVII, both of which, according to the inscriptions carved on them, belong to the fifth century A.D.,⁵ display not only superior intellectual qualities in the conception of themes, but also more refined taste and more perfect technique in the beauty of design, balance of poise, and blending of colours. For example, the scene from the Conversion of Nanda, popularly known as the Dying Princess, on the left wall of Cave XVI (Plate LII),¹ or Indra with his troupe of musicians in the veranda of Cave XVII (Plate LXVII), or the group of heavenly votaries saluting the Lord Buddha in the same veranda (Plates LXXI–LXXII), may, in regard to vividness of expression, or grace of form, or charm of decorative detail, or soft effect of colours, be compared with advantage with the best specimens of Italian art of the fourteenth and fifteenth centuries.

v3.c12.9

⁵ Hyd. Arch. Series, No. 14, and Burgess's Report on the Buddhist Cave Temples, A.S.W.I., pp. 124 ff.

v3.c12.10

¹ This subject has hitherto been called the Dying Princess, but it really represents the deserted wife of Nanda. The other episodes of the story of the Conversion of Nanda are painted on the rock-wall adjoining the former subject.

v3.c12.11

The noble traditions of the art of Ajanta continued for several centuries,² but in the latter half of the sixth century A.D., during the reign of the Chalukyan kings, who, although tolerant to Buddhism in the beginning, professed and enthusiastically patronized the Brahmanical religion, a radical change came in the inner spirit, and the specimens of this period, with a few exceptions,³ are more conventional than spontaneous in character. A large number of frescoes in Caves IX and XVI (Plates XV *b*, LVI *a* and *c*, LVII *a*, and LVIII) representing the Buddha in the teaching attitude, seated on a throne and attended by two *chaurī*-bearers, one on each side, although sometimes conveying a certain religious dignity, by frequent repetition become monotonous, and again such themes betray a lack of freedom and narrowness of vision in artistic effort. The apparent reason for the artists' leaving the wider field of the *Jātakas*, which allowed full opportunity for the display of their artistic fancy and technical skill, was that the ruling class and their attendants, on account of professing a different religion, showed no interest in themes which reflected the glory of Buddhism, while such subjects as gods or kings seated on richly bedecked thrones and accompanied by princely attendants suited the ideas of majesty and grandeur of their own faith. Representations of this class are found in great abundance both in Buddhist and Hindu art from the latter part of the sixth century A.D. onwards, and notable imitations of them may be seen in the murals of Padmanabhapuram in Travancore State, which belong to the eighteenth century or are even later.⁴

v3.c12.12

² As is to be noticed in the paintings at Ellora, Bagh, and Sittanavasal, *vide Annual Reports* of the Archaeological Department, Hyderabad, 1927–8 and 1933–4, Pls. A to D and I–IV, and *The Paintings in the Bāgh Caves*, by Sir John Marshall, Dr. J. Ph. Vogel, and M. B. Garde.

v3.c12.13

³ The figure of a *Bhikshu* in Cave VI, upper story (Plate VII *a*), shows the perfect command of the artist in the selection of pose and the imparting of a spiritual feeling to his work. This cave was, however, hewn in the latter part of the sixth century A.D.

v3.c12.14

⁴ Regarding the artistic merits of the Padmanabhapuram paintings Dr. J. H. Cousins writes as follows: ‘Along with a remarkable exuberance of detail and decoration, they expressed a dignified reserve in the deific figures, and through a strict convention managed to convey a vivid sense of super-personality.’ *Proc. and Trans. of the All-India Orient. Conf.*, 1937, P. 45.

v3.c12.15

The frescoes which are described and studied in this Part are to be found in seven caves, Nos. VI, VII, IX, X, XI, XVI, and XVII, the architectural features of which are on a par with their pictorial embellishment. Two of these caves, Nos. IX and X, are *chaityas* (cathedrals), which on account of their close resemblance to the caves at Bhaja, Bedsa, Kondane, and Karle may be assigned to the first or second century B.C. Fergusson has considered Cave IX to be the earlier of the two, but its façade has no porch, and moreover the curvilinear brackets of the doorway (Plate XIII), and the slender pillars of its front corridor are decidedly later features, so that the cave cannot be placed earlier than the first century B.C. The plan of this temple is rectangular (Plate XII), which is again a novel feature for a *chaitya*, as it always has an apsidal end. The dimensions of Cave IX in comparison with its prototypes mentioned above, or even Cave X at Ajanta, are much smaller, being 45 ft. in length, 22 ft. 9 in. in width, and 23 ft. 2 in. in height. The pillars are octagonal in form, plain in design but slightly decreasing in girth as they rise. Their total number is twenty-one, and each of them is 10 ft. 4 in. high. The *dāgoba*, carved near the end of the hall, has a cylindrical base 5 ft. high, while its dome rises only 4 ft. above it, but it is crowned by a square capital, representing a relic-box, and an abacus consisting of six fillets, each projecting over the one below. The abacus originally supported a wooden umbrella, as at Karle. The inner architecture of the cave is characterized by solemn dignity, befitting a place of worship. The façade of the cave is, however, richly decorated with Buddhist religious motifs, such as the berm-railing, the *chaitya*-window, and the lattice-frame. The colossal figures of the Buddha, carved in the rock along the sides of the cave, are later additions (Plate XIII), and may have been executed in the fifth or sixth century A.D., when the Mahāyāna doctrines of image-worship better appealed to the religious sentiment of the votaries.

v3.c12.16

Cave X, although completely deprived of its front, is more impressive on account of its lofty ceiling and spacious nave.¹ Thirty-nine pillars divide the aisles from the nave, and support an entablature from which the vault of the barrel-shaped roof springs (Plates XXII–XXIII).² The latter originally had curvilinear rafters and cross-beams of wood in imitation of the ceiling of a structural building. Their use was, however, conventional, for they could not have given any support to the rock-roof. Pieces of old rafters still remain in the mortises, but in the *chaitya*-caves at Karle, Bedsa, and Kondane they are seen in perfect form and a better state of preservation. The *dāgoba* of Cave X is of massive proportions. The design is plain, except the capital, which as usual is decorated with Buddhist architectural motifs. Above the capital there was originally a wooden umbrella, which, like the beams of the ceiling, has perished.

v3.c12.17

¹ The inner dimensions of the cave are: depth 95 ft. 6 in., width 41 ft. 1 in., and height 36 ft.

v3.c12.18

² The pillars are octagonal in form and rise to a height of 14 ft. from the floor. The entablature has a height of 9 ft. 6 in. above the pillars, and the vault of the roof rises 12 ft. 6 in. higher still. The pillars, like their prototypes in other early caves of Western India, do not rise perpendicularly, but have an inward lean in imitation of the posts of bamboo-structures, to stop the thrust of the curvilinear roof.

v3.c12.19

As the front of the cave is completely destroyed, it is difficult to determine whether it had a rectangular porch like the caves at Karle and Bedsa. An inscription in Brāhmī characters of the second century B.C. carved on the façade, states that Vāsiṭhīputa made the endowment for the doorway of the temple.³ Traces of old brickwork exist along the side walls of the cave in the front, and it is not unlikely that the doorway mentioned in the inscription was a later construction built of bricks, which are of a large size measuring 20 in. in length, 12 in. in breadth, and 3 in. in thickness.

v3.c12.20

³ Infra, Appendix, p. 86.

v3.c12.21

The simple dignity of Caves IX and X is maintained by the architecture of Cave XII, which is a *vihāra* (monastery) and has no painting. The walls of the monastery above the celldoors are, however, adorned with carving of early Buddhist style: the berm-rail design, screens of floral and wood patterns,¹ the horse-shoe arch-heads, and a parapet of Mesopotamian origin.²

v3.c12.22

¹ The floral designs are identical with those carved above the doorway of the *chaitya* at Nasik (Plate XXV, *Cave-temples of India*), which according to the inscriptions engraved therein belongs to the first century B.C. or the first century A.D. There is a Brāhmī inscription in Cave XII as well, which from the style of the script may be assigned to the same period. *Infra*, Appendix, pp. 85 ff.

v3.c12.23

² The parapet of this design is seen freely in the Moorish buildings of Spain and North Africa, where the Arabs introduced it in the eighth century A.D. and onwards. An earlier prototype has been found recently in the excavations at Ur, and it probably came to the Deccan through Parthian or Scythian immigrants, several waves of whom moved towards India in the first millennium B.C. These foreigners after their settlement in India embraced Buddhism because it had no race prejudice. The parapet in the earliest caves is found only on their richly carved façades, as at Bhaja, Karle, and Kondane, but later it is seen in the embellishment of the interiors of the caves also, as in this *vihāra*—Cave XII, Ajanta. The absence of this parapet in the interiors of the earliest caves may indicate that the holy personages of the *Sangha* permitted the carving of only such motifs and symbols inside the temples as had a direct bearing upon the origin and growth of the tradition of the faith.

v3.c12.24

Cave XI, although situated between Caves X and XII, from the point of view of both architecture and painting is of a much later date than either of the latter two caves.³ Fergusson has included it in the group of earliest caves at Ajanta, but its façade, its pillars, its plan of the main hall and the shrine, and its main doorway and windows place it unmistakably among the latest caves at Ajanta, which belong to the latter part of the sixth century A.D. The cave is approached by a stair, and the high plinth is decorated in the form of a balustrade, which does not have curvilinear posts and rails in the style of earlier caves, but plain flat columns and horizontal bars such as we notice in the parapets of structural temples of the medieval period (Plate XXXVIII).

v3.c12.25

³ For the style of painting of this cave see *infra*, pp. 39-41.

v3.c12.26

The plan of this temple consists of a narrow veranda, a small hall with corridors on all four sides, and a shrine which is without any antechamber.⁴ The front columns of the veranda are plain octagonal, but their bracket-capitals have a design of rolled-up curtain-ends, such as we notice frequently in later caves (Plate XXXVIII). There are four cells attached to the veranda and six more in the interior of the cave, the latter being connected with the left and back walls of the hall. There is also a small recess on a high level in the shrine, which was apparently meant for storing sacred relics or other valuable articles of the temple. The shrine has a sort of *pradakshiṇa* round the figure of the Buddha, which is again a feature of the later caves. In front of the figure of the Buddha there is a statue of a votary kneeling in adoration. The head of this figure being very much damaged, it is difficult to judge of its facial expression, but the pose is very natural and shows devoutness.⁵

v3.c12.27

⁴ *Buddhist Cave-Temples*, A.S.W. I., Plate XXVIII, 2.

v3.c12.28

⁵ *Ibid.*, Plate XXXVIII, 1.

v3.c12.29

The hall is 37 ft. wide by 28 ft. deep and 10 ft. high, and is supported by four columns of a heavy style resembling the columns of Cave VII (Plate X), with which cave it appears to be coeval.⁶

v3.c12.30

⁶ *Cave Temples of India*, by Fergusson and Burgess, Plate XXX, 1.

v3.c12.31

In the fifth century A.D. Buddhist architecture, like the sister arts of sculpture and painting, seems to have reached a very high level, for the design and plan of Cave XVI indicate elegance combined with architectural vigour.⁷ The cave has fortunately an inscription, incised on the left wall near the end of the façade, bearing the name of the Vākāṭaka king Hariṣeṇa, who ruled in the last quarter of the fifth century A.D.¹ The plan of this temple can be best understood with the help of Plate XLIII. It has a spacious veranda, 65 ft. long by 10 ft. 8 in. wide, flanked towards the front by six octagonal pillars and two pilasters, the latter being most beautifully carved (Plate XLII *b*). The hall is entered by three doors, the middle one being larger in dimensions than those on either side of it. There are also two windows between the main door and the side ones. The pilasters on either side of the main doorway have lively female figures standing on *makaras*.

v3.c12.32

⁷ Between Caves XII and XVI are situated Caves

v3.c12.33

¹ For further information regarding this inscription see *Hyd. Arch. Series*, No. 14, and *infra*, p. 44.

v3.c12.34

The front gallery of the interior of the cave is a little longer than that at the back of the hall, the former measuring 74 ft. in length. Another peculiarity of this gallery is that it has a flat ceiling with beams and rafters carved thereon in imitation of their prototypes in wood. In order to make the copying complete, the ends of the heavier beams are supported by brackets designed in the form of fat dwarfs (*gaṇas*), or flying figures (*apsarases*) arranged in pairs. The sculptor has carved these figures with very great skill, showing the *gaṇas* to be over-burdened by the heavy weight of the beams, while the *apsarases* are represented as gay and joyous, their hilarious mood being not affected in the least by the burden which they bear on their necks and shoulders (Plates XLIV *a* and XLV). The sculpture of India has often been criticized as rigid and ungraceful, but who would not admire the rhythmic poise, the pulsating life, and the attractive features of these figures (Plate XLV *b*)?

v3.c12.35

The hall is 66 ft. 3 in. long, 65 ft. 3 in. deep, and 15 ft. 3 in. high. It is flanked by twenty columns, sixteen of which are of plain octagonal form, but the two middle ones in each of the front and back rows have square bases, and change first to eight and then to sixteen sides on the shafts with square heads and bracket capitals. The various lotus designs carved at the places where these shafts change their form are extremely pleasing, particularly the thin stalks of the lotus plant running along the octagonal sides with half-opened flowers at their tops (Plate XLIV *a*).

v3.c12.36

The shrine is entered from the back gallery of the hall, and it has no antechamber. Another unusual feature of the plan of the shrine is that it has two side chambers from which it is

v3.c12.37

XIII to XV. Of these No. XIII is the earliest and was used as a dormitory by the monks who were attached to the great *chaitya-cave*, No. X. It has seven cells and a small hall in front, measuring 13½ ft. in width, 16½ ft. in depth, and 7 ft. in height. The cells have, however, very neatly cut beds which are further carved at one end in the form of a pillow. The doors of the cells are very narrow, nor is there much space inside them, these features indicating that they were used for sleeping only. Cave XIV was inaccessible previously, but the cleaning of Cave XIII in recent times has exposed a stair in its right side leading to the former cave. It was apparently excavated later than Cave XIII and was never finished. The cave has a long veranda, 63 ft. by 11 ft., and an unfinished hall at the back of it, which if completed would have measured 61 ft. by 25 ft. 6 in. Neither Cave XIII nor Cave XIV has any painting. Cave XV is situated quite close to Cave XIII, but in age it is much later than the former. The plan of Cave XV comprises a veranda, 30 ft. long and 6 ft. 6 in. deep, a hall which has no columns and measures 34 ft. each way, and a shrine with an antechamber in front of it. The figure of the Buddha inside the shrine is not of the conventional type; it is rather slim and the legs are crossed and the feet turned upwards (Plate XLII *a*). This pose is more common in later images, particularly among the representations of Jain gods of seven centuries later. On the frieze above the main doorway are carved four pigeons in very natural poses, and between them is a *stūpa* with a halo of cobra-heads, while the base of the *stūpa* is made of the coils of the same reptile (Plate XLI *b*). The roof of the antechamber of Cave XV has traces of painting.

v3.c12.38

separated by a screen of two pillars and pilasters. These chambers are also entered from the back gallery of the hall. The image of the Buddha in the shrine is of colossal size, but it is rather heavy in artistic effect (Plate XLIVb).

v3.c12.39

The architecture of Cave XVII, although in age a quarter of a century later than Cave XVI,¹ maintains the magnificence of the style of the latter. However, as the majority of the paintings of Cave XVII form the subject of the next Part (IV), it will be appropriate to discuss the merits of its architectural detail also in that Part.²

v3.c12.40

¹ There is an inscription in Cave XVII also which has been published. According to the genealogy of Vākāṭaka kings given therein the inscription belongs to the early part of the sixth century. *Buddhist Cave Temples*, by J. Burgess, A.S.W. I., pp. 128–32, and *Cave-Temple Inscriptions*, p. 73 seq.

v3.c12.41

² In this Part (III) only the paintings of the veranda of Cave XVII have been studied.

v3.c12.42

The remaining caves (VI and VII) of which the frescoes have been studied and illustrated in this Part are nearly a century later than Caves XVI and XVII, but Fergusson, and in his wake some modern writers also, have placed the architecture of the former two caves earlier than that of Caves XVI and XVII.³ This opinion does not seem to be correct, the reasons being that the plan of Cave VI both in its upper and lower stories, although conforming to the general plan of the fifth-century *vihāras* of Ajanta,⁴ has certain features which seem to be later developments, indicating more or less the vagaries of the architect's mind. For example, he has increased the number of pillars in the lower story with a view to giving adequate support to the rock-floor of the upper story. The architect has shown considerable ingenuity in the arrangement of these pillars by transforming the middle hall of the *vihāra* into three aisles, the central one of which leads to the shrine. But the presence of these pillars in the middle of the hall is somewhat obstructive, for the main idea in designing the hall was that the congregation should sit comfortably on its floor and be able to look freely at each other and also at the chief *Bhikshu* who addressed them on ceremonial occasions. The design of the arch carved over the door of the shrine again is capricious. It springs from the mouths of *makaras*, and is curvilinear (Plate Va).⁵ The larger numbers of the figures of the Buddha in the upper story of this cave is another feature indicating a later origin (end of the sixth century A.D.). In the corridors of the upper story, besides the cells there are chapels with pillars in front. Over the chapel at the left end of the front corridor there is a frieze representing figures of elephants in different postures, which show the close observation of the sculptor in studying the habits of the animal. The figures are very lively and indicate considerable movement. One of them is killing a tiger. There is another effigy of an elephant, carved on a bracket in the veranda (Plate Vb). In this sculpture the animal holds garlands round his trunk which he evidently is bringing as an offering to the Buddha. The striking feature of the carving is that the animal has been shown almost trotting, animated by feelings of joy at the sacred mission which he is performing.

v3.c12.43

³ *Cave Temples of India*, by Fergusson and Burgess, pp. 299 ff., and *Indian Architecture*, by P. Brown, pp. 66–7.

v3.c12.44

⁴ Caves I, II, XVI, and XVII, which according to contemporary inscriptions belong to the fifth century A.D.

v3.c12.45

⁵ Over the shrine door of Cave XX is also a similar arch, but the latter consists of two curvilinear parts each springing from the mouth of a *makara* and resting in the middle of a lotus-flower. *Buddhist Cave Temples*, Plate XX, Fig. 1.

v3.c12.46

The plan of Cave VII is rather unusual, having a long hall (or veranda) with two porches in front and an antechamber and a shrine at the back (Plate X). There are also cells, four in the back wall, and three in each of the side rooms built at the right and left ends of the hall.¹ The side rooms have been excavated on a higher level than the hall, and as the approach to them is unfinished they appear to be later additions. As this cave is near the old steps which formed the main approach from the valley to the group of the earliest caves, IX, X, XII, XIII, and XIV, it is likely that its plan was changed at various times according to the requirements of the monastic orders which were established there. The sculpture of the antechamber and the shrine of this *vihāra*, however, show it to be a late sixth-century excavation.

v3.c12.47

¹ *Ibid.*, Plates XXVII, XXVIII, Fig. 1, and XXXVII, Fig. 2.

v3.c12.48

In summing up these remarks it may be observed that the main two types of the rock-hewn temples of Ajanta, that is, the *chaitya* and the *vihāra*—the former beginning with a rectangular portico and a nave with side aisles arranged in the form of an apse at the end, in the second century B.C., and the *vihāra* starting with only a square hall having cells excavated on three sides of it—kept to these plans under the Hīnayāna school, which seems to have continued in the Deccan up to the third century A.D. But in the fourth century, under the influence of the Mahāyāna school, the representations of the Lord himself in various *mudrās* (attitudes), either carved in rock or painted, and a well-developed pantheon comprising a number of other gods and goddesses, necessitated the excavation of niches in the aisles and other prominent parts of a *chaitya*, such as the *dāgoba*, for the insertion of these deities. Similarly, the same causes brought about a change in the original plan of the *vihāra*, which, instead of comprising a bare hall, began to have a corridor on all four sides of the hall and a shrine with an antechamber at the end. The shrine was added to house the figure of the Lord, and the antechamber for the accommodation of the principal deities of the faith. At Ajanta the earliest *vihāra* of the latter class is Cave IV,² which has a colossal image of the Buddha in the shrine and some dainty carvings on the doors and the windows of the hall opening on the veranda. The large dimensions of this cave,³ combined with the massiveness of its architectural features and the frugal use of decorative work, present a faithful picture of Buddhist religious dignity, reflecting the restraint of monastic life on the one hand and the expansiveness of spiritual life on the other. The *vihāra*⁴ was probably excavated in the third century A.D. or still earlier;⁵ but the decorative work may have been done at a later date.

v3.c12.49

² It has traces of paintings in the ceiling of its veranda; but they are too faint and fragmentary to be reproduced in this publication.

v3.c12.50

³ The hall of this cave measures 87 ft. each way and is supported by twenty-eight columns, 3 ft. 3 in. in diameter.

v3.c12.51

⁴ The *vihāra* may have been built at the same time as the *chaitya*-cave X, for a temple of colossal size required a large monastery for the dwelling of monks.

v3.c12.52

⁵ Caves III and V are both unfinished and have no paintings. The former was inaccessible until recently; but a stair has recently been constructed by the Archaeological Department of Hyderabad to give access to this cave. Cave V is interesting as showing the method which was followed in excavating the temples.

v3.c12.53

Although in this brief notice of the outstanding features of the paintings and architecture of Caves IV–XVI their serial order could not be observed, yet in describing the stories painted on the rock-walls of these temples and in studying the artistic merits of the frescoes the serial order of the caves will be strictly followed, so that the plan adopted in Parts I and II, dealing with the frescoes of Caves I and II, is maintained in this Part as well.

The Buddha in the Teaching Attitude: First Sermon (?)

v3.c12.s1.2

Plate IIa

v3.c12.s1.3

*The subject is painted on the left wall of the antechamber in the lower story of Cave VI.*

v3.c12.s1.4

THE fresco is much damaged, and hence it is difficult to decide whether the subject represents the First Sermon or the Great Miracle of Śrāvastī.¹ The principal figure of the Buddha is in the teaching attitude, but the smaller representations are painted in other *mudrās* as well. For instance, the Buddha to the left proper of the larger figure is in the *bhū-sparśa* or earth-touching attitude.² They are all seated on large full-blown flowers of the lotus,³ and are separated from each other by stalks of the same plant, which bear flowers in various stages of bloom. There are some figures below those of the Buddha, one of which, apparently representing a woman, is making an appeal to the Great Being, for she has placed her right hand upright on her head. The artist has drawn the fingers of her hand with much skill and feeling, and to show his love of ornament he has painted a ring with a large jewel round the second finger of the lady.

v3.c12.s1.5

¹ This subject is frequently painted at Ajanta, *vide* Part I, pp. 34-5, and Part II, p. 27.

v3.c12.s1.6

² For this as well as other *mudrās* see Part I, p. 34 n.

v3.c12.s1.7

³ The lotus was a symbol of self-creation among the Buddhists. See A. Getty’s *The Gods of Northern Buddhism*, p. 192.

v3.c12.s1.8

An interesting feature of this fresco is that in some places the original outline of the subject in red is preserved, wherein the corrections of the master-artist in black are also seen.

Dvārapālas and a Female Votary

v3.c12.s2.1

Plates IIb, IIIa-b, and IVa

v3.c12.s2.2

*These figures are painted on the back wall of the antechamber, on either side of the shrine-door, in the lower story of Cave VI.*

v3.c12.s2.3

THE door of the sanctuary in the lower story of Cave VI originally had richly dressed *dvārapālas*, whose heads and other parts of the body are almost obliterated now, but there still exist fragments showing that they wore long coats and lavishly bedecked girdles and waistbands. The figure of the *dvārapāla* to the right of the shrine-door is reproduced in Plates III *b* and IV *a* by way of illustration.⁴ The latter is based on a photograph taken direct from the rock-wall, while the former is a faithful tracing of the original fresco. Of this *dvārapāla* a part of the left arm had survived, which has now been carefully preserved by the mechanics of the Department. The arm shows skilful drawing and perfect command in expressing inner feeling, which in this case may be interpreted from the gesture of the sensitive fingers of the *dvārapāla*. The figure represented some rājā, for the Bodhisattvas and *Yakshas* are not generally delineated as wearing coats. The *dvārapāla* has ornaments on his arm, wrist, and little finger, but what strikes one most is the design of his girdle, which contains scrolls, flowers, and check-patterns in alternating divisions. The design was apparently worked out in embroidery, the material being silk or gold and silver threads, or all three.

v3.c12.s2.4

⁴ The two *dvārapālas* are also reproduced in *The Paintings of Ajanta*, by J. Griffiths, vol. ii, figs. 134-5.

v3.c12.s2.5

The *dvārapāla* is further adorned with a belt of pearls or gold beads, and also with chains of the same metal which hang down to the border of his dress. The jewel-rooms of the Deccan kings seem to have abounded in pearls and gold at all times, for beside the lavish display of such ornaments in the paintings of Ajanta, when we come to the medieval period we find ‘Alāu’d-Dīn Khaljī carrying off a booty of 50 maunds of gold and several maunds of pearls in addition to valuable stuffs and a large number of elephants and horses, after his conquest of the Deccan.¹

v3.c12.s2.6

¹ Briggs’s translation of the *Ta’rīkh-i-Firishta*, vol. i, pp. 307-10.

v3.c12.s2.7

The figure of the *dvārapāla* to the left of the shrine-door is still more damaged;² but in the lower part of the fresco the head of a lady can be made out with certainty, and the fresco when complete must have been one of the masterpieces of the later style of Ajanta³ (Plates II *b* and III *a*). The face shows much serenity, and the features which are intact are beautifully drawn. The most attractive parts of the fresco are the ornaments and the coiffure. The former comprise a necklet of large pearls and several strings of smaller pearls round the neck, two types of ear-rings, and a pearl ornament round the forehead, the strings of which along with a garland of flowers are intertwined with the hair. A large flower has been placed above the temple on one side, while the hair above the forehead is arranged in small clusters like the drops of an ornament. The effect is extremely charming. The large ear-rings owing to their weight have been attached to the middle of the ear, but still the ears are elongated, being rather conventional in shape, as noticed in contemporary Buddhist sculpture. The small ear-rings with pearl drops shown in the upper parts of the ear are still in vogue in the Deccan.

v3.c12.s2.8

² A colour-reproduction of a fragment of this fresco has been given by Griffiths in Part II, Plate 134.

v3.c12.s2.9

³ Griffiths has reproduced this figure in Part I, p. 10 (Fig. 12) and apparently reconstructed it with a view to showing the full beauty of the drawing.

Temptation of the Buddha

v3.c12.s3.1

Plates III *c* and IV *b

v3.c12.s3.2

*The scene is painted on the right wall of the antechamber in the lower story of Cave VI.*

v3.c12.s3.3

THE fresco is much damaged and it will be difficult for the reader to trace the drawing of the various figures either with the help of the photograph (Plate IV *b*) or the tracing (Plate III *c*). On the rock-wall a keen eye may, however, notice the figure of Māra to the right proper of the Buddha. He is richly adorned with jewels, as befitting the dignity of a rājā, and is wearing a *dhotī* of striped silk to cover the lower part of his body. In his left hand he holds either a curved sword or a sharp sickle to cut off the head of the Buddha, who is sitting in the middle, unperturbed, in the *bhūsparśa* (earth-touching) attitude. The head of the figure representing Māra is too much damaged to show any expression, but the painter must have tried to show him in an angry mood. Between the figures of Māra and the Buddha there is the representation of a stalwart warrior whose head is missing. Below him, by way of contrast, the artist has painted the figure of a goblin with tiny hands and the head of an animal. He has strange headgear and a waist-band from which little bells are hanging. The goblin is trying to shake the throne of the Buddha. Near the nimbus, round the head of the Buddha, the faces of two demons may be seen, with bulging eyeballs and dumpy noses. Below them are the trunks of two elephants, apparently throwing a strong spray of water to disturb the Buddha in his meditation. Behind them the hands, trunks, and legs of several other figures may be noticed. One of them has thrown up a horse, which is seen in the air over the head of the Buddha. The figure of the horse is intact, and it is shown falling down on its back with its legs and muzzle turned upwards. To the left proper of the Buddha are two more monsters, riding on donkeys. The one nearer to the Buddha has two skulls in his headgear. The other is apparently wearing a mask to frighten the Buddha. The lower end of the mask is seen across the face of the demon. Below these the traces of some more figures may be noticed, and also the head of a cobra which was possibly used as a lash by some demon.

v3.c12.s3.4

Below the throne of the Buddha were painted the daughters of Māra, disporting themselves in an enticing manner. The head of only one figure can be seen; the rest have completely perished. Such subjects gave the artist an opportunity to display his sense of the grotesque, which he has done admirably, inspired by the religious stories which were extant at that time regarding this great event in the life of the Buddha, that is, his enlightenment after protracted meditation.¹

v3.c12.s3.5

¹ There are traces of painting inside the shrine and on the wall of the back corridor. On account of the fires of the *yogīs* and the neglect of centuries combined with climatic deterioration, these paintings have become extremely dim. Still, on the wall of the back corridor, to the right of the antechamber, the figure of the Buddha seated on a throne may be seen.

A Bhikshu

v3.c12.s4.1

Plates VII*a* and VIII*a

v3.c12.s4.2

*This figure is painted near the feet of the Buddha carved on the left wall of the antechamber in the upper story of Cave VI.*

v3.c12.s4.3

THE upper story of Cave VI, like the ground floor, was originally adorned with frescoes, but the bats and other unsympathetic denizens of the cave have subsequently wrought havoc on a large scale, and except a few fragments clinging here and there, the paintings have completely perished. In the upper story of the cave, however, some specimens have survived which retain the original colours and are reproduced in Plate VII *a-b*. The most interesting of these is the figure of a *bhikshu* in the act of adoration. He is kneeling before the figure of the Buddha and looking up towards his face with the light of hope in his eyes. The drawing of the figure is firm, marked by a black outline. The fingers and toes have been painted as usual with very great care, but in this figure the remarkable feature is the large abdomen which generally develops in old age when no physical exercise is taken. The *bhikshu* carries a metal censer in his right hand, while in the left he holds three lotus flowers¹ which he has brought to offer to the Buddha. The colours used in painting this figure are only two, red-ochre and yellow-ochre, relieved by an outline in black. The result is, however, very happy, the former colour very well representing the reddish-brown of the skin of the *bhikshu* and the latter the tint of his scanty garment, comprising a scarf on the left shoulder and chest and a loin-cloth which fails even to reach his knees. The figure is a perfect picture of devoutness and religious hope. The artist could not restrain his sense of humour in painting even a holy person like this, and to evoke laughter he has delineated the tuft of hair on the crown of the *bhikshu*'s head (the *ushnisha*) in the form of a flower.

v3.c12.s4.4

¹ Some scholars have suggested that the three flowers represent the *triratna*, the Buddha, Order, and Law.

Dvārapālas and Pairs of Male and Female Figures

v3.c12.s5.1

Plates VII *b* and IX *a

v3.c12.s5.2

These figures are painted on the jambs and the side-walls of the doorway of the chapel at the right end of the front corridor in the upper story of Cave VI.

v3.c12.s5.3

THE figures of happy couples representing donors or *arhats* with their consorts, or dancers attached to the temple, are carved on the façades of the earliest Buddhist *chaityas*, such as those of Karle, Bedsa, Kondane, and Bhaja. But with the passage of time the practice became more a decorative feature than a means of conveying religious significance or representing any particular deities of the Buddhist pantheon. The doorways of the fifth and sixth century temples of the Buddhist faith are therefore adorned with male and female figures in a variety of most delightful poses.

v3.c12.s5.4

In this subject (Plate VII *b*) we see a lady in a thoughtful mood but with all the charm of her youth, which has been enhanced by scantiness of dress² and profusion of ornaments, and above all by a graceful attitude. She has placed one of the fingers of her left hand on her lip in the typical Indian style. The gesture denotes bashfulness as well as mental absorption. The artist was a master of the technique and he thoroughly understood where to lay the shade to accentuate the plastic beauty of the limbs. Behind the lady to her left is another figure who is fully dressed, the lower garment extending much below the knees and the upper part of the body covered either by a large scarf or a coat. Behind the lady is a dwarf of a fair complexion holding something in his hands which is not clear in the fresco.

v3.c12.s5.5

² She is wearing a tight bodice of a thin material and a pair of tight drawers which do not reach her knees.

v3.c12.s5.6

To the left of this group in another panel is a male figure representing a Bodhisattva or some *arhat* making a religious gesture with the fingers of his right hand. The head of the figure is damaged, but the limbs show careful modelling and the pose is also natural. There is a dwarf in this panel also, but the colours and the drawing are so faint that it is difficult to make out his features with any precision, nor is it possible to appreciate the expression of his face. Above this panel there is another representing two ladies, both of whom are standing, and one of them has crossed her legs in a graceful manner. The upper parts of these two figures are much damaged. Above them there is another couple, in a comparatively better state of preservation. The female figure is looking towards the door of the chapel and has made a loop with the fingers of the left hand, apparently representing some religious mystery.

v3.c12.s5.7

To the right of these panels, on the jamb of the door, there are three panels representing Parthian male and female figures in a joyous mood, sipping wine or love-making. The male figures have beards, and their caps also prove them to be non-Indian. Corresponding to the last three panels there are three on the right jamb of the door again representing Parthian figures whose blue jackets and peaked caps form a prominent feature. Below these is a female figure accompanied by a small male figure with long hair. The latter is resting on a crooked stick which passes through his legs. To the left of the figure of the woman is a dwarf, the drawing of which has become very dim (Plate IX *a*). Close to the right jamb of the doorway, on another facet of the door-frame, there are three more panels, the lowest of which bears the figure of an *arhat* or a Bodhisattva to correspond to the figure on the left side of the doorway. The remaining two panels contain the figures of females carved in beautiful poses (Plate IX *a*). The presence of figures with sunken cheeks and short beards, wearing peaked caps and full-sleeved coats, as shown in some of the panels referred to above and also in several of the later frescoes of Caves I and II, point to a period in Indian history when Partho-Scythian dynasties were established in Surāṣṭra and Malwa, whose territories occasionally included the northern parts of the Deccan.¹

v3.c12.s5.8

¹ *Ajanta*, Part I, p. 46, and *Indian Art of the Buddhist Period*, by G. Yazdani, p. 7 (Oxford Univ. Press).

Miracle of Śrāvastī

v3.c12.s6.1

Plates VIII*b* and IX*b

v3.c12.s6.2

*This subject is painted on the right wall of the chapel in the front corridor of the upper story of Cave VI.*

v3.c12.s6.3

THE miracle of Śrāvastī was a favourite theme of the artists of Ajanta, for it gave the religious-minded novices, as well as the trained craftsmen, an opportunity for painting the Great Being whose figure on account of the various conventional features had become almost stereotyped. The figures of the Buddha painted on the walls of the antechamber of Cave II represent a very poor art, and the work must have been done by beginners with the help of paper-stencils.² But in this chapel the artist has delineated the subject with extraordinary skill, making the features of the Buddha highly refined and giving a spiritual effect to his appearance by the judicious use of high lights (Plates VIII *b* and IX *b*). The head of the figure in the middle has almost a classical grace (Plate VIII *b*) and may be compared with some of the best works of Italian artists of the fourteenth and fifteenth centuries. It bears a striking resemblance to the figure of the standing Buddha on pillar 5 of the right aisle of Cave X,¹ and it is not unlikely that both the figures were painted by the same artist, although some of the frescoes of the latter cave date back to the second century B.C. These two Buddhas, like the other frescoes of Cave VI, are, however, not earlier than the sixth century A.D. This view is confirmed by the characters of a three-line inscription painted on the left wall of the upper story of this cave between the door of the cell and the chapel.²

v3.c12.s6.4

² *Ibid.*, Part II, p. 28 (f.n. 1) and Appendix, p. 65.

v3.c12.s6.5

¹ Cf. Plate XXXVI *b* and Griffiths, 42-3.

v3.c12.s6.6

² Infra, Appendix, p. 88.

v3.c12.s6.7

In this fresco the aesthetic taste of the painter has induced him to deviate from the conventional features of the Buddha, that is, thick lips and nose and an almost roundish face, and replace them by a thin nose with short mouth and full lips and an oval face. The effect of the change is very happy, and the figure has become more vivid than in the case of the ordinary paintings of the Buddha at Ajanta. The artist has also made a slight change in the garment of the Buddha, and instead of covering the upper part of his body with a sheet or a large scarf, has dressed him in a full-sleeved robe which is trimmed at the neck with a black collar. This simple alteration has enhanced the artistic effect considerably.

v3.c12.s6.8

The ceiling of this room was also painted, and traces can still be made out. Besides the usual concentric bands of decorative patterns there are lotus-leaf designs arranged in squares.³

v3.c12.s6.9

³ The room at the left end of the veranda in the upper story of this cave was also painted, but through percolation of water which came from a crack in the rock-ceiling much damage has been done to the frescoes. This crack has been repaired by the Department, and there is no danger of any further damage to the walls and the ceiling of the room. On the right wall there are figures of two *apsarases* and also the representation of a serpent. Two Buddhas are also to be seen, one of whom is in the teaching attitude. On the back wall of the room are some architectural designs, showing wooden pillars and roofs, the latter resembling the canopies of chariots (*ratha*).

First 113 of 894 passages. Read on

Ajanta: The Colour and Monochrome Reproductions of the Ajanta Frescoes Based on Photography v3: Volume 3

Contents

Volume 3

Front matter

Part Three