Chapter V

— Architecture

The Imperial Gazetteer of India, Vol. II, The India Office, 1908 · from page 155 · about 101 minutes

Read by machine, not yet checked by a person. It may hold small errors.

To present, even in the merest outline, any satisfactory account of Indian architecture in the space allotted in the following pages may seem almost impossible. In no other country, perhaps, have so many various styles been employed, nor have developments and changes of the styles been so marked. To separate these various forms into well-defined groups with distinctly recognized characteristics, and to trace their modifications in the course of history, is the task presented to the student. To give any comprehensive outline of the development of these varieties and of their complex relations to one another would necessitate entering into details and the employment of illustrations that would be incompatible with the extent and aims of this chapter : the most that can be attempted is a sketch of the main features of architectural advancement with reference to outstanding examples, to which may be added some notices of less-known groups. For a fuller account the reader may be referred to Fergusson's History of Indian and Eastern Architecture.

The careful study of this art as developed in India is of extreme interest for the general history of architecture ; and, whatever may be his estimate of its aesthetic qualities, the student cannot fail to realize that the designers of Indian structures attained as successfully as their Western contemporaries the aims they had before them, though they used arrangements and adopted forms and details very different from those of Occidental builders in ancient or mediaeval times. These forms and their adaptations of course require study for their proper appreciation ; but once this is understood they become really interesting—for the perception of the suitability of the design to its purpose creates an interest, if not an admiration, for the whole. But besides the scientific advantages of the study, which need not be here enumerated, it has been remarked by the late Mr. Fergusson—to whose genius the science of the history of architecture owes so much—that ‘it will undoubtedly be conceded by those who are familiar with the subject that, for certain qualities, the Indian buildings are unrivalled. They display an exuberance of fancy, a lavishness of labour, and an elaboration of detail to be found nowhere else.’

Without any properly historical chronicle, our knowledge of Indian history and antiquities is hampered by difficulties not perhaps found in the case of any other country. We possess scarcely a landmark in history previous to the invasion of India by Alexander the Great in the fourth century B.C., nor do we know of an architectural monument of earlier date. For later periods there are fortunately a few examples dated by inscriptions, and for others—by applying the scientific principles developed by Thomas Rickman for the discrimination of other styles and relative ages of architectural works—we are now enabled to arrange the monuments of India with considerable certainty in chronological sequence or order of succession.

Architecture, it must be understood, is something more than the mere art of building in any form ; and, if a definition is required, it must be that it is the fine art of designing and constructing ornamental buildings in wood, stone, or other material. It is thus distinct from common building or civil engineering.

Early Architecture—Wooden

It is generally conceded that in the early architecture of India, as in that of Burma, China, and Japan, wood was solely or chiefly employed ; and, if brick or stone were in use, it was only as a building material for foundations and for engineering purposes. Even as late as the end of the fourth century B.C. we find Megasthenes stating that Pātaliputra, the capital of Chandragupta, was ‘surrounded by a wooden wall pierced with loopholes for the discharge of arrows’ ; and if the capital were defended by such palisading, we may fairly infer that the architecture of the time was wholly wooden. And, for all religious or private structures in a tropical climate, wood has marked advantages over stone. On the Sānchī gateways, brick walls are represented, apparently, however, as fences or limits with serrated copings, but not in architectural structures. And at whatever date stone came to be introduced, the Hindus continued and repeated the forms they had employed in the earlier material, and preserved their own style, so that it bore witness to the antecedent general use of wood. Hence we are able to trace its conversion into lithic forms until finally its origin disappears in its absorption in later styles.

The perishable nature of this material readily accounts for the disappearance of all Indian buildings of early date. Memorial stūpas, it is true, have been assigned by some archaeologists to a date previous to the fourth century B.C. ; but they have been excavated with so little conception of scientific method that the main result has been the destruction of such evidences of their real age as might have existed. We have thus no monument of an architectural character that we can cite as certainly belonging to a date before the third century B.C. : one to be noted presently is hardly architectural and shows but little experience in the use of stone. The transition from wood to stone was naturally, as in other countries, made gradually, and at first by the use of brick, to fill in the wooden framing of the structures.

The spread of Buddhism to the westward and, at latest, the invasion of Alexander brought India into contact with Persia, where, in the sixth and fifth centuries B.C., Cyrus and the succeeding Achaemenian kings had hewn out mausoleums in the rocks and constructed palaces with stone basements, pillars, and doorways, filling in the walls with brick, as in the earlier Assyrian buildings. These works would naturally attract the attention of Indian visitors—whether missionaries, ambassadors, or merchants ; and the report of such magnificent structures would tempt Indian princes to copy them. The embassies of Megasthenes to Chandragupta, and of Deimachus to his son, were probably not the only visits of the kind during the interval between the time of Alexander and the accession of Asoka ; and such visitors from the West were specially suited to convey a knowledge of Persian arts to the contemporary Indian potentates. The daughter of Seleucus Nikator, too, who was given in marriage to Chandragupta, may have helped in this.

By the middle of the third century B.C. we find the great Asoka in communication with the contemporary kings of Syria, Egypt, Macedonia, Epirus, and Cyrene ; and to his reign belong the great stone pillars, with capitals of Persian type, that are engraved with his religious edicts. A convert to Buddhism, Asoka is credited with the construction, all over the country, of vast numbers of stūpas—monumental structures enshrining relics of Sākyamuni Buddha or other Buddhist saints ; and with them were erected monasteries and chapels for the monks. We cannot positively identify any of the few still existing stūpas as having been actually built by him ; but there can scarcely be a doubt that the sculptured rails at Buddh Gayā and Bharhut, the caves at Barābar, and the oldest of the cave monasteries in Western India, were excavated during the existence of the Maurya dynasty, or at least within the two centuries following Asoka’s accession.

It was thus partly, at least, to Buddhism, under the impulse of this powerful sovereign, that we owe the inception of all the monuments that have come down to us from that age. Buddhism had not then developed the cult of a personal Buddha farther than to reverence his relics, the representation of his footmarks, the sacred bodhi tree and other symbols, combined perhaps with aboriginal snake-worship. But we must keep in mind that the Jains and other sects, contemporary with the Buddhists, were also protected by this beneficent monarch, and that they raised shrines and constructed cave temples and monastic abodes for their devotees, and further, that these are now recognized by distinctive symbols, by inscriptions, or other evidences of the sects for whom they were prepared.

Stone Architecture—Stūpas

One structural building, close to Rājagriha or Buddh Gayā, is claimed as probably of earlier date than the age of Asoka. This is the great basement known as Jarāsandha-kī baithak. It is about 85 ft. square at the base, and slopes upwards from 20 to 28 ft. to a platform 74 ft. by 78 ft., built entirely of large unhewn stones, neatly fitted together without mortar, and contains fifteen small cells, mostly on the north side, each 6 or 7 ft. in length, and half that width. This is apparently the ‘stone house’ mentioned by Hiuen Tsiang, and the rude cavern behind it would correspond to the traditional Asura’s dwelling. So far, then, as we at present know, this structure may represent the earliest vihāra or monastic dwelling found in India, and its resemblance to the Birs Nimrud has been pointed out by Mr. Fergusson.

On the inscribed pillars or lāts set up by Asoka, besides the Persian form of capital, we find the honeysuckle with the bead and reel and the cable ornaments employed in earlier Assyrian and Persian sculpture; and, though not noticed afterwards in India proper, these continued in use in Gandhāra on the north-west frontier for about four centuries, which seems to indicate that it was from Persia that these forms first came, along with the suggestion that led to the conversion in India of wooden architecture into stone. Many of these lāts, as they are called, have been destroyed ; but it seems probable that they stood originally beside stūpas or other sacred structures. Beside the great stūpa at Sānchī-Kānākhedā, near Bhilsa, there was found a portion of one of Asoka’s pillars, with a fragment of one of his edicts upon it ; but in all other cases the buildings have now disappeared.

The stūpas were more or less conventional or architectural representations of funeral tumuli, and were constructed for the relics of the Buddha and of his disciples. How this relic-worship originated and came to hold so large a place in the Buddhist cult we can hardly conjecture ; the sentiment could not have arisen for the first time on the death of Gautama, when, we are told, eight stūpas were built over the corporeal relics, a ninth over the vessel (drona) by which they were divided, and a tenth over the charcoal of the funeral pile—the erection of such monuments must have been an established custom long before. Asoka, we are told, pulled down the first stūpas over Buddha’s remains, and erected others, which were doubtless different and more architectural. But whether or not we shall yet discover, from actual examination, their real construction, we can hardly doubt that they formed the general model for such objects for the following centuries, and their outward appearance is often represented on later monuments.

The Sānchī-Kānākhedā stūpas, of which two or three were quite entire at the beginning of the nineteenth century, and the second largest of which almost certainly dates from about 200 B.C., may be accepted as examples of the Asoka pattern. On a low circular drum, a hemispherical dome was constructed, with a procession path round the latter, and over the dome a box-like structure surmounted by an umbrella and surrounded by a stone railing. Round the drum was an open passage for circumambulation, and the whole was enclosed by a massive rail with gates on four sides. The dome, surrounded by its drum or berm, was in no sense architectural, and, having but one special purpose, could convey no information as to the art of the age. The railings, however, are most interesting : they are constructed as closely as possible after wooden patterns, and are the only examples of this type that survive here or elsewhere. On the second stūpa, which has been badly ruined by bungling excavators, the sculptures are chiefly on discs upon the uprights of the rails ; and from the difference of character, as compared with those of the toranas of the first and third stūpas, they seem to indicate an earlier stage. But these have not secured the attention they deserve, nor have they been adequately depicted. The uprights are square, and there are as usual three rails between each, and a heavy coping or headrail is placed over the uprights and holds them together by tenons on their upper ends.

To about the same age must belong the remains of the rail round the Buddh Gayā temple, and probably only a little later comes the Bharhut stūpa enclosure. The remains of a rail found at Muttra is of early Jain work, and may perhaps be placed slightly later than the last. The uprights are carved with full-length basso-rilievo figures or with discs ; but the remains found there belong to a lengthened period, and the want of information as to the relative positions in which the various subjects were found deprives us of the basis of any safe induction as to the development of the art.

In the great stūpa at Sānchī we have something more complete : the uprights are still square, and the three cross-bars between each pair are lenticular in section ; but, like the stūpa itself, they are of unusually large dimensions, the rail being 11 ft. in height. This, as well as the smaller rails that were formerly on the berm and round the platform over the dome, were all carved after wooden patterns. But here the erections of lofty toranas at the entrances formed a better field for sculpture than the rail, and it was to them that it was exclusively applied.

These ornamental gateways must have belonged to most if not to all of the larger and more notable stūpas ; though at first they were evidently of wood, and the earliest Buddhist missionaries seem to have carried the idea of such adjuncts with them, for even in Japan they are well known at all temples as tori-i, and in China as p'ai-lus or p'ai-fangs, while wooden toranas are to be met with in villages in Rājputāna and elsewhere down to the present day—an example of what archaeology teaches, that the present is linked to the past in one chain.

The whole of the superstructure of these Sānchī examples is so essentially wooden in character that, as Mr. Fergusson remarked, ‘we rather feel inclined to wonder how men dared to attempt its erection in stone, and are equally astonished that it should have stood’ for twenty centuries ‘nearly uninjured.’

The remains at Sānchī, however, evidently belong to a very extended period, and there are scarcely any reliable data by which to fix the dates of the earliest structures, while much has disappeared, even during the nineteenth century, that might have aided our knowledge. The former presence of one, if not two, Asoka pillars at the great stūpa would point to the stūpa itself being in existence in his time at the latest ; the rail round it may have been added subsequently, and the gateways still later ; but the inscriptions on Stūpa No. II are in the same characters as those of the Asoka inscriptions, and the sculptures appear to be more archaic than those on the great stūpa, so that this rail may be the older of the two. Yet the difference may be small, for several of the inscriptions on the large stūpa seem to be also of the same age.

The gateways would naturally be erected last ; and on the south one we find an inscription on one of the beams stating that it was the gift of an officer under Śrī Śātakarṇi, an Āndhra king who ruled about 160 b.c. The others were probably erected not long before or after this date. Of the buildings that once covered the surrounding area, the ruins of which still remain, our information is defective ; but a small temple to the south-east of the great stūpa is probably the oldest remaining, and may go back even to the third century b.c.

The remains of the Amarāvati stūpa on the lower Kistna river present a still more complicated problem, for of the original work only a few archaic sculptures have survived. Its rail, at least, must have been entirely reconstructed before our era or shortly after,—the sculptures representing the veneration of relics, &c., but no representation of the Buddha ; and then about the middle or end of the second century a.d. a great ‘restoration’ had been effected, when what has been called the inner rail—probably a wainscoting of the stūpa itself—was added, consisting of marble panels sculptured with those figures of Buddha, &c., that were so much favoured by the Mahāyāna school of later Buddhism.

Cave Temples

The earlier rock temples must be of about the same age as these stūpas. Indeed in the Barābar hills, about sixteen miles north from Gayā, we find a group of caves in three of which are short inscriptions of Asoka, dated in his twelfth and nineteenth year, and dedicating them to the Ājīvikas, who seem to have been a naked sect, founded by Makkhali Gosāla, and similar to the Jains. Close by are three more caves, dedicated to the same sect by Asoka’s grandson Dasaratha about 215 b.c. The architectural features of these caves are few : they have vaulted roofs, the walls of some of them are carefully polished, and in the ends of three of the earliest are circular chambers or shrines, the fronts of two of which are carved with overhanging eaves. In the case of the Lomas Rishi cave, the outer apartment of which is 32½ ft. in length, the doorway is surrounded by carving which represents in stone the form of the structural chaityas of the age. They were apparently constructed with strong wooden posts sloping slightly inwards, supporting longitudinal rafters mortised into their heads, while small blocks at the sides were employed to keep the roof in form. Between the main posts was a framework that served to support the smaller rafters, on which lay the roof formed of three thicknesses of plank. The form of this roof was therefore a slightly pointed arch, having a ridge along the centre. The door, like the others in this group, has sloping jambs.

Now when we compare this with the façades of other early caves, we note the identity of construction. Among these the chaitya caves at Kondānā, Bhājā, Pitalkhorā, and No. X at Ajantā are the oldest. The excavators had not yet learnt to carve out these halls leaving a screen wall in front, or they still preferred to retain the wooden fronts. In two of them at least —that at Bhājā and the one at Ajantā—the mortises in the floor indicate clearly where the supports of the wooden screen once stood ; and in the case of the Kondānā chaitya the remains of the wooden framework occupying the upper portion of the façade were still in existence not very many years ago, and were supported by posts rising from the floor, the heads of which were still left. In Bhājā, Kārle, Beḍsā, Kānheri, and other chaityas, the vault of the nave was ornamented by wooden ribs, as if for its support, which proves beyond doubt that these roofs were not copies of any masonry arch, but of a timber construction ; and as time wore on we find these wooden ribs copied in stone in the cave temples of a subsequent date, as in some of the Junnar caves, which were possibly of Jain origin, and in the later Ajantā chaityas.

The next step was to make the pillars of the nave vertical— for they had sloped inwards in the earlier examples—and to carve the front in stone ; and when we look at the instances of this in the chaityas at Kārle, Ajantā No. IX, and Nāsik, we trace the close imitation of previous forms. The arch represented on the front of the Lomas Rishi cave in Bihār continues to be sculptured in all its details in the vihāras and chaityas of Western India till a late date ; and the few buildings pictured on the Sānchī gateways represent it as of the same form. Whence we perceive that every feature and detail of the early caves is copied from a wooden original, and conclude that the early Hindus did not construct their architectural works—whether temples, monasteries, or palaces—in stone or brick, though for foundations and mere walls such materials may have been employed.

The façades of chaitya shrines were, from an early date, covered with sculpture—some of them very richly; and to protect them from the weather a screen was contrived and cut in the rock in front of the façade, with large windows in the upper half for the entrance of light; but, judging from what remains at Kārle and Kānheri, it seems to have been further faced by wainscoting or ornamentation of wood, or a wooden porch was added. In other cases a porch or veranda was only attached below, while a frieze projecting well forward above saved the front from the weather, as in the case of caves XIX and XXVI at Ajantā and the Buddhist chaitya cave at Ellora. The lighting of these chapel caves by a great arch over the entrance has attracted considerable attention, as being admirably adapted to its purpose. As Mr. Fergusson truly remarked : ‘ nothing invented before or since is lighted so perfectly, and the disposition of the parts or interior for an assembly of the faithful . . . is what the Christians nearly reached in after times but never quite equalled.’

The original outward form of the chaitya or chapel when constructed in wood was once a matter of some uncertainty, though the Raths at Māmallapuram (Seven Pagodas), south of Madras, supplied a key—particularly those known as Sahadeva’s and the Ganeśa Rath. Each of these, however, is represented as of several storeys, and has no proper interior, so that certain of the details were somewhat conjectural; but the discovery by Mr. Cousens of an ancient structural chaitya at Tēr in Hyderābād territory, and of others by Mr. Rea at Chezarlā and Vidyādharapuram in Guntūr District and at Guntupalle, have fully confirmed the inferences deduced from the rock-cut examples; and that at Tēr, at least, and the Vishnu Deyyanne Dewale at Polonnaruwa in Ceylon, bear ample evidence of a wooden prototype.

Besides shrines for worship excavated in the rock, such as the Buddhist and Jain chaitya caves, others usually known as vihāras were devoted to the residence of monks and ascetics. These dwellings consisted as a rule of a hall (śālā) surrounded by a number of cells (Bhikshu-gṛihas) or sleeping cubicles. The earliest of these is perhaps that discovered at Bhājā about twenty-five years ago ; it consists of a small hall about 17 ft. each way, with a veranda in front and eight cells irregularly arranged.

The Bhājā group of caves has very little figure sculpture—no other has less—and, but for the form of the chaitya cave and of the groups of dāgabas, it would be hard to ascribe it to any sect. But in this little vihāra cave, except the small rilievo dāgabas, alternating with Caryatids that support a cornice in the veranda, the sculptures are quite different from anything Buddhist. Over the pillar and pilaster capitals in the end of the veranda are sphinxes of Indian form—though derived from Persian prototypes; on the walls are five full-length armed figures, peculiarly dressed; and in the right end of the veranda are two large sculptures, one representing Sūrya the Sun-god in his chariot with attendants, and below a number of monsters; the other probably Indra on his elephant with a group of small figures. These seem to indicate a connexion with the Sauras or Sun-worshippers, who certainly formed an important religious sect in early times.

It can now no longer be assumed that all the earliest caves are of Buddhist origin; the discovery of this early excavation, together with the Ājīvika cave-shrines, suggests that other groups may have to be reconsidered. Certain of the excavations at Junāgarh are almost certainly Jain, and the Lonad cave may not be Brāhmanical. A fuller study of these, and of the sculptures in the excavations at Junnar and elsewhere, may yet lead to some changes in our classification.

Among those of Orissa we find no cave of the properly chaitya pattern ; and as an inscription on the Hāthi Gumphā cave, near the east end of the Udayagiri hill, is of the reign of a king Khāravela of Kalinga (circ. 160 B.C.), a contemporary of Śātakarṇi the Āndhra king, thus bringing it to about the date of the south gateway at Sānchī, this is one of the most important data yet found for the chronology of Indian archaeology. It upsets the whole of the theories advanced in the Antiquities of Orissa, both as to the age and sect to which these caves belonged; for the Hāthi Gumphā and most of the others are not Buddhist but Jain caves—even the figure mentioned by Rājendrāla Mitra in the Ananta-gumphā as ‘of Buddha’ is certainly not Buddhist, nor is it integral, but probably Jain of a late date; while the sculptures in the veranda show no sign of either of these sects, and over one doorway is a representation of the Sun-god Sūrya in his chariot. These Orissa caves are of early date—some or perhaps most of them Jain, and the rest of other Hindu sects ; but an intelligent survey of them has still to be made, and would be of very great importance to the history of Indian art.

Gandhāra Monuments

We come next to a class of remains found on the north-west frontier of India, and generally known as belonging to the ancient province of Gandhāra. Most probably they date from the commencement of the Christian era till about the fourth century, and belong to the Mahāyāna school of Buddhism—a form of religion differing entirely from that early Buddhist cult which had no images of gods or saints, but paid reverence to relics and sacred symbols. Indeed, we have no very clear proof that any of the early Hindu religions had iconographic representations of their divinities.

Be this as it may, we find the first representations of Buddha and the Buddhist pantheon among the sculptures of the Gandhāra monasteries ; and the influence of classical art manifested in many of these images leaves little doubt that they were modelled after foreign and Western patterns. The Graeco-Bactrian kingdom had passed away before the appearance of these sculptures ; but Ionians and other Greeks went far and wide with their merchandise and their art-productions, and Buddhist emissaries had for long travelled westwards as far as into the Levant. But the ethical precepts of Gautama Buddha failed to satisfy the followers of the various sects that acknowledged his tenets, or those who afterwards expounded and developed them ; and the Mahāyāna schools, coming in contact with Western iconography, seem to have embraced the idea of representations for their rapidly multiplying divinities— drawn from aboriginal superstitions as well as from their own legends and imaginations. The development of this pantheon is apart, however, from that of architecture.

The structures in connexion with which these sculptures are found have unfortunately been little regarded by the excavators, whose idea was mainly, if not solely, to secure as many of the sculptures as possible, irrespective of their relations to one another or to the plans, nature, use, and construction of the buildings themselves. Besides their iconographic teaching, however, the sculptures have considerable interest ; for the scenes depicted have frequently an architectural setting or background in which we find pillars, cornices, façades, &c., represented ; while from the débris, capitals, bases, and mouldings have been saved that belonged to the structures themselves, and these must have borne a close resemblance to the style of the buildings of the time.

The separate capitals are distinctly Corinthian, and evidently fashioned on western models, or by western artists. They are not pure Greek, nor yet of very early Roman type, as the little figures of Buddha among the foliage indicate : similar additions were made to capitals in Asia Minor at least as early as the time of Augustus, and were prevalent in other parts of the empire for two centuries afterwards ; and to this period we may on other grounds refer the monasteries at Jamālgarhi and elsewhere on the north-west frontier.

Again, in the sculptures we constantly find representations of architecture, in many of which the bell-shaped Persepolitan capital is represented, and this seems to have been introduced into India at an early date, and to have spread pretty widely in modified forms ; but the Corinthian form does not appear to have extended into India proper. The Persian form of capital, and such as naturally sprang from the necessities of their own wooden construction, therefore gave rise to all the capitals employed in India. The first was the earliest form used in stone architecture in India, and it continued largely in use in Northern India till after the Christian era, and among the Gandhāra monasteries so long as they existed, while in Southern India even till now modifications of wooden forms have been almost exclusively prevalent.

The façades appearing as conventional frames for sculptured scenes represent the fronts of monastic cells, and the form of the wooden framework that filled the great arched windows of the chaitya temples is that represented in these Gandhāra sculptures.

One other type of column, found at Shāhdheri, in the Punjab, is of the Ionic order—the base of the pure Attic type, and the capital with volutes. This is, perhaps, as old as any of the Gandhāra remains, and is a further indication of Western influences. But the remarkably classical character and interest of the Gandhāra sculptures generally makes their age a question of the utmost importance, and this has of late been approaching solution. The era in Western history when Greek art in its minor examples became an object of export ; the introduction of a pantheon into Buddhism ; and the dates found on, or in connexion with, several of the sculptures, if reckoned from the Saṁvat epoch of 57 B.C., all seem to converge on the period of about three centuries between A.D. 50 and 350. The monastic establishments of the Buddhists about Peshāwar and to the west and north must have been very flourishing, and their artistic ornamentation very rich— everywhere covered with carving and gilding. The sudden appearance of representations of Buddha and numerous Bodhisattvas in these establishments, and the Hellenic impress in the sculptures, may raise the question whether iconography in its wider extent, Brāhmanic as well as Buddhist, was not imported from the West.

Probably to about the same age as the Gandhāra remains belong the stūpas at Mānikyāla, between the Indus and Jhelam rivers, excavated by General Ventura and M. Court about 1830. Some of them contained coins of Kanishka, and the inference is that they date from the second century ; but only the great Mānikyāla stūpa had any portion of the outer covering left, and that seems to have been added as the facing of an envelope, 25 ft. in thickness, placed over a smaller stūpa at a much later date, possibly in the eighth century.

Gupta Architecture

By about the fifth century the architectural forms had developed in richness of decoration and variety. For convenience the prevalent style of this later age is sometimes called Gupta, for from about A.D. 319 to 520 the principal ruling dynasty in Hindustān was that of the Guptas, but the style continued long after their extinction. The columns have higher square bases than before, and sometimes a sur-base ; the capitals, which previously had a vase as the chief member, were developed by a foliaged ornament, springing from the mouth of the vase and falling down upon it from the four corners, and so lending strength to the neck whilst converting the round capital into a square support for the abacus. Often, too, a similar arrangement of foliage was applied to the vase so frequently used in early bases, and this form quite superseded the Persepolitan pillar, with its bell-shaped capital, which now disappeared from Indian art. The shafts were round, or of sixteen or more sides ; pilasters were ornamented on the shafts, and the śikharas or spires of the temples were simple in outline, and rose almost vertically at first and curved inwards towards the summit, which was always capped by a large circular fluted disc supporting a vase, whilst the surface of the tower was covered with a peculiar sort of horseshoe diaper, which was usual in early times. This style prevailed all over Hindustān, and was continued with modifications varying with age and locality down almost to the Muhammadan conquest, being often best marked in Jain structures. How far south it extended is uncertain ; for but few examples have survived of the many that must have existed previous to the fourteenth century, when the Muslim armies desolated the Deccan and ruined the Hindu shrines.

Whether the Buddhist chaitya temple, with its nave and side aisles, its sacred dāgaba in the apse, and circumambulatory passage, was derived from an early Hindu form, or vice versa, we can trace the connexion in plan between the early Buddhist shrines and the later Jain and Hindu temples. This is, perhaps, most distinctly brought out in the old Vaishnava temple at Aihole in Dhārwār, belonging to about the year A.D. 700. There the nave has side aisles lighted through the walls of the temple, which was impossible in the rock-cut chapels ; the dāgaba, or chaitya proper, is superseded by a cella for the image with a semicircular back, also separated from the outer wall by the continuation of the aisles in the passage for pradakshiṇā or circumambulation ; this passage also is lighted from without. In front is a porch, and round the whole is a raised veranda on square pillars and plain bracket capitals. How the śikhara or spire and roof of this early temple were finished, we have, unfortunately, no means of knowing, as it was long since ruined to convert it into a place of defence in the troublous times of a century ago : careful removal of the débris that covers it might, perhaps, reveal part of the structure of the spire.

If we turn next to the temple of Pāpanātha or Sangamesvara, at Pattadkal, which is also of early date—leaving out of consideration the large square outer mandapa that has been joined to it as a great portico—we find the plan almost repeated, except that the shrine or cella is now square, and the passage behind it narrower than the side aisles, but still lighted as at Aihole. The next step was to widen the temple by double side aisles, as in the temple of Virūpāksha at the same place, belonging to the Dravidian style, and built in the latter half of the eighth century ; and from this plan we see how readily the later temples all over the country—both Hindu and Jain—were evolved.

Kashmīr Architecture

From the eighth century, if not earlier, till about the Muhammadan conquest, we find in Kashmīr and the vicinity a style of architecture having in it a certain classical element, which at once reminds us of more western forms and has little if any connexion with the art of the rest of India. No sufficiently complete examination has yet been made of the examples of this style, and the hypotheses of unscientific surveyors are of doubtful value. A full knowledge of the details and peculiarities of such a quasi-classical style would afford valuable data for the history of architecture in this region.

The most notable type of this Kashmīr style is the temple of Mārtand, about three miles east of Islāmābād or Anantnāg, the old capital. It stands in a court about 220 ft. by 142 ft., surrounded by the ruins of some eighty small cells, with a large entrance porch at the east end. The temple itself was 60 ft. long by 38 ft. wide, with two wings, and consisted of two apartments— a naos and a cella. The trefoiled or cusped arch on the doors of the temple and cells is a striking peculiarity of the style, and may perhaps have been derived from the section of the Buddhist chaitya. It is used decoratively, however, rather than constructively. The pillars and pilasters of the portico and temple bear a close resemblance to some of the later forms of the Roman Doric, and have usually sixteen shallow flutes on the shafts, with numerous members in the base and capital. A triangular pediment surmounts the doorways, and on gableends or projecting faces are representations of double sloping roofs, much in the style of modern Kashmīr wooden roofs, and of which many of the temple-roofs in Nepāl are also exaggerated examples. The Mārtand temple has long been roofless ; but the probability is that when built in the eighth century (A.D. 725–60) it had a sloping wooden roof, while the cells surrounding the court were small enough to be covered by flat stone roofs. The name given it implies that it was a temple of the Sun, and we know that, till the eleventh century at least, the worship of the sun was very prevalent in the north-west of India.

It was contended by General Cunningham that this and other Kashmīrian temples of the class, at Avantipur, Bhanyur, Vāngath, Pāndrethan, &c., were Nāga or snake shrines, because he supposed they had originally been surrounded by shallow basins of water, kept at a uniform level, and approached by raised pavements across the courts. But there is no proof of this ; nor does their situation render it at all probable that the traces adduced in support of his theory were other than necessary drainage arrangements. Snake-worship, indeed, appears as early as the Yajurveda, and probably was prevalent among the original inhabitants of Kashmīr ; but surrounding water was not an indication of a Nāga shrine. The sculptures here are much decayed, and have not been represented in such detail as to indicate the divinities worshipped. We do know, however, from history that all the older examples must have been erected between A.D. 720 and 1000.

Jain Temples in Kanara

Another departure from the style of Hindu architecture has been remarked in certain Jain temples and tombs at Mūdbidri in South Kanara. These works have double and triple sloping roofs; indeed the tombs consist of a basis with quite a series of converging roofs, and remind us at once of Nepāl chaityas or Chinese towers. The whole style, in the form of the pillars of the temples, the blinds between them, and the reverse slope of the eaves above the veranda roof, is closely in imitation of wooden originals, and must have been copied either from a foreign source or from local wooden models; and one has only to notice the style of the native thatched dwellings to see whence these forms were directly derived. The interiors of the Kanara temples are often very rich in carving, the massive pillars being sculptured like ivory or the precious metals.

Associated with these temples are elegant monolithic pillars placed on square bases, the shafts richly carved and the capitals wide-spreading, and supporting, on four or five very small colonnettes, a square roof elaborately modelled. These stambhas or pillars are the representations of the early Buddhist lāts and other columns raised at their temples. We had an example of a Jain stambha in the Indra Sabha court at Ellora, and of a Brahmanical stambha in the court of the great Kailāsa temple there. The Jain example at Ellora was of the Svetāmbara sect, while the Kanara Jains are Digambaras, and the Kanarese columns belong chiefly to the fourteenth and fifteenth centuries.

Dravidian Architecture

It would be difficult to follow a strictly chronological order in noting the development of the Hindu styles, since, though they may have reacted on one another, they developed naturally among the various races and more or less independently. We might at this point, then, take up first either the Northern or the Southern developments.

Dravidian is a term applied to the people in Southern India who speak the Tamil, Malayālam, Telugu, and Kanarese languages, and is conveniently applied to the style of art practised over the larger portion at least of the area inhabited by this race. We may trace approximately the northern boundary of the style along the course of the Kistna river, to Dhārwar District, and thence south-east, past Vijayanagar and to the east of Sravana Belgola and north of Mysore city westwards to the coast. Much of the Kanarese country lies to the west of this, and part of the Telugu area to the north of it. Of course examples of the style are to be found beyond this line, and of other styles within it. In this area flourished the early dynasties of Pāndyas, Cheras, and Cholas—the first in the south, the Cholas on the east, and the Cheras in the west. These families were often at war, and by the tenth century the Cholas had overcome the Cheras and, somewhat later, the Pāndyas.

The architecture of this area, however, was essentially different from that of other regions of India, and of one type, gradually changing, but becoming worse rather than better. So far as yet known, we cannot point to any building within the Dravidian area of very early date, or before the sixth or seventh century, if indeed quite so early. Yet there may be still unnoticed structures that careful survey may bring to light, and for the present the materials are not available to enable us to trace the evolution of the style.

One of the best-known groups of monuments in the southern part of the Peninsula is that of the Māmallapuram Raths, or ‘Seven Pagodas,’ on the seashore to the south of Madras. They have often been figured, by Chambers, Goldingham, Babington, Braddock, &c., but the Government survey still remains unpublished. The raths are each hewn out of a block of granite, but none of them has ever been quite finished, nor have any of the numerous excavated caves at the same place. How this is to be accounted for we cannot explain. They have been ascribed to various dates, some too early, others very late ; the most probable view, to judge from their style and the character of the alphabets in which the inscriptions on them are carved, being that they belong to the seventh century A.D. Though evidently of Brāhmanical origin, they are certainly very like Buddhist temples as we know them from the early caves and such structural examples as have been found. But their special interest lies in their being the earliest forms of Dravidian architecture. If we compare the whole arrangement of parts in the great rath with some of the typical Dravidian temples, we at once see how the latter have been derived from the earlier type. The square raths were evidently models of Buddhist vihāras, and became the designs from which the temples proper or vimānas of Southern India were for long copied ; and further, the oblong raths, like Arjuna’s, appear to have given the first form to the great gateways or gopurams which are so notable a feature in the enclosures surrounding the larger Dravidian temples.

The next landmarks in Dravidian art are the temple of Virūpāksha at Pattadkal and the rock-cut example of the Kailāsa at Ellora. The latter is well-known as a great monolithic temple hewn out of the rock, and perhaps a century older than the first. Why we find a purely Dravidian style of temple so far to the north of the Tamil country is not readily accounted for. The site was in the Rāshtrakūta kingdom, but the style of work is that of the Cholas; and we ask, was this temple the sequence of a conquest, or of an alliance, or was the architect brought from the South?

Buildings of the Dravidian style are very numerous in proportion to the extent of the area in which they are found. The temples generally consist of a square base, ornamented externally by thin tall pilasters, and containing the cell in which the image is kept. In front of this may be added a mantapam or hall, or even two such, but they are not characteristic of the style. Over the shrine rises the śikhara, of pyramidal form, but always divided into storeys and crowned by a small dome, either circular or polygonal in shape. Another special feature of these temples is the gopurams, or great gateways, placed in front of them at the entrances to the surrounding courts, and often on all four sides. In general design they are like the vimānas or shrines, but about twice as wide as deep, and very frequently far more important than the temples themselves. Another feature is the cornices of double curve; in other Indian styles the cornices are mostly straight and sloping downwards.

The style is distinctly of wooden origin, and of this the very attenuated pilasters on the outer walls and the square pillars—often of small section—are evidences. But as the contemporary Northern styles are characterized by the prevalence of vertical lines, the Dravidian is marked by the prevalence of horizontal mouldings and shadows, and the towers and gopurams are storeyed. Then the more important temples are surrounded by courts enclosing great corridors, or prakāras, and pillared halls. In the early Kashmīr temples, in many of the Jain temples of Western India, at Brindāban, at the great temple of Jagannāth in Orissa, and others—probably in early times very many more—there are courts surrounded by cells; but in the great Dravidian temples, such as those at Madura, Rāmeswaram, Tinnevelly, Srīrangam, Tiruvallūr, Chidambaram, Kānchipuram (Conjeeveram), &c., the courts are very extensive, and are one within another. This system of enclosure within enclosure, with pillared corridors, was also carried across to Siam and Kāmboja, where the largest and most magnificently sculptured temples perhaps ever raised were executed in this Dravidian style, developed and more fully adapted to lithic materials, with complete symmetry of arrangement, a consideration disregarded in South India, where they are too often a fortuitous aggregation of parts, arranged as accident required during the long course of their erection.

The later examples of the style were overloaded with carving: every part of the building was covered with ornamentation in the most elaborate and intricate designs the artist could invent; but while the imagination may be impressed with the evidence of power and labour so lavished on ornament—much of it truly elegant—the better judgement is offended by want of architectural design in the arrangement of the constituent parts of the whole.

One of the best examples of this order is the great temple at Tanjore. It would appear to have been begun on a definite plan, and not as a series of extensions of some small temple which, by accident, had grown famous and acquired wealth by which successively to enlarge its courts, as that in Tiruvallūr seems to have grown—by a series of accretions. The body of the Tanjore temple is of two storeys and fully 80 ft. high, whilst the śikhara or pyramidal tower rises in eleven storeys to a total height of 190 ft. This dominates the gopurams over the entrances to the court in which it stands, and to an outer court, added in front of the first, but which does not, as in other cases, surround it. On the left of the principal shrine stands a smaller one of Śubrahmanya, the war-god, which is an admirable illustration of the style in its later and decorative stage, in which aspect it is as exquisite an example as exists in Southern India. The central shrine, so far as we know, was erected in A.D. 1025, and this separate one may be placed at least as late as A.D. 1150.

The Śrīrangam temple, the largest in India, is architecturally the converse of this; it is one of the latest in date, the fifth court having been left unfinished in the middle of the eighteenth century. The shrine is quite insignificant and distinguished only by a gilt dome, while, proceeding outwards, the gopurams to each court are each larger and more decorative than the preceding. The circumstances of successive independent additions and the ambitions of successive donors proved incompatible with any considered design or arrangement of parts.

The earlier Dravidian structures had lions or yālis and elephants placed as supports for pillars ; and these were gradually enlarged, made affixes to pilasters or pillars, and the animal forms multiplied and conventionalized with riders and human and other figures introduced as supporters or attendants, until about the fourteenth century or earlier they had obtained a permanent place in the architecture : at a later date figures of gods, demons, and patrons or donors sometimes took their place. Well-known examples of these occur in the temples of Vellore, Madura, Vijayanagar, and Rāmeswaram.

But though we can trace the beginnings of Dravidian art back to a pretty early date, we have as yet little help in following its development up to the thirteenth or fourteenth century ; and most of the temples, of which published plans and details are as yet available, belong to dates subsequent to the great Muhammadan invasions in the first quarter of the fourteenth century.

Chālukyan Architecture

Leaving the Dravidian, we come to the next great architectural area—that which Mr. Fergusson has called the Chālukyan style—prevailing over the whole of the basin of the Godāvari, the northern boundary being drawn roughly from the south end of the Chilka Lake in Orissa to the north-west, following for a considerable distance the course of the Mahānadī river, along the Sātpurā Hills to the Tāpti, and then south-west to the coast, eighty miles south of Surat.

The Chālukyan dynasty, whose name is applied to this style, begins to figure in the history of the Deccan early in the fifth century. About 615, a brother of Pulikesin II, who ruled at Bādāmi, set up an eastern kingdom at Vengi, on the lower Godāvari, and about the same time another branch of the family became established in the south of Gujarāt. The area of the Chālukyan style, then, includes the Hyderābād territory, the Central Provinces, Berār, and the Marāthī and part of the Kanarese-speaking Districts of the Bombay Presidency.

In the middle of the eighth century the Rāshtrakūtas dispossessed the Chālukyas of their territories and made them feudatory ; but late in the tenth century they reasserted their power, which continued for about two centuries, and was finally overthrown in A.D. 1184 by the Hoysalas, who next ruled the south-west of the earlier Chālukya domain, while the Kākatīyas had established themselves a little earlier at Warangal to the east. Both these kingdoms were conquered about 1320 by the Muhammadans.

The earliest temples within this area, however, are not very clearly marked off from the Dravidian and the more northern styles—some of them have distinctly northern spires, and others are closely allied to the southern style ; and it was perhaps only gradually that the type acquired its distinctive characteristics. Till a late date we find temples with towers differing so little in form from Dravidian vimānas, that, other details apart, they might readily be ascribed to that order. Unfortunately many of the finer examples must have perished during the Musalmān invasions and during the rule of the Muhammadan dynasties of Bijāpur, Gulbarga, Bīdar, Hyderābād, and Burhānpur, and, as we might expect, round these cities most of the earlier works have disappeared. Still in Mysore, Dhārwār, and Belgaum, as well as in Berār and the Marāthā districts, sufficient remains still exist to illustrate the various developments of the style.

The old temple of Pāpanātha at Pattadkal presents a curious combination of styles. The body of the temple is Dravidian, and must have been a fine specimen, of as early a date as the early part of the eighth century ; but the śikhara is a curious approximation to the form of the early Northern Hindu or Indo-Aryan order, while in details the temple shows a strong leaning to the Dravidian. One is almost tempted to suppose that the architect of the temple had died and left the spire to another who, having a preference for the northern form, had tried to adapt it to a Dravidian substructure. The temple of Virūpāksha at the same place is an excellent example of the pure Dravidian, built about A.D. 740, while close by is another that might have been transferred from Orissa.

On the temple of Kuchchimalligudi at Aihole is a somewhat similar śikhara. This temple is small and plain, with a sloping roof over the side aisles, and belongs to about the seventh century. The Meguti temple also at Aihole must have been a fine work, but unfortunately it has lost all above the wall heads.

Among Chālukyan temples a prevalent form is that of three shrines round one central mandapa or hall. The arrangement for supporting the roofs of the halls almost always follows the Dravidian mode of four pillars, or multiples of four, in squares ; the device of twelve columns so disposed in a square that, omitting the corners, the remaining eight could be connected by lintels to form the octagonal base of a dome, is almost unknown. It is employed, however, in the outer hall of the great temple at Hāngal. In the Dravidian and northern temples the projections on the walls are generally formed by increments of slight thickness added flatly to their faces, and, however thick, they are so placed as to leave the true corners of the shrines, &c., more or less recessed.

In the Chālukyan temples the corners are often made prominent by increments placed over them, or the whole plan is star-shaped, the projecting angles having equal adjacent faces lying in a circle, as in the temple of Belūr in Mysore, built about A.D. 1120; in that at Somnāthpur on the Cauvery, thirteen miles east from Mysore city, finished in A.D. 1270; in that of Kaitabheśvara at Halebīd—lately the gem of Chālukyan art, now, alas ! a shapeless ruin ; and in a modified form in that of Galteśvara in Gujarāt. The great temple of Hoysaleśvara at Halebīd, begun about A.D. 1250, was left unfinished at the Muhammadan conquest. It is a double temple, measuring 160 ft. by 122 ft., and is covered with an amazing amount of the richest sculpture. But the spires were never raised over the shrines. The Kedāreśvara temple at Balagāmi is perhaps one of the oldest of the style in Mysore, and there are other good examples at Kubattur, Harnhalli, Arsikere, Kōravangala, and elsewhere—surveys of none of which have been published. But the plans vary greatly. The śikhara did not preserve the southern storeyed form but was rather stepped, forming a square pyramid with breaks corresponding to the angles in the walls, and with a broad band answering to the larger face in the middle of each exposed side of the shrine.

Some of the details of this style are very elaborate : in fact, most of the finer temples were completely overlaid with sculptural ornament. The pillars are markedly different from the earlier Dravidian forms ; they are massive, richly carved, often circular and highly polished. Their capitals are usually spread out, with a number of circular mouldings immediately below ; and under these is a square block, while the middle section of the shaft is richly carved with mouldings in the round. In many cases the capitals and circular mouldings have been actually turned in a sort of lathe, the shaft being held in a vertical position. They are almost always in pairs of the same design, the whole effect being singularly varied and elegant.

As we see at Ajantā and elsewhere, doorways were, from a very early period, objects on which much artistic skill was lavished ; and this taste was maintained in the utmost elaboration bestowed on the sculptures surrounding the doors of Dravidian and Chālukyan shrines. Pierced stone windows were employed in Dravidian temples at Pattadkal, Ellora, and other places ; but the richly carved and highly ornamented pierced windows belong specially to this style. Generally the temples stand on a terrace from 10 to 15 ft. wide, quite surrounding them, and from 3 to 6 ft. in height—a feature which adds considerably to the architectural effect. The buildings were erected without mortar, and, in the earlier examples at least, the joints were carefully fitted. The whole was covered with sculpture, often of geometric and floral patterns, intermixed with numerous mythological figures ; and, in the later examples, the courses of the base were carved with the succession of animal patterns prescribed for them in the Śilpa Śāstras. This is very fully exemplified in the great temple of Hoysaleśvara at Halebīd. This temple, though unfinished, is one of the most remarkable in India, and, in an artistic sense, is unmatched in the variety of its details and the wild exuberance of fancy displayed in its ornamentation ; while the combination of horizontal with strongly-marked vertical lines and the play of outline and of light and shade are hardly surpassed in any style.

Owing to our still imperfect knowledge of the antiquities in the Hyderābād territory, we can as yet refer only to a few, such as the Hanamkonda temple near Warangal, one at Buchanapalli to the west of Hyderābād, and others at Ittagi in the south-west, at Nilanga, Nārāyanpur, &c., though we know there are many other ruins all over the area that belong to this style. In the south of the Bombay Presidency we may instance those at Dambal, Rattihalli, Tiliwalli, and the large temple at Hāngal ; in the Bellary District of Madras, at Magala, Kuruvatti, Nīlagunda, &c. ; and in Mysore those at Belūr, Somnāthpur, Halebīd, Balagāmi, Kōravangala, Harihar, and others.

Indo-Aryan Style of Architecture

Of Northern India, or that area which is usually designated as Hindustān, lying to the north of the Tāpti and the Mahānadī rivers, the Hindu architectural style or styles, besides being more widely spread than either of the preceding, is also more varied and wanting in marked and characteristic individuality. Mr. Fergusson, whose nomenclature has necessarily become impressed upon Indian archaeology, has applied the term Indo-Aryan to the Hindu style prevailing over this area ; and it would be difficult to find a better, since this type of architecture was ‘invented and used in a country which Aryans once occupied, and in which they have left a strong impress of their superior mental power and civilization.’

Within this large area there are, of course, many examples of other styles, whilst south of it there are also buildings belonging to this more northern type. At Pattadkal, for example, the temple of Pāpanātha, as already noticed, has a śikhara belonging to this Indo-Aryan style ; whilst at the same place is another temple of the early northern class, as are also the temples of Kuchchimalligudi at Aihole, the smaller temple at Hāngal, and others in the northern Kanarese districts. This sporadic appearance of temples of a style removed from their proper area may be accounted for in various ways : great temples were constantly being visited by pilgrims on their way from one shrine to another, and the repute of any new fane was soon spread over all India ; and thus, when a prince undertook to build a new temple, an architect (sthapati) of acknowledged ability might occasionally be sent for from the most distant province, and engaged to design the work, which, of course, would be in his own style. This, too, may possibly account for the mixture of styles we find in some temples.

But under this Indo-Aryan style are classified monuments of very various orders, and we might, if necessary, separate them into two or more distinct types. The characteristic that first appeals to our notice is the curvilinear spires of the temples, and next to this the absence of that exuberance of sculpture seen in the great Chālukyan temples of the South ; while in many cases, as in the Jain temples, a greater central area has been obtained by arranging twelve columns so as to support a dome on an octagonal disposition of lintels. The shrines and mandapas are square, and only slightly modified by additions to the walls of parallel projections, which, in the earlier examples, were thin ; the walls were raised on a moulded plinth (pīṭha) of some height, over which was a deep base, the two together rising, roughly, to about half the height of the walls. Over this is the panelled face of the wall, usually of less proportionate height than in the Chālukyan style, and though devoted to figure sculptures in compartments, the tall, thin pilasters of the southern style have disappeared. Over this is the many-membered architrave and cornice, above which rise the spire and roof. The spires follow the vertical lines of the wall, and present no trace of division into storeys, but vary in details with the age. In the earlier, we have a broad band in the centre of each face, running up and curved inwards towards the summit, which was crowned by a large, fluted, circular block called amalaśilā— probably mistaken for āmalaka (Phyllanthus Emblica),—the word amala meaning ‘pure,’ ‘shining.’ The finial over this is the shape of a vase, known as the kalaśa or karaka. The central band on the śikhara was carved, usually with a reticulated pattern composed of minute arches, but occasionally interrupted by bands of larger ornament, as at Kanārak in Orissa, and on some of the Bhubaneswar temples. The corners of these spires were generally in courses, also carved in successive patterns, each third or fourth course being alike, and one of these was usually fluted if not also circular like the āmalaka.

What is known as the Jain style of architecture in Western India is a development or variety of this Indo-Aryan order, and was used by Hindus and Jains alike all over Rājputāna, Mālwā, and Gujarāt. It was employed in its most ornate form by the Jains in their famous marble temples on Mount Abu, and by both Jains and Hindus at Nāgdā near Udaipur, where is a group of little-known but remarkably fine deserted temples. At Girnār also and Śatrunjaya in Gujarāt are clusters of temples of this order; but as they are mostly restorations of earlier shrines destroyed by the Muhammadans, they are much less lavishly ornamented with sculpture. One of the most striking features of the style is the richly carved domes over their mandapas or porches. Nothing can exceed the elaboration and delicacy of details in the sculptured vaults of the temples at Abu and Nāgdā. These, with the diversified arrangement of variously spaced and highly ornamented pillars supporting them, produce a most pleasing impression of symmetry and beauty.

The plain of north Gujarāt was so often devastated by war from the eleventh to the fifteenth century that its more notable temples have perished, though the once magnificent Sun temple at Mudherā still witnesses in its ruins to the architectural style and grandeur of the period when it was erected. What fragments still survive there have been illustrated in the volume of the Archaeological Survey of that district.

Another considerable group of from thirty to forty temples in this style is found also at Khajurāho in Bundelkhand. In the early part of the last century they were much more numerous than now—many having been removed for building material. They belong to both the sects of Hinduism as well as to the Jains, and date mostly from the tenth and eleventh centuries. The older temples are covered outside and inside with the most elaborate sculptures, and architecturally they may justly be regarded as ‘the most beautiful in form as well as the most elegant in detail' of the temples of Northern India ; indeed, the only others that can well be compared with them is the earlier group at Bhubaneswar in Orissa.

The temples at Bhubaneswar exhibit the Indo-Aryan style perhaps in its greatest purity, and they differ very markedly from those in the West in being almost entirely astylar—pillars having been introduced in later additions. They have the early form of śikhara—nearly perpendicular below, but curving in near the summit ; and the crowning member has no resemblance to anything like the small domes on Chālukyan spires. The surface of the sides is entirely covered with carving in the most elaborate style, every single stone having a pattern engraved upon it ; and much of the sculpture on the earlier temples is of considerable merit and much beauty of design. The older and finer ones were erected probably in the seventh and eighth centuries, and the series was continued, by additions, down to the eleventh. From the light they seem calculated to throw on the history of art, no temples in India probably would better repay a complete scientific survey, and an attempt was made in 1869 to supply this want, but the result was an unfortunate failure. The drawings made were mostly of mere details, chosen without sequence or meaning ; and no plans of any of them were prepared until a second effort was made three years later, when some ground-plans on a small scale and of doubtful accuracy of detail were drawn and printed together, on two plates in the second volume of the Antiquities of Orissa.

The temple of Kanārak, known as the Black Pagoda, on the coast of Orissa, appears to belong architecturally to the ninth century, though it has been by some attributed to the reign of Narasimha in the thirteenth—possibly because he repaired it or made some addition. A detached mandapa that stood in front of it, occupying a corresponding place to that at Mudherā, was removed to Purī, in the eighteenth century, by the Marāthās. A corner of the śikhara was still standing in 1839, but within the next thirty years had disappeared ; and the great lintel over the entrance to the principal hall, carved with the Navagraha or nine planetary divinities, with other parts about the doorway, had fallen or were removed, and an abortive attempt was made to carry the lintel to Calcutta. Now this famous monument, which for its size is 'the most richly ornamented building—externally at least—in the whole world,' has lately been treated in a way that has very seriously injured it. The historical and artistic interest of it and of the two groups previously noticed cannot be fully estimated until complete surveys have been published with detailed plans and sections.

In later examples the spire is still a square curvilinear pyramid, to the faces of which are added smaller copies of the same form, carrying up the offsets of the walls ; and in some examples these are multiplied to an extraordinary extent. The earlier temples were apparently astylar, then—like the southern forms—with columns arranged in the mandapas in groups of four, and later, especially in Western India, the larger domes on twelve pillars formed the central area of the halls. These mandapas in early examples were roofed with long, sloping slabs ; but, to provide for carved conical roofs inside, their outer forms represented courses of masonry, which were carved, —as we find in the older temples of Kanārak and Bhubaneswar, in the mediaeval shrines at Ambarnāth, Baroli, Khajurāho, Abu, and Chitor, and in the more recent forms at Nāsik, Benares, Udaipur, Śatrunjaya, &c.

Muhammadan Architecture

What is popularly known as Saracenic architecture is the style which was adopted by the Muhammadans when they became the ruling race in India, from about the end of the twelfth century. But while largely applied to mosques and tombs, it varied much at different periods and under the various local Muslim dynasties in different parts of the country. The Delhi emperors, for the first three centuries of their domination, were of Turkī or Pathān stock, and were succeeded in the early part of the sixteenth century by the Mughal dynasty founded by Bābar, when the latter materially influenced the architectural style of the previous dynasty.

Then there were local kingdoms which had styles more or less their own : Bengal became a separate kingdom at the beginning of the thirteenth century ; the Bahmani dynasty at Gulbarga and Bīdar dates from the middle of the fourteenth century ; the kingdoms of Jaunpur, Gujarāt, and Mālwā from about A.D. 1400 ; Bijāpur and Ahmadnagar from about 1490, and Golconda from some twenty-two years later. Exclusive of other varieties of less extent and individually not so distinctly marked off, we have thus some ten more or less fairly different styles of Saracenic structures.

In all the varieties, the distinctive features of each may be traced at once to the employment of local native Hindu workmen, and the use of their own materials and methods. The conquerors were of Turkish descent, and apparently had strong architectural instincts ; accordingly they began at once to found mosques for the glory of Islām and to mark their triumph over the idolaters.

Their first mosques were accordingly constructed of the materials of Hindu and Jain temples, and sometimes with comparatively slight alterations. The colonnade of a temple court required little more than a wall on the west side fitted with mihrābs or kiblas, and the removal of the idol shrine, to adapt it for a mosque. In other instances they demolished the temples, and, by adding to the height of the columns, obtained the greater elevation and airiness they required. Thus in the great mosque at Ajmer, erected between 1200 and 1230, three tiers of pillars are piled above one another, and the roof is largely formed of slabs from the temples to which the columns originally belonged : in plan it is an adaptation of that of a Jain temple. And in the still larger mosque at the Kutb Minār near Delhi, built about the same time—so far as it remains—we have the same features ; while in both a richly carved screen of pointed arches was added in front, and the whole enclosed by massive walls. The arches, which the Muhammadans seem to have insisted on, are built after the system of the Hindu domes, of horizontal courses as far as practicable, and then closed by long slabs meeting at the apex—an evidence that the workmen, being Hindus, were unused to building arches and modified their own methods to meet the new form of construction. The arches are circular segments up to about two-thirds of their height, and constructed in horizontal courses. Above come one or two half voussoirs, and the head is closed in by sloping slabs.

The Kutb Minār itself is one of the finest pillars in the world. Erected by order of Altamsh (not by Kutb-ud-dīn Aibak), it was probably completed about a.d. 1231, which is the date of an inscription on the adjoining mosque. It is still about 240 ft. in height and ornamented by four projecting balconies with richly sculptured and engraved belts between, and the whole of the lower three storeys are cut up by twenty-four projecting ribs that add greatly to its beauty. Behind the north-west extremity of the Kutb mosque is the tomb of Altamsh, who died in 1236 ; it is thus perhaps the earliest Musalmān tomb to be found in India, and is profusely ornamented with carving, and altogether of extreme beauty in its details. A still finer example of the Pathān style is to be seen in the eastern annex of the mosque—the splendid southern gateway or Alai Darwāza, built in 1310 : this and the now ruined tomb of ‘Alā-ud-dīn Khilji, erected soon after, mark the style at the period of its greatest perfection : indeed during his reign (1296–1316), and that of Fīroz Shāh Tughlak (1351–88), palaces, forts, mosques, mausoleums, baths, universities, and all sorts of public and private buildings multiplied in an extraordinary manner. But after the death of the former, for fully a century, the Pathān buildings are marked by a stern simplicity of design and a solemn gloom and nakedness, in marked contrast to the elaborate richness of ornamentation of the preceding period. In 1321 New Delhi or Tughlakābād was founded by Ghiyās-ud-dīn Tughlak I, all the buildings of which are characterized by a severe simplicity, as contrasted with those of the preceding century. The sloping walls and massive solidity of the founder’s tomb, together with the heavy towers of the fortified citadel surrounding it, form an unrivalled model of a warrior’s tomb.

But by this time the builders had got rid of the imitation arch of the Hindus, and had learnt to construct true arches, and their architecture had now developed into a new and complete style of its own. To this style belong many of the finest mausoleums of Northern India. Like that of Sher Shāh Sūri (1540–1545), built in a spacious tank at Sasarām, which is one of the best examples, they are very often octagonal, with an outer veranda and crowned by a dome over the inner walls, and the whole surrounded by a square terrace ornamented by small pavilions at the corners. Round the drum of the domes, also, are placed octagonal kiosks that accentuate the beauty of the outline. Other examples are numerous, among which that of ‘Alā-ud-dīn Ālam Shāh at Tijāra in Alwar territory, and of Mubārak Shāh (1540-45) at Kotila near Old Delhi, may be instanced.

The Pathān mosques of the fourteenth century were as severe in the simplicity of their style as their tombs, as we may see in the Kalān mosque at Delhi, finished in the time of Fīroz Shāh in 1386. In the first quarter of the fifteenth century, however, a reaction had set in, and the later style was hardly less rich and much more appropriate for its purposes than the earlier in the end of the twelfth and early thirteenth century. The façades of the mosques became more ornamental, were often encrusted with marble, and usually adorned with rich and beautiful sculpture. Minarets had not become a feature of the mosques, and the corners of the structure were relieved by little kiosks instead. At Ahmadābād, minarets came into use for the mu‘azzin in the fifteenth century. The body of the mosque became generally an oblong hall, with a central dome flanked by two or four others of the same span, but not so lofty, and separated from it by an arch whose mouldings formed a principal feature of the building. The pendentives are remarkable for their variety of design and elaborateness of detail. The style in the later Pathān period, as Mr. Fergusson has remarked, was marked by a return to the elaborateness of the past, but with every detail fitted to its place and its purpose, ‘and we recognize in this last development one of the completest architectural styles of the world.’

The Sharkī Style

In 1397 Khwāja Jahān, who governed Jaunpur, assumed independence, and founded the Sharkī or Eastern dynasty, which ruled there for about eighty years. Of the palaces or public buildings of the Sharkī dynasty no trace is left; for Sikandar Lodī ibn Bahlol razed them all to the ground, his courtiers using the materials for building their own mansions, and what has come down to us is little more than three great masjids—the Jāmi, Atala, and Lāl Darwāza —besides a fort and bridge with a number of tombs. Of these mosques the cloisters that surround the open courts and the galleries within are almost purely Hindu in style, with short square pillars and bracket capitals supporting horizontal lintels, and roofs formed of flat slabs; but the gateways and principal features of the masjids are in a completely arched style. There is sufficient evidence that, for the earlier of these at least, the materials of Hindu temples were largely used, and the workmen were probably mostly Hindus by birth and inclined to the old trabeate forms. The fusion of the two styles was thus incomplete. The masjid proper consists of a central square hall covered by a lofty dome of the whole width of it, in front of which stands the great propylon, of massive outline and rising to the full height of the central dome. This propylon had a large recessed arch between the two piers at the sides, in the lower portion of which was the entrance to the mosque, whilst the upper formed a pierced screen. On each side of the dome is a compartment divided into two storeys by a stone floor supported on pillars, and beyond this, on each side, is a larger apartment covered by a pointed ribbed vault. The gateways into the courts on the three sides were only copies on a smaller scale of the propylons of the mosques.

The whole of the ornamental work on these mosques has a character of its own, bold and striking rather than minute and delicate, though in some of the roof-panels there are designs that may bear comparison with similar patterns in Hindu and Jain shrines. The mihrābs are marked by their severe simplicity; they are simply patterns of the entrances and of the niches on the outer walls, with flat backs and structural arches over them. They form a link, however, in the evolution of the favourite form under the Mughal rule.

The Sharkī buildings have been pretty fully illustrated in the volume of the Archæological Reports on the subject, and need not be further described here. They afford a marked expression of strength combined with a degree of refinement that is rare in other styles. Examples of this style are met with also at Benares, Kanauj, and other places within the Jaunpur kingdom.

Mālwā

Dilāwar Khān Ghorī, the governor of Mālwā, assumed independence in 1401, and the state continued under its own rulers till 1531, when it was conquered by Gujarāt, and was finally re-annexed to imperial rule under Akbar in 1570. The capital of the province had been first at Dhār; but Dilāwar Shāh resided a good deal at Māndu or Māndogarh, about 22 miles south from Dhār, placed on an elevated plateau detached from the mainland by deep ravines and surrounded by walls on the brink of the cliffs; and under the second king, Hoshang Shāh, Māndu became the permanent capital. During his reign (1405–34) the most important of the buildings were erected. Among these, which are in a modified form of the Pathān style, the finest is the great Jāmi Masjid, which was only finished by Mahmūd Shāh I in 1454. It covers a nearly square area, measuring 290 ft. from east to west by 275 ft. from north to south, exclusive of the porch on the east, which projects about 56 ft. Inside, the court is an almost exact square, about 163 ft. each way, surrounded by arches on each side of about 12½ ft. in span, standing on plain square piers 10 ft. high, each of a single block of red sandstone; behind these are triple arcades on the north and south, a double one on the east, and on the west the masjid—five arcades in depth, and having three great domes on the west side. This court, in its simple grandeur and expression of power, may be taken, as Mr. Fergusson has well remarked, as one of the very best specimens of this style now to be found in India.

The tomb of Hoshang Shāh, adjoining the west side of the Jāmi Masjid, is a fine bold specimen of a Pathān mausoleum. It was revetted both outside and inside with white marble, which, however, has much peeled off, and is now being restored. Near the mosque, on the west, is a splendid hall, 230 ft. in length by 28 ft. wide, supported by eighty-four pillars, in three rows, of which the pattern must have been suggested by the usual forms in Hindu and Jain shrines ; only on the capitals the kīrttimukh or horned face of the Hindus has been hewn into a group of leaves of the same outline. The porch on the north side of it is purely Hindu in style.

The Delhi gate on the north of the fortifications, by which they are entered, has been a fine lofty structure, though now much ruined : it also is purely Pathān in style, but unusually elegant in proportions and decoration. Close by this gateway are the remains of an enclosure, within which are the ruins of the royal palaces—the Jahāz Mahal, the Hindola Mahal, the Tawilī Mahal, and the Nahār Jharokhā, with the Champā well or baoli, &c. The palaces are specially interesting as remaining examples of Pathān secular architecture, though, unfortunately, no proper survey of them has as yet been published, and it is hardly possible from such sketches as have from time to time appeared to form a just estimate of them or their arrangements.

The Jahāz Mahal, the ‘ship’ or ‘water-palace,’ built between two great tanks, is the chief of these. It is a massive structure, the eastern façade being about 40 feet in height, in the centre of which is the arched entrance, faced with marble, and still in fair preservation ; over it is a projecting cornice supported on brackets, above which is a bracketed balcony under an oblong pavilion. In the front of the lower storey on each side are five arches under a deep overhanging cornice, and over each end of the façade is a domed pavilion. On one side is a ruined wing of the palace branching off from it ; and on the opposite side were other apartments and a stair leading up to the roof. Seen from the west, where it overhangs the lake, this is altogether a striking building, one of the most remarkable of the period, and well worthy to be the residence of an independent Pathān chief.

North of this, about a quarter of a mile, stands the Hindola Palace, which, with its massive masonry, is in rather better preservation than the others. The sloping buttressed walls, projecting balconies, and deep-set windows of this fine building present an appearance of great strength ; and the great hall within, 108 ft. in length by 22 ft. wide, its roof supported on arches, was a splendid apartment. To the north of this were the zanāna apartments : and at some distance to the west are the large underground cisterns and tah-khānas, or hot-season retreats, of the Champā well or baoli. These indicate the care and taste bestowed on such appendages of a Muhammadan palace 500 years ago.

The Nahār Jharokhā Palace is to the north of the Hindola Mahal, and also within the walled enclosure ; and outside is Dilāwar Khān Ghori’s mosque, the oldest in Māndu (1405), constructed of materials taken from Hindu or Jain shrines. It has, however, a simplicity of structure about it characterizing it as a typical Pathān work.

About eighty yards to the south of the Jahāz Mahal is the Tawilī Mahal, a three-storeyed building, with its rows of lofty Saracenic arches below deep stone eaves and heavy windowless upper storeys. It lies across a beautiful foreground of water and ruins.

About a mile and a quarter south of the Jāmi Masjid, on the east of a great talāv or lake, is a group of buildings among which is the so-called Dhāī-ka Mahal, a substantial square tomb, and the Chhotā Jāmi Masjid of Malik Mughīs-ud-dīn, built in 1432 largely with materials taken from Hindu or Jain shrines, as the pillars in the porch and colonnade bear witness. This mosque must have been one of great beauty and interest, its entrance porch, though in ruins, being still an elegant structure. Opposite to it is the ruin of Malik Mughīs-ud-dīn’s palace, and also, a little farther off, his tomb, the dome still enlivened by a belt of blue enamel.

Still more to the south are the remains of the palace of Bāz Bahādur, the last king of Mālwā,—which was built apparently by Sultān Nāsir-ud-dīn Khiljī in 1509, and of which some portions of the courtyards remain intact, as well as the cupolas over the colonnades. On the hill above is what is now known as Rūpmatī’s chhatri, still in fair preservation.

Here, as elsewhere, the available materials have exercised a marked influence upon the architecture ; the prevalence of a red sandstone is emphasized in the piers of the Jāmi Masjid —more than 300 of them being each of a single block of this material ; and for more decorative purposes marble, both white and coloured, was freely used to revet the walls and piers. An adequate survey of the remains at Māndu, and of a few others of the same age in Mālwā, would form an interesting monograph on this style of architecture, together with its constructional methods, which deserve attention and study. We have here a strictly arcuate style, without admixture of the general trabeate structural methods followed by the native Hindus ; and while at Jaunpur and Ahmadābād, at the same period, we find the strong influence of native methods copied in the Muhammadan architecture, at Māndu the borrowing or imitating of such forms seems to have been suppressed, and the builders clung steadily to the pointed arch style, without any attempt, however, at groining—so successfully employed at a later period by the Mughal architects.

Bengal

The Bengal province was placed under governors appointed from Delhi as early as A.D. 1194, the first of these being Muhammad-i-Bakhtyār, under the emperor Kutb-ud-dīn Aibak. In 1282 Nāsir-ud-dīn Bughra Khān, the son of the emperor Ghiyās-ud-dīn Balban, was appointed governor and the office became hereditary in his family. In 1338 Fakhr-ud-dīn Mubārak slew the governor Kādir Khān and assumed sovereignty, but was successfully opposed by ‘Alī Mubārak, who reigned from 1340 to 1346. He was assassinated by Shams-ud-dīn Ilyās Shāh, who then defeated Ikhtiyār-ud-dīn Ghāzī Shāh in 1352, and may be regarded as the founder of the Purbīya dynasty, which ruled Bengal for about a century and a half, or till 1487, when the throne was usurped by Habshīs and subsequently by ‘Alā-ud-dīn Husain Shāh III. But in the reign of his son, Mahmūd Shāh III, Sher Khān, the Afghān ruler of Bihār, invaded Bengal in 1537 and laid siege to Gaur, which was then completely sacked, and this once great and wealthy city, thus plundered, began to decay and its buildings were neglected. The kingdom was annexed by the great Akbar in 1573, and the city was depopulated by plague in 1575.

But long before the advent of the Muhammadans, Gaur, or Lakhnautī, had been the capital of the Pāl dynasty in the ninth and tenth centuries and of the Sens of Bengal in the twelfth century; it was then of great extent, and doubtless contained many temples and palaces that were destroyed by the Muslims. As the country is practically without stone, they would be mostly of brick and would afford material for the conquerors ; but probably pillars, images, and details were of that hornblende, basalt, or hard potstone, which takes a high polish and is employed in the later structures. In the Eklākhi mosque or tomb (A.D. 1414) at Pandua, and in the Chhoti Sonā Masjid at Gaur (circ. 1500), the stones used have largely been taken from earlier Hindu buildings. The use of brick forced the builders to elaborate a local arched style of their own, and further, as Mr. Fergusson pointed out, to introduce a new mode of roofing, which, though but little agreeable to our tastes, came to be regarded by the natives, whether Hindu or Muhammadan, as a most elegant form, and spread, in the seventeenth century, as far up the Gangetic valley as Delhi, and a little later, even to Amritsar. The curvilinear form given to the eaves, descending at the corners of the structure, was almost certainly suggested by the form of the huts, constantly roofed with bamboo and thatch, in which the Bengalis always use a curvilinear form of roof.

The erection of large buildings of brick required heavy piers for the arches and thicker walls than those constructed entirely of stone. Such piers and walls, when enriched by a casing of moulded tiles, would appear still heavier ; and for tiles, when opportunity offered, a facing of carved stone might be substituted. This was doubtless the kind of buildings before the Muhammadan conquest, and the style was only modified by that event ; hence this Bengal style is not like any other, but a purely local one, with heavy short pillars faced, at least, with stone, supporting pointed arches and vaults of brick.

Ilyās Shāh (1345) made Pandua, to the north of Gaur in Mālda District, his capital, and there his son and successor, Sikandar Shāh (1358–89) built the great Adīna Masjid within the first ten years of his reign. It measures nearly 500 ft. in length by 285 ft. from east to west, containing in the centre a court measuring nearly 400 ft. by 154 ft., surrounded by a thick curtain wall of brick, pierced by eighty-nine arched openings, with one on the west side much wider and more dignified than the others. The roof was supported by 266 stone pillars—the cloisters on three sides had a double row of pillars in each— that on the west, or the mosque proper, having four rows of pillars and thirty-five mihrābs on the west wall. The pillars are about 2 ft. square at the base and 10 ft. 5 in. high, some consisting of one block of hornblende, but wanting in variety of pattern. North of the central mihrāb is a platform known as the Bādshāh-ka takht, that is, the mulūk khāna, or royal gallery. This is supported by twenty-one short pillars of a much heavier form, and has others of a better type above. The roof consisted of 378 domes, all of the same form and construction. Such a design has little architectural merit, though its size and the elegant richness of its details make it an interesting study; and the same character runs through most of the works of these Purbīya rulers.

Next to the Adīna Masjid comes the Eklākhi mosque or tomb, at a distance of about two miles to the south-west. It is said to be the tomb of Ghiyās-ud-dīn Azim Shāh (1390–7), but there is no inscription to show this, and it may have been the work of Jalāl-ud-dīn Muhammad Shāh (1414– 43), who was a great builder. It is 80 ft. square and covered by one dome. Much of the materials have been taken from Hindu temples, the structure being built of hornblende slabs and brick, with much embossed brick used in the decoration. It has richly carved buttresses at the corners, reminding one of the bases of minarets, but they had only a capstone above the level of the roof, the corners of which curve downwards on each face. Though much smaller, this was altogether a bolder and architecturally finer structure than the Adīna Masjid.

To the south-west of the preceding is the Sōnā Masjid, a small but once elegant mosque, built of hornblende or, perhaps, basalt. It has five arched doorways, and was roofed by fifteen brick domes, but the trees that were allowed to take root in them have wrought its destruction.

Five miles south of Pandua is Mālda, where also are remains of mosques, tombs, and gateways belonging to the times of the Purbīya rulers.

Among the ruins of the once great city of Gaur, six miles south-east of Mālda, are more interesting remains, of the same style and period. Of these may be instanced the Dākhil or Salāmī gateway, the north entrance into the fort, supposed to have been built by Rukn-ud-dīn Bārbak Shāh (1460–74)—as grand a structure of the kind as is to be found anywhere. It is built of small bricks, decorated with embossed terra-cotta facings, is 70 ft. across the façade, with a depth of 110 ft., having rooms for the guard on each side the passage and lofty towers at the corners, whilst a recessed arch 34 ft. high encloses the entrance on each face.

Just outside the fort to the east is a minār, about 85 ft. in height and 21 ft. in diameter, which, for two-thirds of its height, is a polygon of twelve sides, and above that contracts and is circular. Probably a platform some 15 ft. in height once surrounded the base, but it has entirely disappeared, and the door is now at a considerable height from the ground. Inside, a spiral stair leads to the small chamber on the summit, once roofed by a dome. At one time this tower was encompassed by a revetting of stone, and the cupola was covered with blue and white tiling, but now these are entirely gone. Indeed, most of the tiles with which the mosques and tombs at Gaur and Pandua were originally ornamented have long since disappeared, for (as mentioned in Grant's Fifth Report, p. 285) the Nizāmat daftar contained an entry of 8,000 rupees under the head of kīmat khishtkār, annually levied from a few land holders who had the exclusive right of 'dismantling the venerable remains of the ancient city of Gaur or Lakhnautī, and conveying from thence a particular species of enamelled bricks, surpassing in composition the imitative skill of the present race of native inhabitants.'

This minār Mr. Fergusson regarded as a Jayastambha or pillar of victory, comparing it with the Kutb Minār at Delhi, that raised by Ghiyās-ud-dīn Balban at Koil in 1253, and one at Daulatābād with a high marble platform round it. This was erected by Saif-ud-dīn Fīroz Shāh II (1488–90); and it may be remembered that about 1443 Mahmūd Khiljī, after his victory over the Rājputs at Kumbhalmīr, erected a tower of victory at Māndu faced with marble; and again on Rānā Kūmbha gaining a victory over Mahmūd he erected the fine Kīrtti-stambha at Chitor, 1448–58.

Among the mosques at Gaur the Kadam Rasūl Masjid at the south-east gate of the fort is the only one at all cared for, because it contains a stone bearing the supposed impression of Muhammad's footprint, brought from Madīna by Husain Shāh (1493–1519). The mosque was built by his son Nusrat Shāh, A.D. 1530. It has three arched entrances in the front, separated by massive piers, and is about 35 ft. in length inside; the central portion of the roof is covered by a single dome, and it had four minarets at the corners, the upper portions of stone, of which only one survived into the last century. The façade is relieved by horizontal mouldings and panels of moulded brick, and string-courses of the same extend its whole length.

South from this is the half-fallen Tāntipāra mosque, which must have been a building of considerable architectural merit. The mihrābs are elegantly carved, the roof was supported by massive stone pillars, and the façade richly decorated with ornamental terra-cotta facing. It is ascribed with probability to Yūsuf Shāh, about A.D. 1480.

Southwards from this again is the Lattan or ‘painted mosque,’ so called from its walls being cased inside and out with glazed tiles of different colours—dark blue, green, yellow, and white, admirably arranged for effect in varied patterns. The exterior has been much defaced, but inside, if still in fair preservation, it is fast going to decay. It has four entrances in each end as well as in front ; and eight double buttresses relieve the exterior by their mouldings and encrusted tile decoration. It is also ascribed to the reign of Yūsuf Shāh in 1475.

The Kotwālī Darwāza is a handsome and imposing gateway of brick, leading from the south side of the old city, and, except above, is in pretty good preservation. To the apex of the arch is 31 ft., the depth is 51 ft., and on the south it was provided with semicircular abutments on each side for the military guard.

About two miles farther south is the Khwāja-kī Masjid, known as the Chhoti Sonā Masjid, built by Walī Muhammad during the reign of Husain Shāh, that is, in the beginning of the sixteenth century. Constructed entirely of hornblende, which has been largely taken from earlier Hindu temples, it is in fairly good preservation—better than any other in Gaur. Inside it is divided lengthwise into three aisles, and five across, the arches over which rise on somewhat massive stone pillars to a height of 20 ft., and above this are the fifteen domes. The five mihrābs are of black hornblende, and were once gilt ; while in the north-west corner is a carved takht or throne. It is ornamented outside by carving in low relief, of most elaborate and artistic designs, and inside it is beautifully finished. In this mosque and the next we have the style probably at its best.

The Sonā Masjid, outside the fort to the north-east, is perhaps the finest memorial left at Gaur. Built by Nusrat Shāh in 1526, it was 170 ft. in length by 76 ft. deep, with walls 8 ft. thick, faced inside and out with hornblende. In front it has eleven arched entrances, 8½ ft. wide and 14 ft. high. These lead into a corridor, the arches of which support the twelve domes of its roof. Within this is the masjid, of which the whole roof has now fallen : it had three aisles in length, supported by twenty pillars, and had eleven mihrābs in the west wall. At both sides of the end doorways to the corridor and at the back corners were polygonal minārs of brown basalt, but how far they rose above the walls is uncertain. The front had carved panels between the doorways and mouldings above them. From its massive solidity and size this must have been an imposing building. Indeed, this characteristic of the Gaur buildings stands out in striking contrast to the somewhat slight architectural arcades of much of the Saracenic architecture.

Perhaps nowhere else, even in India, is the effect of unchecked luxuriant vegetation upon the most substantial structures to be seen in a more striking and withal melancholy scale than in the vast mass of ruins that run almost continuously for more than twenty miles from Pandua to Madhāpur. But besides these remains, there are other examples of this style scattered over the area of what was the Bengal kingdom of the fourteenth and fifteenth centuries. And the style was continued by the natives, whether Hindu or Muhammadan, and employed with excellent effect in temples of later date, as in that of Kānta at Kāntanagar near Dinājpur, erected 1704–22, or in the now-ruined one of Rās Mohan at Gopālganj near the same place, built in 1754; the latter has twelve sides, but is cramped in plan, which is sacrificed to excessive ornament.

Gulbarga and Bīdar

The Bahmani dynasty, founded by Hasan Gangū Bahmani in 1347, had Gulbarga for its capital till about 1428, when it was transferred to Bīdar, a little over sixty miles to the north-east. This kingdom stretched from Berār to the Kistna river and from the Telingāna or Warangal territories on the east to the Arabian Sea on the west, and in the latter half of the fifteenth century it included all the Western Deccan from Mysore to Gujarāt.

During the eighty years (1347–1428) that Gulbarga was the capital, it was adorned with important buildings, of which the most notable now remaining is the great mosque, one of the most striking in India. It measures over all 216 ft. from east to west by 176 ft. from north to south, 45 ft. on the west being occupied by the masjid proper. It differs from all the great mosques in India in having the whole central area covered over—as in the great mosque at Cordova—what in others would be an open court of about 126 ft. by 100 ft., being roofed by sixty-three small domes. The light is admitted through the side-walls, which are pierced by great arches on all sides except the west. This plan protects the worshippers from the heat and glare of the Indian sun. The central area of the masjid is covered by a dome about 40 ft. in diameter, raised on a clerestory, and the side sections by six small domes each, whilst at each end of the corridors are domes of the width of 25 ft. The style is plain and substantial, with but little ornament, and it is built wholly of original materials.

In the east of the town are the tombs of the Bahmani kings—massive square-domed structures with handsome stone tracery on their outer walls, and elaborately finished inside ; they are now used as State offices. Farther out is the shrine of Banda Nawāz, built about 1640, and other dargāhs are close by.

On the removal of the capital to Bīdar, mosques, palaces, and dargāhs were erected there also, of which most have perished. In the citadel the most entire, perhaps, is the mosque, which is 295 ft. in length by 77 ft. deep, with nineteen arched entrances in front, and inside eighty round piers, each 4¼ ft. in diameter, which support the groins of the roof. In the middle, enclosing the mihrābs and a pulpit of three steps, is an apartment 38 ft. square, which is carried up as an octagon a storey above the roof of the mosque, and covered by a large dome. Parts of the roof—which was covered by some eighty-four small domes—have fallen in.

During the reign of Muhammad Shāh II, Khwāja Mahmūd Gilanī (or Gāwān), an old noble, in 1478–9 built a madrasa about 180 ft. by 205 ft., with lofty minārs at the corners of the east face. This must have been a striking building, three storeys in height, with the towers, if not the whole façade, covered with enamelled tiles. In 1656 the city was taken by Aurangzeb, and the madrasa was appropriated as a cavalry barrack, and part of it as a powder magazine, which exploded and wrecked the building.

In 1492 the rule was seized by Kāsim Barīd, who really founded a new dynasty, his son, Amīr Barīd Shāh, assuming the title of king. The dargāh of the latter, about half a mile to the west of the city, stands on a large solid platform, and is nearly 57 ft. square, with walls 9 ft. 8 inches thick, rising to a height of 57 ft. from the platform, crowned with a sort of honeysuckle border, and the dome is about 37 ft. in height. The dome is ornamented inside with belts of coloured tiles, and further decorated with interlaced Arabic sentences.

The ten tombs of Bahmani kings, about five miles north-east from the city, are of the like pattern and of considerable splendour, the largest being that of Ahmad Shāh I, who died in 1435. They are not much ornamented, but are structurally good, and impressive by their massive proportions.

Gujarāt

Of the style of Gujarāt Mr. Fergusson has truly remarked, that ‘of the various forms which the Saracenic architecture assumed, that of Ahmadābād may probably be considered as the most elegant, as it certainly is the most characteristic of all. No other form is so essentially Indian, and no one tells its tale with the same unmistakable distinctness.’ It is the less necessary to enter into detail regarding this style of architecture, since it is one of the very few that have as yet been treated with anything approaching to fullness, in three volumes of the Archaeological Survey of Western India.

The Hindu kingdom of Gujarāt had been in a high state of civilization before its subjugation by the Muhammadans, and the remains of their temples at Sidhpur, Pātan, Modherā, and elsewhere testify to the building capacity of the race. Under Muhammadan rule they introduced forms and ornaments into the works they constructed for their rulers, superior in elegance to any the latter knew or could have invented. Hence there arose a style combining all the beauty and finish of the previous native art with a certain magnificence of conception which is deficient in their own works. The elevations of the mosques have usually been studiously arranged with a view to express at once the structural arrangements, and to avoid monotony of outline by the varied elevation of each division. Instead of the propylon of the Sharkī style, the central portion of the façade was raised by a storey over the roof of the wings, and to this was attached two richly carved minarets, rising in the very earliest mosques only as small turrets above the façade, but soon after to towers of considerable height. The central dome was raised over a gallery above the central part of the hall by two rows of dwarf pillars, of which the outer row was connected by open stone trellis-work, admitting a subdued light and providing perfect ventilation. This second storey rose to about the height of the central façade, and upon it was the principal dome.

By and by the style changed much : the arched entrances in front were often omitted, and only a screen of columns formed the façade, the minarets being removed to the corners, and were no longer for the mu’azzin but only architectural ornaments. This was partly a return to the Indian trabeate construction, and it was carried out in its best form in the Sarkhej group of buildings belonging to the second half of the fifteenth century.

The Muhammadan architecture of Gujarāt is notable for its carved stonework ; and in the perforated stone windows in Sīdī Saiyid’s mosque, the carved niches in the minārs of many other mosques, and the sculptured mihrābs and domed and panelled roofs, we have ornamental work that will stand comparison with, and much of it will rival, anything of the kind employed elsewhere in any age.

Their tombs were a natural product of the style. There were many brick mausoleums, like that of Daryā Khān and of Āzam and Muazzam Khān, just as there were brick mosques like ‘Alif Khān Bhūkāī’s at Dholkā ; but all the stone tombs were pillared pavilions of varying dimensions, the central area over the grave covered by a dome standing on twelve pillars. These pillars were connected by screens of stone trellis-work carved in ever varying patterns, and round this there might be a veranda with twenty pillars in the periphery, or a double aisle with thirty-two in the outer square. And as these were irregularly spaced in order to allow the inner twelve to support the lintels of a regular octagon for the dome, the monotony of equal spacing was avoided. In larger tombs, as in Saiyid Usmān’s, the dome was supported on a dodecagon, and coupled pillars introduced for structural purposes also increased the variety of aspect. The finest example that has come down to us of this class is the tomb of Mubārak Saiyid, erected at Mahmudābād in 1484, which is wholly in the earlier arched style and one of the most splendid sepulchres in India, simple in plan, with a solidity and balance of parts about the whole that has rarely been equalled.

The step-well or wāv of Bāī Harīr, though a Muhammadan work, is strictly Hindu in design, and almost a copy of another at Adālaj ; but it is ornamented with pillars and galleries having carved wall panels in every way as the mosques are. The sluices, too, of the great artificial tanks are really works of art—designed in suitable forms, and highly ornamental.

Bijāpur

The Musalmān dynasties of the Deccan were short-lived, and about the same time that the Barīd Shāhis supplanted the Bahmanis at Bīdar, the Nizām Shāhis set up as rulers at Ahmadnagar ; and in 1492 Yūsuf ‘Ādil Shāh, a Turk—said to have been a son of the Ottoman Sultān Murād II—who had found service under Amīr Barīd, founded the kingdom of Bijāpur. The Berār and Golconda kingdoms arose respectively before and after this; but of their architecture we have as yet no really satisfactory survey. Of Bijāpur we have the excellent accounts by Fergusson, Capt. Hart, and Meadows Taylor (1859 and 1866). It is, therefore, the less necessary to enlarge upon it.

The foreign origin of the ‘Ādil Shāhī dynasty, and their partiality for the Shiah form of Islām prevailing in Persia rather than the Sunni, together with their ready employment of Persian officers, probably influenced their architecture, and led to that largeness and grandeur which characterized the Bijāpur style.

About twenty years ago the Bombay Government adapted a number of these old buildings to modern requirements: the Bukhāra Masjid has been used as a post office, and the mosque belonging to Muhammad’s great tomb was turned into a travellers’ resthouse, but has recently been restored; then the ‘Adālat Mahal was converted into the Collector’s residence; the Chīni Mahal into public offices; Yāqūt Dabalī’s mosque into a residence for the Assistant Collector; Khawāss Khān’s dargāh and mosque into house and office for the Executive Engineer; the Chhotā Chīnī Mahal into a house for the Police Superintendent; and the ‘Arash Mahal into the Civil Surgeon’s residence.

The more notable buildings now left at Bijāpur are the Jāmi Masjid, begun by ‘Alī ‘Ādil Shāh (1557–79), and his unfinished tomb; the Gagan Mahal (1561); the Mihtar Mahal; the Ibrāhīm rauza and mosque (1580–1627); the ‘Asar Mahal; and the Gol Gumbaz or great tomb of Muhammad ‘Ādil Shāh (1626–56).

The style of the buildings differs markedly from those of Agra and Delhi, but is scarcely, if at all, inferior in originality of design and boldness of execution. There is no trace of Hindu forms or details; the style was their own, and was worked out with striking boldness and marked success. The mode in which the thrusts are provided for in the giant dome of Muhammad’s tomb, by the use of massive pendentives, hanging the weight inside, has drawn the admiration of European architects. And this dome, rising to about 175 ft. from the floor, roofs an area 130 ft. square, covering 2,500 sq. ft., larger than the Pantheon at Rome, where stability is secured only by throwing a great mass of masonry on the haunches, and so hiding the external outline.

The plan of the Jāmi Masjid is of the usual form, except that the east wall and corridor was never built ; but, notwithstanding, it is one of the finest mosques in India. It was commenced early in the reign of ‘Ālī ‘Ādil Shāh I (1557-79). The masjid proper is about 240 ft. in length by 130 ft. deep, divided longitudinally into five aisles, by nine across ; but the centre, occupying a square space of three bays each way, is covered by the great dome, supported in the same way as that over Muhammad ‘Ādil Shāh’s, rising to a height of about 96 ft. inside, and is the earliest example of this style of dome—being nearly a century earlier than that on the great tomb. The court is about 187 ft. from east to west, and has a corridor on the north and south sides. At the east corners two minārs were to be erected, but only that on the north was properly begun. At a later date the court was extended eastwards, and a large gateway constructed about 115 ft. in advance of the original court, with part of a corridor on the south of it.

In the Gagan Mahal, again, the central arch has a span of 61 ft., but the whole structure is ruined, and the wooden roof, &c., were carried off by the Marāthās. The Asar Mubārak, too, is largely of wood, the façade being open, with two wooden pillars supporting the roof, while inside the decoration was of the same material and richly painted. Again, in the Mihtar Mahal—really a splendid gateway to a mosque— and in the Ibrāhīm rauza group, we have every detail of the structure in stone covered with the most delicate and exquisitely elaborate carving, the windows filled with tracery, and cornices supported by wonderfully rich brackets. In the dargāh, too—as if in defiance of constructional demands—the room, 40 ft. square, is covered by a perfectly level stone roof, supported only by a cove-bracketing from the walls on each side.

Mughal Saracenic Style

The Mughal phase of Indian Saracenic architecture began under Bābar (1526–31), but we have no important work of his left, nor of his son Humāyūn. The first examples of the style belong to the time of Sher Shāh (1539–45), one of the most characteristic of which is the Kila-kohna or Sher Shāh Masjid (1541) at Purāna-Kila, near Delhi, and there are a few other fragments there and at Rohtās. But though the later developments of the style in the rich remains at Delhi, Agra, Fatehpur Sīkri, and some other places have been largely surveyed and illustrated, these earlier structures, though so interesting as the initial forms of the style, have hitherto been neglected for the more ornamental examples of later date. The first, too, seems to have suffered most under our own rule. During the whole period of the Mughal dynasty, as Mr. Fergusson has well remarked, there is a ‘unity in the works and a completeness in their history which makes the study of their art peculiarly fascinating ; and some of their buildings will bear comparison, in some respects, with any architectural productions in the world.’

With the emperor Akbar (1556–1605) the Mughal styles made a great advance ; he built very largely, and art was living and developing so vigorously during his long reign that it would be difficult to enumerate all the peculiarities of his numerous buildings. As in the Gujarāt and other styles, there is a combination of Hindu and Muhammadan features in his works which were never perfectly blended. Like their predecessors, the Pathāns, the Mughals were a tomb-building race, and those of the latter are even more splendid than those of the former, more artistic in design, and more elaborately decorated. The most splendid of these, and the most renowned building in India, is the far-famed mausoleum, the Tāj Mahal at Agra—the tomb of Mumtāz Mahal, the wife of Shāh Jahān ; it is surrounded by a garden, as were almost all Muslim tombs.

In the fine tomb of his father Humāyūn, and at Fatehpur Sīkri, Akbar’s buildings are best seen ; and as the latter have been well illustrated by Mr. Ed. W. Smith, in four volumes of his Survey Reports, it is unnecessary here to enter upon details. Three small pavilions, said to have been built for three of his wives, are gems of picturesque structures, carved and ornamented to the greatest extent without being in the least overdone, and are unsurpassed by anything of the kind elsewhere. Then the great mosque is scarcely matched in elegance and architectural effect ; the south gateway is well-known, and from its size and structure excels any similar entrance in India. Akbar’s pavilion, the Chālīs Sitūn at Allahābād, was destroyed for materials to repair the fortifications ; but his tomb at Sikandra near Agra is a unique structure of the kind and of great merit, the plan probably suggested by some native design.

With Akbar’s death the style underwent a change : the Hindu features disappeared entirely, as if outgrown. Jahāngīr made Lahore his principal residence, and Agra and Delhi have little to show belonging to his rule. His great mosque at Lahore is in the Persian style, covered with enamelled tiles ; his tomb near by (1630–40) was made a quarry of by the Sikhs from which to build their temple at Amritsar ; and the capital he built at Dacca in Bengal, being mostly of brickwork, in so moist a climate, has gone to utter decay. At Agra, the tomb of I‘timād-ud-daula belongs to this reign, and being built entirely of white marble and covered wholly by pietra dura mosaic, it is one of the most splendid examples of that class of ornamentation anywhere to be found.

Under Shāh Jahān (1628–58) a remarkable change came over the style : its force and originality gave way to a delicate elegance and refinement of detail. This is well illustrated in the magnificent palaces he built at Agra and Delhi—the latter once the most exquisitely beautiful in the East. Unfortunately, no adequate survey of what remains of these buildings has yet been published. Of the Tāj Mahal (1632–54), fortunately so well preserved, nothing need be added ; its extreme delicacy, the richness of its material, and the complexity of its magnificent design have been dwelt on by writers of all countries. So also of the surpassingly pure and elegant Motī Masjid in the Agra Fort, all of white marble : it is among the gems of the style. The Jāmi Masjid at Delhi (1650–66) is a really imposing building, and its position and architecture have been carefully considered so as to produce a pleasing effect and feeling of spacious elegance and well-balanced proportion of parts. In his works Shāh Jahān presents himself as the most magnificent builder of Indian sovereigns.

With the reign of Aurangzeb (1658–1707) the decline of taste set in at once. He was more disposed to insult the religion of the Hindūs than to glorify his reign by splendid monuments ; and with all his fanaticism on behalf of Islām, it is said he lowered the mimbars of the mosques that the khatībs might not stand in a commanding position in his presence. With little true reverence, it was hardly to be expected he should delight in architectural magnificence. Spending much of his time in camps, he built no palace of importance ; the tomb of his favourite wife at Aurangābād—vulgarly believed to be like the Tāj at Agra—is commonplace to a degree, and he erected no tomb for himself, though he lived to a great age. The works of his reign seem mostly to have shared in the same decline of style : squared stone and marble gave way to brick or rubble with stucco ornament.

The buildings at Seringapatam and Lucknow are of still later date and are in certain respects imposing, but in detail are often tawdry. Yet architecture is not dead in India. Even in recent years there have been erected tombs and temples of purely native origin and of much elegance in detail, while retaining the essential elements of structural design ; in others again, these elements have parted company, and no amount of elegant ornament can compensate the want of propriety in such structures. Otherwise the imitation of a foreign style is rapidly proving fatal to indigenous art.

Much remains to be done to make us fully acquainted with Indian architecture in its many and interesting phases, more especially in Hindustān or India north of the Vindhya range, and in the extreme South as well as in Hyderābād territory. In the North there has been too little system in the surveys ; we want a few monographs on entire styles and districts to enable us to grasp their real merits and characteristics. Surveys of buildings here and there without any links of architectural or historical connexion may serve to illustrate the traveller's route, but contribute little to a full or scientific delineation, the publication of which is a serious desideratum.

JAS. BURGESS.

BIBLIOGRAPHY

Capt. Ernest Barnes.—‘ Dhar and Mandu,’ in Jour. Bombay B. R. As. Soc., vol. xxi, pp. 339–391.

Capt. Blake.—History of Mandu, the ancient Capital of Malwa. Bombay, 1844 ; reprint. 1879.

J. W. Breeks.—An Account of the Primitive Tribes and Monuments of the Nilagiris. 4to. London, 1873.

  • J. Burgess.—The Rock-cut Temples of Ajanta. Pamphlet, 1868.—The Temples of Satruñjaya, photographed by Sykes and Dwyer, with historical and descriptive introduction (one plate and forty-six photographs). Atlas fol. Bombay, 1869. Reprint of Text. 8vo. Ahmadabad, 1878.—Notes of a Visit to Somanâth, Girnâr, &c., in Kâṭhiâwâḍ. 18mo. Bombay, 1869. —Notes of a Visit to Gujarât. 12mo. Bombay, 1870.—Photographs from Somanâth, Girnâr, Junâgadh, &c., in Kâṭhiâwâḍ, with descriptive Text. Bombay, 1870.—The Rock-cut Temples of Elephanta or Ghârâpurî. Illustrated (with drawings and photographs) 8vo and oblong fol. Bombay, 1871. —Photographs of Architecture and Scenery in Gujarât and Rājputāna (by Bourne and Shepherd), with historical and descriptive letterpress. Fol. Calcutta, 1874.—Report of the Archaeological Survey in Belgaum and Kaladgi Districts. Roy. 4to. London, 1874.—Report of the Archaeological Survey : Kâṭhiâwâḍ and Kachh. Roy. 4to. London, 1876.—The Rock-cut Temples of Elura or Verul, with twelve photographs. Cr. 8vo. Bombay, 1877.—Report of the Archaeological Survey : Bidar and Aurangâbâd. Roy. 4to. London, 1878.—Notes on the Rock-temples of Ajanta and their Wall-paintings. Demy 4to. Bombay, 1879.—Inscriptions, &c., from the Cave-temples of Western India. Demy 4to. Bombay, 1881. —Notes on the Amarâvati Stûpa. 4to. Madras, 1882.—Buddhist Cave-temples and Brahmanical and Jaina Caves in Western India. 2 vols. Roy. 4to. London, 1883.—List of Antiquarian Remains in the Bombay Presidency. Demy 4to. Bombay, 1885 : 2nd ed., revised by H. Cousens. 1897.—Buddhist Stûpas of Amarâvati and Jaggayyapeta, Roy. 4to. London, 1887.—‘Archæological Research in India,’ Actes du Congrès Int. des Orientalistes, 1889.—Epigraphia Indica, vols. i and ii. 4to. Calcutta, 1891–4.—Muhammadan Architecture of Bharoch, Cambay, Dholka, &c., in Gujarât. Roy. 4to. London, 1896.—The Ancient Monuments, Temples, Sculptures, &c., of India. Pt. i, 170 plates. Fol. London, 1897.—Muhammadan Architecture of Ahmadabad. 2 vols. Roy. 4to. 1900–5.—Buddhist Art in India, translated from A. Grünwedel’s Handbuch, revised and enlarged : 154 illustrations. 8vo. London, 1901.—‘Indian Architectural Details.’ Jour. of Indian Art, vol. iii, No. 32, fifteen plates. Fol. London, 1890.—The Gandhâra Sculptures, twenty-five plates and thirty-eight text-blocks. Ibid., vol. viii, pp. 23–40 and 73–92.—‘The Great Stūpa at Sânchi-Kânâkheda.’ Jour. R. As. Soc. 1902, pp. 29–45.—‘Sketch of Archæological Research in India during half a century.’ Jour. Bom. B. R. As. Soc. Centenary vol., 1905, pp. 131–148.

J. Burgess and H. Cousens.—The Antiquities of Dabhoi. Fol. Edinburgh, 1888.—The Architectural Antiquities of Northern Gujarât. Roy. 4to. London, 1903.

J. Burgess, E. W. Smith, and A. Führer.—Sharqi Architecture of Jaunpur. 4to. 1889.

Capt. T. S. Burt.—‘Description, with drawings, of the ancient Stone Pillar at Allahabad.’ Jour. As. Soc. Beng., vol. iii, pp. 105–123.

A. C. Campbell.—Glimpses of the Nizam’s Dominions. Fol. Bombay, 1898.

Sir J. Campbell.—‘Mandû,’ in Jour. Bom. B. R. As. Soc., vol. xix, pp. 154–201.

A. M. Cantrell.—Account of the Ruins at Bijâpur. Bombay, 1872.

Major H. H. Cole.—Illustrations of Ancient Buildings in Kashmir. Roy. 4to. London, 1869.—Architecture of Ancient Delhi. 28 plates. 4to. London, 1872.—Illustrations of Buildings near Muttra and Agra. Roy. 4to. London, 1873.—Preservation of Ancient Monuments, India. Ten parts, 104 plates. Fol. Calcutta, 1884–5.—Eighty-six plates from the preceding, and fourteen others. Fol. London, 1896.—Three Reports on Preservation of Ancient Monuments in India. 8vo. Simla, 1882–4.

H. Cousens.—Bijâpur, the Old Capital of the ‘Adil Shahî Kings. 8vo. Poona, 1889.—Lists of Antiquarian Remains in the Central Provinces, Berar, and Haidarabad. 4to. Calcutta, 1897–1900.—‘Chālukyan Temples.’ Jour. Ind. Art., vol. ii (1888), No. 17, and vol. iii (1890), No. 28.—‘Conservation of Ancient Monuments in the Bombay Presidency.’ Jour. Bom. B. R. As. Soc., Centenary vol., pp. 149–162. Sind Tiles, coloured plates. Fol. London, 1906.

W. G. Cowie.—‘Notes on Some of the Temples in Kashmir.’ Jour. As. Soc. Beng., vol. xxxv, pt. i, pp. 91 ff.

J. Crawfurd.—Journal of an Embassy to the Court of Ava. 4to. London, 1829.

H. Creighton.—The Ruins of Gour, Described and Represented in eighteen Views. 4to. London, 1817.

Sir A. Cunningham.—The Bhilsa Topes, or Buddhist Monuments of Central India, &c. 8vo. London, 1854.—The Stûpa of Bharhut, a Buddhist Monument illustrative of Buddhist Legend and History. 4to. Simla, 1877.—Mahabodhi ; or, the Great Buddhist Temple at Buddhagayā. 4to. London, 1892.—Archæological Survey of India Reports. 1862–84. 23 vols. 8vo. Calcutta, 1871–1887.—‘Essay on the Arian Order of Architecture, as exhibited in the Temples of Kashmir.’ Jour. As. Soc. Beng., vol. xvii, pt. ii, Sept., 1848.

Capt. J. D. Cunningham.—‘Notes of the Antiquities of the Districts within the Bhopal Agency.’ Jour. As. Soc. Beng., vol. xvi, pt. ii, pp. 739–63.

T. and W. Daniell.—Antiquities and Views in India. 143 plates. Folio. London, 1799.

H. C. Fanshawe.—Delhi Past and Present. 8vo. 1902.

J. Fergusson.—Illustrations of the Rock-cut Temples of India. Folio. London, 1845.—Picturesque Illustrations of Ancient Architecture in Hindustan. Folio. London, 1847.—The Rock-cut Temples of India, &c., illustrated by photographs by Major R. Gill. 8vo. London, 1864.—On the Study of Indian Architecture. 8vo. London, 1867.—Tree and Serpent-Worship : Illustrations of Mythological Art in India. 2nd edition. Roy. 4to. London, 1873.—History of Indian and Eastern Architecture. 8vo. London, 1876.—Archæology in India. 8vo. London, 1884.

J. Fergusson and J. Burgess.—The Cave-Temples of India. 8vo. London, 1880.

J. Fergusson and P. Meadows Taylor.—Architecture at Beejapoor. Folio. London, 1866.—Architecture in Dharwar and Mysore. Folio. London, 1866.

A. Kinloch Forbes.—Râs Mâlâ : Hindu Annals of the Province of Goozerat. 1st ed. 2 vols. 8vo. London, 1856.

J. Forbes.—Oriental Memoirs. 4 vols. 4to. London, 1813.

L. de Forest.—Indian Architecture and Ornament (50 photographs). 4to. Boston, 1887.—Indian Domestic Architecture. n.d.

Lieut.-Col. Forrest.—Picturesque Tour along the Rivers Ganges and Jumna. Twenty-four views, &c. London, 1824.

G. W. Forrest.—Cities of India. 8vo. Westminster, 1903.

A. Foucher.—Les Bas-reliefs du Stûpa de Sikri. 8vo. Paris, 1903.— L’Art Grêco-Bouddhique du Gandhâra. 1905.

A. Führer.—Monumental Antiquities and Inscriptions in the North-West Provinces and Oudh. 4to. Allahabad, 1891.—Monograph on Buddha Sakyamuni’s Birthplace. 1897.

Sir Lepel Griffin.—Famous Monuments of Central India. Obl. fol. London, 1886.

E. M. Grindlay.—Scenery, Costumes, and Architecture, chiefly in Western India. Thirty-six plates. Fol. London, 1826.

F. S. Growse.—Mathurâ, a District Manual. 2nd ed. 4to. Allahabad, 1880.

B. R. Harrington.—Portfolio Studies from Ancient Hindu Architecture. Fol. London, 1888.

Capt. C. Harris.—Ruins of Mandoo. Fol. London, 1860.

Capt. C. Hart and J. Fergusson.—Architectural Illustrations of the principal Buildings of Bijapur. Fol. London, 1859.

T. H. Hendley.—‘Decorative Art in Rajputana.’ Jour. Ind. Art, vol. ii (1888), No. 21.

William Hodges.—Select Views in India. Forty-eight plates. Imp. fol. London, 1786.

Sir T. C. Hope.—Surat, Bharoch, and other Cities of Gujarât. 4to. Bombay, 1868.

Sir T. C. Hope and J. Fergusson.—Architecture of Ahmedabad, photographed by Col. Biggs, R.A. 4to. London, 1866.

Sir S. Swinton Jacob.—Jaypore Portfolios of Architectural Details (10 parts). Atl. fol. London, 1890, &c.

J. L. Kipling.—‘Indian Architecture of To-day.’ Jour. Ind. Art, vol. i (1886), No. 3.—‘Mosque of Wazir Khân at Lahor.’ Ibid., vol. ii (1888), No. 19.

E. Impey.—Views in Delhi, Agra, and Rajputana. London, 1865.

Sir S. Swinton Jacob.—Jeypore Portfolios of Architectural Details. Ten parts. Folio. 1898.

Dubois de Jancigny and Xavier Raymond.—Inde (in L’Univers); with eighty-four plates. 8vo. Paris, 1845.

H. G. Keene.—Handbook for Visitors to Delhi. Calcutta, 1874.— Handbook for Visitors to Agra. Calcutta, 1874.

Major Markham Kittoe.—Illustrations of Indian Architecture from the Muhammadan Conquest. Fol. Calcutta, 1838.—‘Ruins and Pillar at Jajipur,’ Jour. As. Soc. Beng., vol. vii, pp. 53–60 and 200–5.— ‘Notes on the Caves at Barâbar.’ Jour. As. Soc. Beng., vol. xvi, pt. i, pp. 401–16.

L. M. Langles.—Monuments Anciens et Modernes de l’Hindustan. 144 plates. 2 vols. Fol. Paris, 1821.

Gust. Le Bon.—Les Monuments de l’Inde. 4to. Paris, 1893.

List of Ancient Monuments of Bengal. Fol. Calcutta, 1896.

Sylvain Levi.—Le Nepal. 2 vols. 8vo. Paris, 1905–6.

Capt. Lyon.—Notes to accompany photographs to illustrate The Ancient Architecture of Southern India. Edited by J. Fergusson. 8vo. London, 1870.—Notes to accompany photographs to illustrate The Architecture of Gujarat. 8vo. London, 1870.

Capt. T. S. F. Mackenzie.—Description of the Halabid Temple. 12mo. Bangalore, 1873.

Genl. F. C. Maisey.—Sānchi and its Remains. 4to. 1892.

Sir C. Malet.—‘Description of the Caves or Excavations of Ellore.’ Asiat. Res., vol. ii, pp. 389–420.

J. H. Marshall.—Archaeological Survey of India: Annual Report for 1902–3. 4to. Calcutta, 1904.—Annual Report for 1903–4. Calcutta, 1906.

R. Montgomery Martin.—History, Antiquities, and Statistics of Eastern India. 3 vols. 8vo. London, 1837.

Sayyid Muhammad Latif.—Lahore, its History, Architectural Remains, and Antiquities. 8vo. Lahore, 1892.

John Muir, LL.D.—Account of a Journey from Agra to Bombay. 8vo. Edinburgh, 1854.

H. A. Oldfield.—Sketches from Nipal, Historical and Descriptive. 2 vols. 8vo. London, 1880.

Oriental Annual. 7 vols. 8vo. London, 1834–40.

Prof. F. Orlebar.—The Monolithic Temples of India, with some Account of the Mural Paintings at Ajunteh. 12mo. London.

Major-General Sir A. P. Phayre.—History of Burma. London, 1883.

Jas. Prinsep.—Benares, Illustrated in a Series of Drawings. Sm. folio. Calcutta, 1830–31.

Rajendrala Mitra.—Antiquities of Orissa. 2 vols. Fol. Calcutta, 1875–80.—Buddha Gaya, the Hermitage of Śakya Muni. 4to. Calcutta, 1878.

Ram Raz.—Architecture of the Hindus. 4to. London, 1834.

  • J. H. Ravenshaw.—Gaur, its Ruins and Inscriptions. 4to. London, 1878.
  • A. Rea.—South Indian Buddhist Antiquities. 4to. Madras, 1894.—Chalukyan Architecture of Southern India. 4to. Madras, 1896.—‘Dravidian Architectural Details,’ in Jour. Ind. Art and Industry, vol. v, No. 44, six plates. Fol. 1893.
  • Lewis Rice.—Mysore, a Gazetteer compiled for Government. Revised ed. 2 vols. 8vo. 1897.
  • E. Roberts.—Views in India, China, and the Shores of the Red Sea, with Descriptions. 2 vols. 4to. London, 1835.
  • L. Rousselet.—L'Inde des Rajahs. 317 gravures, &c. 4to. Paris, 1875.
  • R. Sewell.—Lists of Antiquarian Remains in the Madras Presidency. Madras, 1882.
  • Rev. M. A. Sherring.—The Sacred City of the Hindus. 8vo. London, 1868.
  • Wm. Simpson.—‘Indian Architecture.’ Trans. R. Inst. Brit. Architects, 1861–62, pp. 165–78.—‘Classical Influence in the Architecture of the Indus Region and Afghanistan.’ Ibid., 1894, pp. 93–115 and 191–3.—‘Buddhist Architecture in the Jellalabad Valley.’ Ibid., 1879–80, pp. 37–64.
  • E. W. Smith.—Mughal Architecture of Fathpur Sikri. 4 parts. Ar. Sur. 4to. Allahabad, 1894–98.—Portfolio of Indian Architectural Drawings. Fol. London, 1897.—Mughal Colour Decoration of Agra. Pt. i. 4to. 1901.—‘Chaurâsi Gumbâz, Kalpi.’ Eight plates. Jour. Ind. Art and Industry, vol. v (1893), No. 42.—‘Wall-paintings from Fathpur Sikri.’ Eleven plates. Ibid., vol. vi (1895), pp. 65–8.
  • V. A. Smith.—The Jain Stupa and other Antiquities of Mathura. 1901.
  • R. Phenè Spiers.—Architecture East and West. 1905.
  • Carr Stephen.—The Archaeology of Delhi. 8vo. Ludhiana, 1876.
  • Major W. H. Sykes.—‘An Account of the Caves of Ellora.’ Trans. Lit. Soc. Bombay, vol. iii, p. 265, 1st ed.
  • Major M. Symes.—Account of an Embassy to the Court of Ava in 1795. 4to. London, 1800.
  • Lieut.-Col. Jas. Tod.—Annals and Antiquities of Rajast’han. 2 vols. 4to. London, 1829–32.—Travels in Western India. 4to. 1839.
  • Views of Beejapore and the Deccan. 1845.
  • G. T. Vigné.—Travels in Kashmir, Ladak, Iskard, &c. 2 vols. London, 1842.
  • H. H. Wilson.—Ariana Antiqua; A descriptive Account of the Antiquities and Coins of Afghanistan. 4to. London, 1841.—The Oriental Portfolio. Eleven litho. plates. Fol. London, 1841.
  • John Wilson, D.D.—Lecture on the Religious Excavations of Western India. Bombay, 1875.—‘Memoirs on Cave-Temples and Monasteries, &c.’ Jour. Bom. B. R. As. Soc., vol. iii, pt. ii, p. 36 ff., and vol. iv, p. 340 ff.
  • Dr. W. H. Workman and Mrs. F. B. Workman.—Through Town and Jungle, among the Temples and People of the Indian Plains. Illustrated. 8vo. London, 1904.
  • Col. H. Yule.—Mission to the Court of Ava in 1855. 4to. London, 1858.