History of Indian and Eastern Architecture

CHAPTER II.

Chapter 5233,221 wordsPublic domain

PAGODAS.

CONTENTS.

Temple of the Great Dragon--Buddhist Temples--Taas--Tombs--Pailoos--Domestic Architecture.

If we had the requisite knowledge, or if the known examples of Chinese temples were sufficiently numerous, we ought, before describing them, to classify the buildings, apportioning each to that one of the three religions to which it belongs. For the present this must be left to some one on the spot. Meanwhile there is no difficulty in recognising those which belong to the religion of Fo or Buddha. These are generally the nine-storeyed towers or taas, which, as will be explained hereafter, are merely exaggerated tees of the Indian dagobas. The temples, properly so-called, of this religion, are not very magnificent, nor are they generally built in a permanent style of architecture. This is still more the case, apparently, with the temples of Confucius. The only one that has been carefully described and photographed is that at Pekin, which is also probably the most magnificent. Judging from our present information, it more resembles a university than a temple. There are neither images nor altars, but great halls, on which are hung up the names of the emperors and of the most distinguished literates of the kingdom. There are no priests; and though ceremonies are there performed annually by the emperor in honour of the great philosopher, these scarcely can be called worship, or the hall a temple.

TEMPLE OF THE GREAT DRAGON.

The most magnificent temple in the capital, so far as we know in the empire, is that known as the Temple of Heaven, or the Great Dragon.[656] It is situated close to the southern wall of the city in a square

enclosure measuring about a mile each way. From the outer gate a raised causeway leads to the temple, on either side of which are numerous buildings for the accommodation of the priests, which are approached by frequent flights of steps leading down to a park beautifully planted. At its inner extremity stands the temple itself, a circular building, three storeys in height, with broad projecting roofs, the upper terminating in a gilt ball, directly under which stands the altar.

The temple is raised on a circular pyramid, the three terraces of which are seen in the woodcut. There are several handsome gateways at intervals across the causeway, so arranged that from the entrance the circular temple itself can be seen through the long vista, framed as it were by them; and as the whole of the upper part is covered with blue tiles and gilding, the effect is said to be very pleasing.

In the same enclosure is another temple called that of the Earth, where sacrifices of animals are annually offered to the gods, whoever they may be, to whom this temple is dedicated.

These temples are said to have been erected about the year 1420, and, if so old, seem to be in a very fair state of preservation, considering the manner in which they are now neglected.

In reading Mr. Michie’s, or any other description of the Dragon Temple of Pekin, it seems impossible to avoid feeling that there are so many points of resemblance between it and the Serpent Temple of Nakhon Wat, that the coincidence can hardly be accidental. The variations are hardly greater than might be expected from difference of age, and the fact that the one was erected by Chinese at the northern extremity of their empire, the other by Cambodians near the southern limit of theirs. All the links, however, which connect the two temples are still wanting; yet, as we have the assertion of the Chinese traveller in 1295 that the Tao-tze religion[657] existed in Cambodia while he was there, we should not feel surprise at any similarity that may be traced between the temples of the two countries.

BUDDHIST TEMPLES.

The only Buddhist temple in China of which any plans have been made, or which I have myself had an opportunity of inspecting, is that at Honan, opposite Canton. Unfortunately it is very modern, and by no means monumental. It is a parallelogram enclosed by a high wall, measuring 306 ft. by 174 ft. In the shorter front facing the river is a gateway of some pretension. This leads to a series of halls opening into each other, and occupying the whole of the longer axis of the internal court. The first and second of these are porches or antechapels. The central one is the largest, and practically the choir of the building. It contains the altar, adorned by gilt images of the three precious Buddhas, with stalls for the monks and all arrangements necessary for the daily service. Behind this, in the next compartment, is a dagoba, and in its rear another apartment devoted to the goddess Kuan yin, principally worshipped by women--in fact, the Lady Chapel of the church. Around the court are arranged the cells of the monks, their kitchen, refectory, and all the necessary offices of the convent. These are generally placed against the outer wall, and open into the court.

Any person familiar with the rock-cut examples in India will easily recognise in this temple all the features he is accustomed to in the earlier Chaityas and Viharas, though strangely altered by their Chinese disguise. The figure which stood in front of the dagoba (Woodcut No. 61) is moved forward and placed on an altar by itself, with two companions added, in accordance with modern Chinese theology; but the general arrangements remain the same. The most interesting part, however, is the arrangement of the cells, &c., relatively to the temple. In one of the caves at Dhumnar (Bhim ka Bazar) something like this has been attempted, but it is evidently so difficult of execution in the rock, that we are not surprised to find it not repeated. It is evidently what was intended to be represented on the central rath of Mahavellipore (Woodcut No. 181), and must indeed have been the general arrangement of Buddhist ecclesiastical establishments. What is now wanted is, that some one should supply information regarding the earlier temples of the Chinese, say of the 12th to the 16th centuries. They no doubt exist, and would throw great light on the earlier Indian examples. In the meanwhile, however, it is curious to refer back to the Woodcut No. 129. From it it will be perceived that as early as the 11th century the Buddhist Chaitya in India, standing in the centre of its Vihara, had already been sublimated into an idol temple, surrounded by a series of idol niches, since there cannot be a doubt that the Jaina temple of Vimala Sah is a reproduction for another purpose of an old Buddhist monastery. The curious point is, that the 18th-century temple of Honan reproduces, for their original purpose, forms which in India had, seven centuries earlier, passed away to another faith, and became wholly conventional. It is still more strange that, if we leap over the intermediate period, and go seven centuries further back, we shall find in India the same ceremonies performed in the same form of temples as those at which any one may assist in China at the present day.

At Pekin there are several Lamaseries or Buddhist monasteries, of a much more monumental character than that at Honan, but it is very difficult indeed to guess at their arrangement from mere verbal descriptions without dimensions. The gateway of one, represented in Woodcut No. 380, gives a fair idea of the usual mode of constructing gateways in China.

It has three openings of pleasing proportions, and is as well designed as any to be found in China. Behind it is to be seen the dagoba, to which it leads: a tall form, with a reverse slope, and an exaggerated tee, so altered from those we are accustomed to in the earlier days of Indian architecture, that it requires some familiarity with the intermediate forms in Nepal and Burmah to feel sure that it is the direct lineal descendant of the topes at Sanchi or Manikyala. Around it are minarets, with a cross-legged seated figure of Buddha on each face. But without a plan or description it is impossible to say whether they come down to the ground, or on what kind of basement they rest.

The ordinary form of a temple, as seen in the villages or towns in China, is extremely simple, and seems to be the same, whether dedicated to Buddha, or to the Queen of Heaven, or to any other deity of the strange pantheon of the Celestial Empire. It generally consists of a square apartment with a highly ornamented roof, and with one of the side-walls removed. The entrance is never at the end, nor the end wall ever removed, as would be the case in the West, but always the side; and it is by no means clear that this is not the right and reasonable way of arranging matters. In very small temples a single beam supports the eaves, and a screen inside forms the back of the porch and the front of the temple. In larger temples two or more pillars are introduced, but the other arrangements remain the same. Both these may be seen in the annexed woodcut (No. 381), and when arranged as picturesquely as in this group, and with their gateways and subsidiary adjuncts, they become very pleasing features in the landscape. As architectural objects, they depend for their effect principally on colour, which is applied with an unsparing hand in the form of glazed tiles, painted ornaments, and frequently also paintings, such as landscapes and figure subjects. Gilding is also employed to a great extent, and with good effect.

TAAS.

The objects of Chinese architecture with which the European eye is most familiar are the taas, or nine-storeyed pagodas, as they are usually called. In the south they generally have that number of storeys, but not always, and in the north it ranges from three to thirteen. As before hinted, these are nothing but exaggerated tees of dagobas, and it is easy to trace them through all the stages of the change. In India we can easily trace the single wooden chattah or umbrella of Karli (Woodcut No. 56) to the nine-storeyed tower at Chittore (Woodcut No. 143), and from that the transition is easy to the Chinese examples, although the elaboration of the two was simultaneous, and the Chinese had probably erected tall towers as early as the Jains.

Of those which existed in China in our own time the best known is the celebrated porcelain tower at Nankin.[658] Commenced in the year 1412, and finished in 1431, it was erected as a monument of gratitude to an empress of the Ming family, and was, in consequence, generally called the Temple of Gratitude. It was octagonal in form, 236 ft. in height, of which, however, about 30 ft. must be deducted for the iron spire that surmounted it, leaving little more than 200 ft. for the elevation of the building, or about the height of the Monument of London. From the summit of the spire eight chains depended, to each of which were attached nine bells, and a bell was also attached to each angle of the lower roofs, making 144 bells in all, which, when tinkling in harmony to the evening breeze, must have produced an effect as singular as pleasing. It was not, however, either to its dimensions or its bells that the tower owed its celebrity, but to the coating of porcelain which clothed its brick walls, as well as the upper and under sides of the projecting roofs, which mark the division of each storey. The porcelain produced a brilliancy of effect which is totally lost in all the representations of it yet published, but which was, in fact, that on which the architect almost wholly relied for producing the effect he desired, and without which his design is a mere skeleton.

Another celebrated pagoda is that known as “Second Bar Pagoda,” on the Canton river. It is a pillar of victory, erected to commemorate a naval battle which the Chinese claim to have gained near the spot. It is, in design, nearly identical with that of Nankin, but of smaller dimensions, and is now fast falling to ruin.

These two are of the usual and most typical form, and so like hundreds of others, that it is impossible to deduce any sequence from them with such representations as we now possess. Though pleasing and purposelike, as well as original, they are somewhat monotonous in design. A tower divided into nine equal and similar storeys is a very inferior design to that of the minars of the Mahomedans, or the ordinary spires of Christian churches; and, if all were like these, we should be forced to deny the Chinese the faculty of invention in architecture. In the north, however, the forms seem much more various. One in the Summer Palace (Woodcut No. 383) is divided into either three or seven storeys, as you choose to count them. Four of the sides of the octagon are longer than the other four, and altogether there is a play of light and shade, and a variety about the ornaments in this tower, which is extremely pleasing. It is much more like an Indian design than any other known in China, and with the circle of pillars round its base, and the Lât or Stambha, which usually accompany these objects further west, it recalls the original forms as completely as any other object in this country.

In direct contrast to this is the Pagoda of Tung Chow (Woodcut No. 384). Its thirteen storeys are almost more monotonous than those of the Nankin Pagoda; but they are merely architectural ornaments, string-courses, in fact; and as the tower is not pierced with windows above the base, it becomes, like an Orissan temple, an imposing object of architectural art without any apparent utilitarian object. It thus escapes the charge of littleness in design, which only too justly applies to most of its compeers.

It is extremely difficult to form a correct estimate of the artistic merits of these towers. Edifices so original and so national must be interesting from that circumstance alone, and it seems almost impossible to build anything in a tower-like form of great height, whether as a steeple, a minar, or a pagoda, which shall not form a pleasing object from its salience and aspiring character alone, even without any real artistic merit in itself. Besides these qualifications, I cannot but think that the tapering octagonal form, the boldly-marked divisions, the domical roof, and general consistence in design and ornament of these towers, entitle them to rank tolerably high among the tower-like buildings of the world.

TOMBS.

Like all people of Tartar origin, one of the most remarkable characteristics of the Chinese is their reverence for the dead, or as it is usually called, their ancestral worship. In consequence of this, their tombs are not only objects of care, but have frequently more ornament bestowed upon them than graces the dwellings of the living.

Their tombs are of different kinds; often merely conical mounds of earth, with a circle of stones round their base, like those of the Etruscans or ancient Greeks, as may be seen from the woodcut (No. 385) borrowed from Fortune’s ‘China’--which would serve equally well for a restoration of those of Tarquinia or Vulci. More generally they are of a hemispherical shape, surmounted with a spire, not unlike the Indian and Ceylonese examples, but still with a physiognomy peculiarly Chinese. The most common arrangement is that of a horseshoe-shaped platform, cut out of the side of a hill. It consequently has a high back, in which is the entrance to the tomb, and slopes off to nothing at the entrance to the horseshoe, where the wall generally terminates with two lions or dragons, or some fantastic ornament common to Chinese architecture. When the tomb is situated, as is generally the case, on a hillside, this arrangement is not only appropriate, but elegant. When the same thing is imitated on a plain, it is singularly misplaced and unintelligible. Many of the tombs are built of granite, finely polished, and carved with a profusion of labour that makes us regret that the people who can employ the most durable materials with such facility should have so great a predilection for ephemeral wooden structures.

When the rock is suitable for the purpose, which, however, seems to be rarely the case in China, their tombs are cut in the rock, as in Etruria and elsewhere; and tombs of the class just described seem to be a device for converting an ordinary hillside into a substitute for the more appropriate situation.

Occasionally, however, the Chinese do erect tombs, which, though ornamental, are far from being in such good taste as the two forms just quoted. A tumulus is considered appropriate for this purpose all the world over, and so is the horseshoe form under the circumstances in which the Chinese employ it; but what can be said in favour of such an array of objects as those shown in the Woodcut No. 387? Judged by the standard of taste which prevails in China at the present day, they may be considered by the natives as both elegant and ornamental, but it would be difficult to conceive anything which spoke less of the sepulchre, even from a Chinaman’s point of view; while, on the other hand, their dimensions are such as to deprive them of all dignity as architectural objects.

PAILOOS.

The Pailoos, or “triumphal gateways,” as they are most improperly called, are another class of monument almost as frequently met with in Chinese scenery as the nine-storeyed pagodas, and consequently nearly as familiar to the European eye. Their origin is as distinctly Indian as the other, though, from their nature, being easily overthrown, but few examples can be found in a country that has so long ceased to be Buddhist. Fortunately, however, we still possess in the gateway of Sanchi (Woodcut No. 10) the typical example of the whole class; and we find them afterwards represented in bas-reliefs and in frescos in a manner to leave no doubt of the frequency of their application.

In China they seem almost universally to be employed as honorific monuments of deceased persons--either men of distinction, or widows who have not married again, or virgins who have died unmarried. Frequently they are still constructed in wood, and when stone is used they retain to this hour the forms and details of wooden construction. Whatever the material, they consist of either two or four posts, set either on the ground, so as to allow a passage through, or on a platform, as in Woodcut No. 388. This is as usual a form as the other, and shows how inapplicable the term gateway is to these monuments. The posts always carry a rail or frieze, bearing an inscription, which is, in fact, the object for which the monument was erected. Above this are various architectural details, which complete the design in a manner both original and artistic.

One serving as the portal to a dagoba has already been given (Woodcut No. 380), and though rich, can hardly be considered as superior to that in Woodcut No. 389, which spans a street in Amoy. Instead of leading to a dagoba, as was the case at Sanchi, and generally in India, we have, in this instance, what appears to be a simulated coffin placed under a canopy, and above the principal cornice, which is an essentially Chinese idea. With them a handsome coffin is an object of the highest ambition, and is, consequently, a luxury which the rich take care to provide themselves with during their lifetime. So far as we know, no great structural dagobas ever existed in China, so that their form is generally unfamiliar to the people.

Probably the Chinese would have spent more pains on their tombs had they not hit on the happy device of separating the monument from the sepulchre. We do so in exceptional cases, when we erect statues and pillars or other monuments to our great men on hill-tops or in market-places; but as a rule, a man’s monument is placed where his body is laid, though it would probably be difficult to assign a good logical reason for the practice. The great peculiarity of China is that in nine cases out of ten they effect these objects by processes which are exactly the reverse of those of Europe, and in most cases it is not easy to decide which is best. In erecting the Pailoo, or monument, in a conspicuous place apart from the sepulchre, they seem to have shown their usual common sense, though an architect must regret that the designs of their tombs suffered in consequence, and have none of that magnificence which we should expect among a people at all times so addicted to ancestral worship as the Chinese.

In an historical point of view, the most curious thing connected with these Pailoos seems to be, that at Sanchi, about the Christian Era, we find them used as gateways to a simulated tomb. In India both the tumulus and the Pailoo had at that time passed away from their original sepulchral meaning; the one had become a relic-shrine, the other an iconostasis. Two thousand years afterwards in China we find them both still used for the purposes for which they were originally designed.

DOMESTIC ARCHITECTURE.

It is in their domestic architecture, if in any, that the Chinese excel; there we do not look either for monumental grandeur or for durability, and it is almost impossible to resist being captivated by the gaiety and brilliancy of a Chinese dwelling of the first class, and the exuberant richness and beauty of the carvings and ornaments that are heaped on every part of it.

One of the most remarkable peculiarities of their houses is the almost universal concave form of roof, which writers on the subject have generally referred to as a reminiscence of the tent of the Tartars, who are supposed to have introduced it. The authors of this theory, however, forgot that the Chinese have been longer out of tents, and know less of them, than any other people now on the face of the globe. The Tartar conquest, like our Norman one, has long been a fusion rather than a subjection, and does not seem to have produced any visible effect on the manners or customs of the original inhabitants of China. It may also be observed that the typical form of the roof of a Tartar tent was and is domical, like those represented in the Assyrian sculptures, and seldom, if ever, constructed with a hollow curve; so that the argument tells the other way. Be this as it may, the form of roof in question arose from a constructive exigence, which others would do well to imitate. In a country like China, where very heavy rains fall at one season of the year, tiled roofs, such as they almost universally use, require a high pitch to carry off the water; but the glaring sunshine of another season renders shade to walls and windows absolutely necessary. If (as on the left of the annexed diagram) the slope of the roof is continued so far out as to be effective for the last purpose, the upper windows are too much darkened, and it is impossible to see out of them. To remedy this defect, the Chinese carry out their eaves almost horizontally from the face of the walls, where a leak becomes of slight importance; and then, to break the awkward angle caused by the meeting of these two slopes, they ease it off with a hollow curve, which not only answers the double purpose of the roof more effectually, but produces what the Chinese think--and perhaps rightly--the most pleasing form of roof.

The only parts of such a roof that admit of decoration by carving are evidently either the central or angular ridges; and here they exaggerate their favourite hollow curve to an extent unpleasing to a European eye--the angles being, in some instances, actually turned back, and the ridge being also ornamented by upturned ornaments at its ends, to an extent we cannot reconcile with our notions; nor indeed is it possible we should, when they are overloaded with grotesque ornaments to the extent too often found.

Another peculiarity that gives a very local character to their architecture is their mode of framing a roof, so unlike that of any other people. This arises from the timber most easily available for the purpose being a small pine, which has the peculiarity of being soft and spongy in the inside, while the outer rims of wood, just under the bark, retain their hardness and strength; it is thus practically a hollow wooden cylinder, which, if squared to form a framing as we do, would fall to pieces; but merely cleaned and used whole, it is a very strong and durable building-material, though one which requires all a Chinaman’s ingenuity and neatness to frame together with sufficient rigidity for the purposes of a roof.

The uprights which support these roofs are generally formed of the same wood, though not unfrequently they are granite posts--they cannot be called pillars--of the same dimensions, and strengthened, or rather steadied, by transverse pieces of wood, the space between which and the roof is generally filled with open-work carving, so as to form a species of frieze.

The roof is usually constructed (as shown in diagram No. 390) by using three or four transverse pieces or tie-beams, one over the other, the ends of each beam being supported on that below it by means of a framed piece of a different class of wood. By this method, though to us it may look unscientific, they make up a framing that resists the strongest winds uninjured. Sometimes, as shown in the dotted lines of the same woodcut, they carry the curve across the top of the roof; but, when this is done, they are obliged to have recourse to metal roofing, or to tiles of a greater length than are usually found or easily made.

As before remarked, however, it is not so much on its forms that Chinese architecture depends as on its colours--the pillars being generally painted red, the friezes and open work green; blue marks the floors and stronger lines, and gilding is used profusely everywhere. Whether this would improve a finer or more solid style of art may admit of doubt; but it is certainly remarkably pleasing in China, and singularly appropriate to the architecture we have been describing; and grouped as these buildings usually are around garden courts, filled with the gayest flowers, and adorned with rock-work and fountains more fantastic than the buildings themselves, the fancy may easily be charmed with the result, though taste forbids us to approve of the details.

The same ephemeral system of construction which prevailed in dwellings of the rich merchants and mandarins was carried out in the royal palaces without any increase of monumental character, but, of course, with greater richness of ornament, and upon a larger scale. Like most Oriental palaces, however, those at Pekin consist of a number of detached pavilions, rather than of numerous suites of apartments grouped under one roof, as is usually the case in Europe; and they consequently never attain the magnitude essential to architectural dignity. In the Summer Palace at Pekin there were many detached pavilions similar to that represented in Woodcut No. 391, which, when interspersed with trees and water and rocky scenery, aid in making up a very fairy-like landscape, but in themselves can hardly be considered as objects of dignified architecture.

Occasionally, however, the Chinese attempted something more monumental, but without much success. Where glass is not available of sufficient size and in sufficient quantities to glaze the windows, there is a difficulty in so arranging them that the room shall not be utterly dark when the shutters are closed, and that the rain shall not penetrate when they are open. In wooden construction these difficulties are much more easily avoided; deep projecting eaves, and light screens, open at the top, obviate most of them: at least, so the Chinese always thought, and they have consequently so little practice, that when they tried solid architecture in a palace they could only produce such a pavilion as that figured in Woodcut No. 392, which, though characteristic of the style, cannot be praised either for the elegance of its form or the appropriateness of its ornamentation.

Perhaps their most successful efforts in this direction were when they combined a solid basement of masonry with a light superstructure of wood, as in the Winter Palace at Pekin (Woodcut No. 393). In this instance the height and solidity of the basement give sufficient dignity to the mass, and the light superstructure is an appropriate termination upwards.

This last illustration is interesting, because it enables us to realise more distinctly than any other example yet known, what must have been the effect of the palaces of Nineveh and Khorsabad in the days of their splendour. Like this palace, they were raised on a solid basement of masonry, and were themselves composed of pavilions of light and ornamental woodwork; the great difference being that they had flat-terraced roofs instead of those covered with tiles, as in snowy Pekin; but the resemblance is curious, and examples even more nearly akin might probably be found if looked for.

The engineering works of the Chinese have been much extolled by some writers, but have less claim to praise as works of science than their buildings have as works of art. Their canals, it is true, are extensive; but with 300 millions of inhabitants this is small praise, and their construction is most unscientific. Their bridges, too, are sometimes of great length, but generally made up of a series of small arches constructed on the horizontal-bracket principle, as nine-tenths of the bridges in China are, and consequently narrow and unstable. When they do use the true arch, it is timidly, and without much knowledge of its principles.

Their most remarkable engineering work is certainly the Great Wall, which defends the whole northern frontier of the country, extending over hill and dale for more than 1200 miles as the crow flies. It is, however, of very varying strength in different places, and seems to be strongest and highest in the neighbourhood of Pekin, where it has generally been seen by Europeans. There it is 20 ft. in height, and its average thickness is 25 ft. at the base, tapering to 15 ft. at the summit. There are also towers at short distances whose dimensions are generally about double those just quoted for the wall.

However absurd such a wall may be as a defensive expedient, it proves that 200 years B.C. the Chinese were capable of conceiving and executing works on as great a scale as any ever undertaken in Egypt. The wonder is, that a people who 2000 years ago were competent to such undertakings should have attempted nothing on the same scale since that time. With their increasing population and accumulating wealth we might have expected their subsequent works to have far surpassed those of the Egyptians. It, however, remains a problem to be solved, why nothing on so grand a scale was ever afterwards attempted.

In the rear of the Great Wall, in the Nankau Pass, there is an archway of some architectural pretension, and which is interesting as having a well-ascertained date, A.D. 1345.[659] Its dimensions are considerable, and it is erected in a bold style of masonry (Woodcut No. 394). The upper part is a true arch, though it was thought necessary to disguise this by converting its form into that of a semi-octagon, or three-sided arch. On the keystone is a figure of Garuda, and on either side of him a Naga figure, with a seven-headed snake hood, and beyond that a class of flowing tracery we are very familiar with in India about the period of its erection. Its similarity to the Nepalese gateway at Bhatgaon (Woodcut No. 174) has already been remarked upon, and altogether it is interesting, as exemplifying a class of Indian ornamentation that came into China from the North. If we had a few specimens of art penetrating from the south, we might find out the secret of the history of Buddhist art in China.

* * * * *

A few years hence it may be possible to attempt to write a history of architecture in China. At present, all that can be done is to describe the style as practised at the present day, and to point out in what respect it differs from the styles prevailing in neighbouring countries. Beyond this we shall not be able to advance till some qualified person, accompanied by a photographer, is enabled to visit the central and western provinces of the empire. Even then his visit will be of very little use, unless he is sufficiently familiar with the style as now known, to be able to discriminate between what is new and what is old, and by an extended series of inductions to check the absurdities of native tradition, and form his own opinion on the facts presented to him. Assuming all this, it is still doubtful whether the materials exist in China for any extended history of the art. Such facts as have come to light are not encouraging. Wood has been far too extensively used throughout for any very permanent style of architecture ever having been employed. But there are things in Cambodia, and other neighbouring states, which seem to have come neither from India, nor from any other country we are acquainted with, but are nevertheless of foreign origin, and must have been imported from some extraneous land; and it is difficult to say where we are to look for their originals if not in central or western China.

The same remarks apply to Japan. So far as our knowledge at present extends, there is not a single permanent building in the island of so monumental a character to deserve being dignified by being classed among the true architectural examples of other countries. It may be that the dread of earthquakes has prevented them raising their buildings to more than one or two storeys in height, or constructing them of more solid materials than wood. It may be, however, that the Japanese do not belong to one of the building races of mankind, and have no taste for this mode of magnificence. It is the same story as in China; we shall not know whether it is true that there are no objects worthy to be styled architecture in Japan till the island is more scientifically explored than it has been; nor, if they do not exist, shall we till then be able to say to which of the two above causes their absence is to be ascribed. Such information as we have is very discouraging; and it is to be feared that, though quaint and curious in itself, and so far worthy of attention, it is of little interest beyond the shores of the islands themselves. On the other hand, it is to be feared that the extent of our knowledge is sufficient to make it only too clear that the art, as practised in Japan, has no title to rank with that already described in the preceding pages, and consequently no claim to a place in a general history of architectural art.

However admirable and ingenious the modern Chinese may be, it is in the minor arts--such as carving in wood and ivory, the manufacture of vessels of porcelain and bronze, and all that relates to silk and cotton manufactures. In these they certainly excel, and reached a high degree of perfection while Europe was still barbarous, but in all the higher branches of art they take a very low position, and seem utterly unprogressive.

They have no poetry, properly so called, and no literature worthy of the name. Their painting never rose much above the scale of decoration, their sculpture is more carving than anything we know by the higher name, and their architecture stands on the same low level as their other arts. It is rich, ornamental, and appropriate for domestic purposes, but ephemeral and totally wanting in dignity and grandeur of conception. Still it is pleasing, because truthful; but after all, its great merit in the eyes of the student of architecture will probably turn out to rest on the light it throws on the earlier styles, and on the ethnographic relations of China to the surrounding nations of Eastern Asia.

APPENDIX.

APPENDIX A.

ON SOME DISPUTED POINTS OF INDIAN CHRONOLOGY.

Throughout the preceding pages the dates of kings’ reigns, where quoted, have been assumed as known, and the eras from which they are calculated as ascertained. This has been done in order not to interrupt the narrative of events by introducing a chronological disquisition at every point where a date occurs; but no one at all familiar with the subject needs to be told that the dates of mediæval dynasties in India are far from settled, and that few are universally acquiesced in. Great progress has, it is true, been made in the last ten or twenty years in clearing away the difficulties that surround the subject. So much is this the case, that there are only one or two dates of sufficient importance to affect our reasoning which still remain in doubt; but though this may be true, there are many others about which the world in general feel considerable hesitation. It consequently becomes almost indispensable to state briefly the grounds on which the chronology used throughout this work is based, in order that the correctness of most of the inductions stated in it may be estimated at their true value.[660]

* * * * *

The earliest reasonable statement bearing on the subject which we possess is in the 9th chapter of Arrian’s ‘Indica.’ It is there stated--quoting from Megasthenes--“That from Bacchus (Ixwaku) to Sandrocottus (Chandragupta), the Indians reckon one hundred and fifty-three monarchs, who reigned during the space of six thousand and forty-two years.”

The first part of this statement is eminently satisfactory, as it seems clear from it that we possess in the Puranas the same lists as were submitted to the Greeks in the fourth century B.C. In the Solar lists, we have in the Treta Yug sixty-two reigns, from Ixwaku to Rama.[661] There is no complete Lunar list in that age. For the Dwapar age we have three Solar lists: one for Kusha to Vrihadsana, thirty-five reigns; another from Dishta to Janamejaya, thirty-three reigns; and a third, from the son of Swadhaja, the father of Sita, wife of Rama, to Mahabasi, thirty-four reigns. In the Kali Yug we have no complete Solar list, but the Lunar list gives fifty descents from Jarasandha to the last Nanda. This gives 145 or 146 reigns, or rather too few. But the Lunar lists, from the Dwapar Yug, give forty-four from Puru to Yudhishihira, and fifty from Yadu to Krishna, so that the average is as nearly as may be that stated by Megasthenes.

The second part of the statement, giving these kings’ reigns an average duration of nearly forty years, must of course be rejected, but it is satisfactory to find that, at that early age, the falsification of the chronology had only gone to the extent of duplication, and that the monstrous system of Yugs, with all their attendant absurdities, had not then been invented.

Though it may not at present be capable of direct proof, I have myself no doubt that the date assigned by the Hindus for the Kali Yug (3101 B.C.) is a true date, though misapplied. It either was the date when the Aryans assumed that their ancestors had first crossed the Indus, or when they had first settled on the banks of the Saraswati or the Ghoghra. It forms no part of any subsequently invented system, and seems the only one fixed point in a sea of falsification. Assuming it for the present, and deducting Chandragupta’s date from it, we have 3101-325 = 2776 years from Ixwaku to Chandragupta, which, divided by 153, gives the reasonable number of eighteen years for the duration of each king’s reign. Of course it is not contended that these lists are absolutely to be depended upon--many names may be lost, and many misplaced, from the carelessness of copyists, or from other causes; but, on the whole, when treated in this manner, they afford a reasonable framework for the reconstruction of the ancient history of India, and one that accords perfectly with all we at present know about the ancient history of the immigrant Aryans.

If this view can be sustained, the events which are described in the Ramayana--not of course the poem, which is comparatively modern--took place about 2000 years before Christ. Adhering to the above average, we gather that the events described in the ‘Mahabharata,’ in like manner, occurred 900 years before Chandragupta, or 1225, or more precisely, according to the Puranic chronology, thus--

B.C. Chandragupta 325 Sisunagas, 360 years 360 Sunakas 128 Sabadeva to Ripunjaya, 23 reigns at 18 years 414 ---- 1227 ----

which may probably be taken as very near the true date.

It must for the present remain an open question whether the dates just quoted can be so established as to stand the test of the exigencies of modern critical acumen. It would be very satisfactory if this could be so accomplished. In the first place, because it would afford a firm basis for all our reasoning regarding the ancient history and ethnography of India, but also because it would prove that the Puranas do contain the germs of truths which, when properly investigated, may lead to the most important deductions. My own impression is entirely in favour of the existence of the requisite materials for the purpose; but the fashion has been lately to pooh-pooh the whole thing, and no attempt has been made--so far as I know--by any competent scholar, to investigate the matter on scientific principles.

Be this as it may, when we come to the Anjana era, 691 B.C.,[662] and the life of Buddha, we tread on surer ground; and it is fortunate for our purposes that it so, as with the life of Buddha the mediæval history of India may be said to commence, and unless his date and that of his successors can be established with at least approximate certainty, the history of architecture in India must remain unintelligible. In this instance, however, the materials, I believe, exist in abundance. They have not, it is true, been as yet investigated to such an extent as to render any point certain, but the difficulties are daily disappearing, and as every point gained adds materially in throwing light on others that have hitherto been considered unsettled, we may hope before long to see the whole satisfactorily resolved.

* * * * *

There is perhaps no single point in the whole early history of India on which the chronicles of Ceylon and Further India are so distinct and unanimous than that Buddha died--as they express it, attained Nirvana--at the age of eighty years, in the year 543 B.C., or in the year 148 of the Eetzana[663] or Anjana epoch.[664]

Attempts have recently been made, it appears to me on the most illogical and insufficient data, to invalidate this conclusion. There is an admitted falsification in the Ceylonese annals, as set forth in the ‘Mahawanso,’ of sixty years about this date; but as Turnour, who first pointed it out, explained also the reason for it,[665] the rectification is easy, and the result clear. It seems that Vijaya, the first Indian immigrant or conqueror of Ceylon, landed in the island 483 years B.C., or thereabout; and the reigns of his successors, down to Devenampiyatisso, the contemporary of Asoka, when added together, amount to only 236 years. When the annals came to be expounded in the ‘Mahawanso,’ it was thought expedient, for the good of religion, that the coming of Vijaya should be coincident with the death of Buddha; and as the sacred era could not be disturbed, Asoka’s reign was carried back so as to admit of the adjustment. This was effected principally by reducing the epoch of the nine Nandas from 100 years, at which the Puranas place them, to forty-four, and by other slight alterations. The sixty years was afterwards recovered by small increments to subsequent reigns, not of much consequence, but injuriously affecting the correctness of the whole chronology of the ‘Mahawanso,’ down to about A.D. 400, when it was compiled in its present form. As the date of Asako’s reign is perfectly well known (272-236 B.C.), we have only to reject the most improbable coincidence of Vijaya landing on the day of Buddha’s Nirvana, which there is nothing to support, and the whole becomes clear, and everything falls into its place.[666]

Besides the Ceylonese lists, and those quoted by Crawfurd from the Burmese annals,[667] the Puranas afford us two, quoted below, which are of great interest to us, and the whole are so marvellously coincident, that there seems very little doubt of their general authenticity.

With regard to the first or Solar list, Professor Wilson remarks, that “Sakya is no doubt the name of the author or reviver of Buddhism, but is out of place, as he was the son and not the father of Suddhodana.”[668] This, however, is only one of the numerous instances in which the grandson takes his grandfather’s name, and which is an interminable cause of confusion in Indian chronological inquiries.[669] Gautama, as we know, never ascended the throne, but devoted himself to his religious duties, but his son Ratula succeeded his grandfather. In like manner, the Prasenajit in the list is not the cousin and companion of Buddha, but the grandson, or grand-nephew of that earlier king of the same name. Sumitra, the last name mentioned in the Bhagavat Purana, seems to have ascended the throne about 451. There are no exact dates for fixing this event, and with him perished the long line of Solar monarchs, who for more than twenty-six centuries--if our chronology is correct--had influenced in so marked a manner the destinies of India.

It was during the reign of Kalasoka, the eleventh king of this dynasty, that the second convocation was held, 100 years after the Nirvana. This, too, it has recently become the fashion to doubt. The accounts, however, in the ‘Mahawanso,’ and the pointed mode in which it is referred to in the Burmese annals, seem sufficient to settle the point. Like Vijaya’s landing in Ceylon on the day of Buddha’s Nirvana, Prome is said to have been founded 443, the year of this convocation.[670] They must have believed strongly, or they would not have attempted the adjustment.

As before mentioned, we have neither buildings, nor coins, nor inscriptions belonging to this period, nor indeed any material facts that would enable us to verify the chronological data. It is, however, so near the time when these became abundant, that it does not seem unreasonable to hope that some such evidences may turn up. Till something is found, the absence of all such materials must remain as a curious piece of evidence regarding the important influence that the contact of the nations of the West had on the arts and civilization of India at the time.

MAURYA, SUNGA, AND KANWA DYNASTIES.

CHRONOLOGY. BUILDINGS.

_Maurya Dynasty, 130 years._

The chronology of these three dynasties, as recorded in the Puranas, may admit of some adjustment in detail; but the whole is so reasonable and consistent that it can hardly be to any great extent. The whole, too, is now found to be so perfectly in accord with the architecture of their age, and with such inscriptions as have been found, that I see no reason whatever for doubting its general correctness.

The cardinal point on which the whole hinges is the twelfth year of Asoka’s reign after his consecration--the sixteenth from his inauguration. In that year he published his rock-cut edicts, in which he mentions his allies, Antiochus and Antigonus, Ptolemy (Philadelphus), Magas (of Cyrene), and Alexander (of Macedonia).[671] As it happens, all these five names are mentioned together in Justin’s abridgment of Trogus Pompeius (xxvi. 2, 3 and xxvii. 1), though without giving any date. As Magas, however, died B.C. 257, and the only year in which all five were alive together was either that year or the preceding, we may safely assume that the sixteenth of Asoka was B.C. 256 or B.C. 257. If that is so it seems impossible to bring down the date of the accession of Chandragupta to a time more modern than one or two years after B.C. 325. The Ceylonese annals allow him thirty-four years,[672] but our knowledge of what happened in India in Alexander’s time forbids any such extension. On the other hand, his accession happening in the year, or the year after, the defeat of Porus, is not exactly what we would expect from the context; but there is nothing, so far as I know, to controvert it.

Even if it were not so certain as it appears to be from the statements just quoted, there can be no doubt that the chronology of this period can easily be settled from the numerous inscriptions found in the rock-cut excavations quoted in the table, as well as from coins and other materials that exist. These dynasties thus become a fixed starting-point for all our inquiries, either backwards or forwards.

ANDRA, OR ANDRABRITYA DYNASTY.

CHRONOLOGY. BUILDINGS.

For this dynasty, as for the preceding three, we are dependent on the Puranas; but its chronology, like theirs, is so reasonable and so consistent with what we learn from other sources that I see no reason whatever for doubting its general correctness. There are slight discrepancies of course, not only as to names but as to the duration of this dynasty in the different Puranas. Thus the Vishnu Parana, according to Wilson, enumerates thirty kings, reigning 456 years; the Vayu and Bhagavat the same. The Matsya gives only twenty-nine kings, but makes them reign 460 years; but none of them give all the names, nor does the addition of the longest list extend beyond 435 years.[673] The whole, from Chandragupta to the last, are also added together (p. 232), and make up 751 years, or bringing the last of the Andras down to A.D. 426. The actual fixation of these dates will probably be found in Nassick cave inscriptions. Two of these bear dates: one, apparently in the reign of Pulomavi, or Padma, is dated nineteen from an unspecified era; the other is in the twenty-fourth year of the “modern era,” and the act recorded is, apparently, by order of Gautamiputra.[674] As it is, however, almost certain that the Gupta era, A.D. 319, was established in the reign of the last-named king, it seems probable that when these inscriptions are more carefully examined than they hitherto have been, they will fix these reigns with even greater certainty than we obtain from the Puranic dates; the one element of uncertainty being that the new era does not seem to be dated either from the accession of the king or from any great event, but four cycles of sixty years, or 240 years from the Saka era it was intended to supersede.[675]

However this may be settled, it cannot disturb either the initial or the final dates of this dynasty, nor affect to a greater extent than say ten or twelve years the period of 751, which extended from the accession of Chandragupta to the final overthrow of the Andras in or about A.D. 426.

This being so, it is evident that these four dynasties form the backbone of our mediæval chronology of India to which all minor events must be fitted, and fortunately most of them do so without any difficulty. It was the great period of Buddhist supremacy in India. There were, it is true, Buddhists in India before Asoka, but they were then only a sect, and Buddhism was a religion for two centuries after the fall of the Andras. It was then, however, a struggling faction. The modern Hindu religion was gradually raising its head under the Gupta and Ujjain princes, and in the 8th century it superseded Buddhism in most parts of India.

A great part of the uncertainty that of late years has crept into the chronology of this period is owing to the neglect with which these dynasties have been treated by modern investigators. This has arisen principally from the extreme rarity of their coins, while it has been principally from numismatic researches that progress has been made in the elucidation of many dark passages of Indian history. Coinage was, however, a most distinctly foreign importation into India. The Bactrian Greeks were the coiners _par excellence_, and it is through their coins, and those only, that complete lists of their kings down to 130 B.C. have been compiled. It is only from their coins also that we know the names of the barbarian kings who succeeded them, or those of the Sah kings, who appear next in our list. But the four dynasties from Chandragupta to Chandrasri were of native kings, who had only indirectly, if at all, come in contact with the Greeks, and had never learnt the art of coining, or, at least, used it to a sufficient extent to enable us to identify their names or succession from their coins. Their caves, and the inscriptions with which they covered their walls, are fast supplying the information their coins, if they had existed, would have afforded; but the investigation has not been taken up by those who have the ear of the public to the same extent as the numismatists. Enough, however, has been done to show that the materials exist for establishing the history of these dynasties on a sure basis; and when this is done from inscriptions combined with architecture, the results are more satisfactory than when dependent on numismatic evidence alone.

SAH KINGS OF SAURASTRA.

COIN DATES. A.D. Nahapana 79 -- Ushavadata -- -- Swami Chastana -- -- Jaya Dama -- -- Jiva Dama -- -- Rudra Daman 72 151 Rudra Sinha 102 181 Rudra Sah 104 183 Sri Sah -- -- Sangha Daman -- -- Daman Sah 144 223 Yasa Daman -- -- Damajata Sri -- -- Vira Daman -- -- Isvara Datta -- -- Vijaya Sah 170 249 Damajata Sri -- -- Rudra Sah 197 276 Visva Sinha -- -- Atri Daman -- -- Visva Sah 200 279 22. Rudra Sinha 270 -- Asa Daman 271 280 Swami Rudra Sah 292 371[676] Swami Rudra Sah II. -- --

The evidence on which the dates in the above list are founded is in curious contrast with that on which those of the previous dynasties rest. It is almost wholly numismatic. The founder of the dynasty, Nahapana, describes himself as the viceroy or satrap of King Kshaharata,[677] certainly a foreigner, who conquered the country and held it in subjection for nearly 300 years.

The one point that interests us here is to ascertain from what era the dates on the coins are to be calculated. When I previously wrote on the subject,[678] I felt inclined to adopt a suggestion that Nahapana was the founder of the era known afterwards as that of Vicramaditya, B.C. 56. I did this principally because I felt certain that no king of that name reigned in the first century B.C., and I could discover no event occurring about that time so important as to deserve to be commemorated by an era.

On the other hand, a foreign conquest and the foundation of a new dynasty were just such events as would be so celebrated; and, pending further evidence, this assumption seemed to account for what was otherwise inexplicable in the foundation of this era. Since then, however, a more careful study of Rudra Daman’s Bridge inscription,[679] and the architectural evidence detailed in the preceding pages, have convinced me that such a theory was untenable. The Bridge inscription is dated in the year 72, from the same era from which all the coins of these kings are dated. In it he boasts “that, after twice conquering the Sata Karni, Lord of Dakshinapatha, he did not completely destroy him on account of their near connexion, and thus obtained glory.” And he boasts of conquering, among other countries, Anupa, Saurastra, Asva Kutcha, Kukura, Aparanta, &c.[680]

A little further on in our history, Gautamiputra, in whose reign the era was established which was afterwards adopted by the Guptas and Ballabhis, boasts, in an inscription in a cave at Nassick, that he had conquered, among others, all the countries above enumerated, and as having re-established the glory of the Satavahana dynasty, and destroyed the race of Khagarata.[681] All this reveals a state of matters that will not accord with the Vicramaditya era, but does perfectly agree with that of Salivahana.

Assuming that the Sata Karni dynasty is correctly represented in the Puranas, as enumerated above, Rudra Dama would, on the assumption that the dates were Samvat, have been reigning A.D. 16 (72-56), immediately after the establishment of the dynasty, and before the long and prosperous reign of Sata Karni II., which could hardly have taken place had his family been smitten so early in their career. But if we assume that it was A.D. 151 (79+72), it would coincide with the reign of the third king of that name, and at a time when, so far as we can judge from the length of the reigns, and the careless way they are enumerated in the Puranas, the fortunes of the family were considerably depressed; and it is little more than a century and a half after this time that Gautamiputra restored the fortunes of his family. Had 300 years elapsed between these two events, the family could hardly ever have attained the position it did.

Another point of more importance is, that the dates on the Sah coins--from whatever era calculated--extend only to 270-271, or doubtfully to 292.[682] If these are calculated from the Vicramaditya Samvat, they must have ceased to reign in A.D. 214, or at the latest A.D. 236, and there would have been no Khagaratas for Gautamiputra to humble after A.D. 312. On the other hand, if calculated from A.D. 79, their final extinction would have been in A.D. 349, or at latest A.D. 371. So that, though humbled by Gautamiputra, they overlap the Gupta era to some extent, which it seems is almost indispensable to account for the mode in which the Sah coins overlap and run into those of the Gupta series, on which Mr. Thomas so strongly and, it appears to me, so correctly insists.[683]

One of two things seems necessary: either that the Guptas shall be carried back so as to overlap the Sahs, dating either from the Vicramadityan or Selucidan eras, or that the Sahs be brought down so as to overlap them, if dating from the era bearing their name. Mr. Thomas and General Cunningham prefer the former hypothesis. For the reasons just stated, and others to be given further on, I feel convinced that the latter hypothesis is the only one that is in accordance with the facts of the case as we now know them.

This substitution of the Saka era for the Samvat brings what we know of the history, with what we learn from the inscriptions, and gather from the coins, so completely into accordance, that I can hardly doubt now that it is the correct view of the matter, and certainly more in accordance with the facts than that I previously adopted.

GUPTAS.

Although the Puranas conduct us in so reasonable and satisfactory a manner to the end of the Andrabritya dynasty, their guidance forsakes us there. After that, all the subsequent contemporary dynasties were thrown into hotch-pot--to use a legal expression--and a system of fraud and falsification commenced which is the reproach of Indian history. It is not, however, difficult to see the causes of this new and monstrous invention. For six centuries and a half Buddhism had reigned supreme in India, and the system of the Brahmans, though probably never extinct, was at least subdued and subordinate. With the decline of the Andras this state of affairs was altered. The Guptas, who immediately succeeded them, are shown, both by their coins and inscriptions, to have been followers of Vishnu and Siva,[684] and their buildings at Erun tell the same story.[685]

Though the Guptas may have inaugurated the new system, it was by the great Vicramaditya of Ujjain that it was established, A.D. 515-550. He did for the new religion what Asoka had done for Buddhism some seven and a half centuries before his time. He made a state religion in India, and established it so firmly that little more than a century after his death it seems to have superseded Buddhism altogether. It is in his reign, apparently, that the Puranic system was invented--not that the Puranas were written or all the falsifications of history invented in his day, but a commencement was then made, and by the 10th or 11th century of our era it was brought to the complete perfection of fraud in which it is now found.

One of the first necessities of the new system was to throw back the period when India was Buddhist, and to place a gulf between them and their successors. To effect this, the Puranas enumerate the following:--“After these” (the Andrabrityas) “various races will reign--seven Abhiras, ten Gardabhilas, sixteen Sakas, eight Yavanas, fourteen Tusharas, thirteen Mandas, eleven Maunas or Hunas[686]--seventy-nine princes will be sovereigns of the earth for 1399 years. Then eleven Pauras will be kings for 300 years; when they are destroyed, Kailakila Yavanas will be kings, the chiefs of whom will be Vindhya Sacti, &c.--106 years.” After various others: “The nine Nagas will reign in Padmavati, Kantipura, and Mathura; and the Guptas of Magadha along the Ganges to Pryaga.”[687] Although we cannot identify all these dynasties with certainty, we know, at all events, that, instead of succeeding one another during more than 2000 years, they were all more or less contemporary--certainly that none were earlier than the Gupta era (A.D. 319)--and that none of them survived Vicramaditya (A.D. 550). The Sakas and Maunas, or Hunas, may be those destroyed by him, but of this hereafter. The Vindhya Sactis were contemporary with the Guptas, and the Gardabhilas are somehow connected with Bahram Gaur the Sassanian; and others we recognise dimly, but they are not sufficiently important to be discussed here.

* * * * *

Of all these the most important are the Guptas, and fortunately their date is one of the most clearly established facts in mediæval Indian chronology.[688]

DYNASTY. COINS AND DATES ON INSCRIPTIONS. A.D. Sri or Raja Gupta -- -- Maharaja Ghatotkacha -- -- M. R. adhiraja Chandra Gupta I. 82, 93+319 = 401, 412. Caves 16 to 20 Ajunta. Buildings at Erun. “ Samudra “ -- -- “ Chandra Gupta II. -- -- “ Kumara “ 124+ “ = 443 “ Skanda “ 130, 137, 141, 146+ “ = 449, 456, 460, 465 Mahendra a minor -- -- Maharaja Sri Hastina 163+ “ = 482 Raja Buddha 165+ “ = 484 M. R. adhiraja Toramana 182+ “ = 501

The three last named can hardly be considered as belonging to the great dynasty, though they date from the same era, and the two first were comparatively insignificant characters. It was only Chandra Gupta I., A.D. 401, who assumed the title of Maharaja adhiraja, and founded the greatness of his race on the ruins of that of the Andrabrityas.

In addition to the above chronology, compiled from coins and dated inscriptions, Major Watson has recently supplied a most important item to their history from written records existing in Gujerat.

From this we learn that Chandra Gupta II. reigned twenty-three years after the conquest of Saurastra by his son; that Kumara Pal Gupta reigned twenty years; and that Skanda Gupta succeeded him, but lost Saurastra by the rebellion of his Senapati Bhatarka, the founder of the Ballabhi family. Two years after this event Skanda Gupta died, and, as we are informed, “at this time the Gupta race were dethroned by foreign invaders.”[689]

The era from which these dates are taken never appeared to me doubtful; and this confirms more and more the conviction that it was from the era that bears their name, A.D. 319. It could not be from the Saka era, as has generally been assumed, from the fact that Albiruni asserts that the era that bears their name, was “apparently” that of their destruction,[690] because in that case Skanda Gupta must have lived and reigned for ninety-four years in addition to the sixteen we already know, from inscriptions, he occupied the throne. A reign of 110 years seems impossible; and, if it is not so, it seems certain, for the reasons stated in my previous paper, that the Gupta era, 319, is that from which their coins and inscriptions are dated.

Besides this, there is an inscription on the rock at Junaghar, engraved by the same Skanda, the last of the great Guptas. This was not translated by Prinsep, though a copy of it was in his hands before his last illness.[691] Had he lived to translate it, my impression is that the controversy as to the age of the Guptas never would have arisen--its evidence seems so absolute. Be this as it may, it never appeared, so far as I know, in a complete form and translated, till this was accomplished by the late Bhau Daji in the sixth volume of the Bombay Journal of 1862. In it we have three dates--the Sadarsana lake is said to have burst its banks in 130, to have been repaired in 137, and a temple to Vishnu built in 138, and twice it is repeated “_counting from the era of the Guptas_” (Guptasya Kala). The stone is worn where the middle date occurs, but there is just space enough for these words. The same king, on the Kuhaon pillar, dates his inscription in 141,[692] but without mentioning the era, which seems to have been so usual in Bengal as not to require being specified.

Besides this, the 146[693] years from 319, which we know from their dated inscriptions that they reigned, is just the interval that is required to fill up the gap between the Ballabhis and their era which they adopted on usurping the inheritance of the Guptas, two years before Skanda Gupta’s death.[694]

One other point of considerable importance to Indian history which arises from the fixation of this date (A.D. 465-70) for the destruction of the Guptas is, that it was almost certainly the White Huns who were the “foreign invaders” that struck the blow that stopped their career. At least, we learn from Cosmas Indicopleustes, writing seventy years after this time, that the Huns were a powerful nation in the north of India in his day, and we may infer, from what he says of them, had been settled there some time.[695]

On the Bhitari Lât, Bhau Daji reads--somewhat doubtfully, it must be confessed--the fact that Skanda Gupta had fought, apparently with success, against the Hunas.[696] But the great point is that it was just about this time that the White Huns broke loose and extended their incursions east and west, so that there is not only no improbability of their being the “foreign invaders” alluded to, but every likelihood they were so. No one, indeed, can, I believe, with the knowledge we now possess, read De Guignes’ chapter on the White Huns,[697] without perceiving that it contains the key to the solution of many mysterious passages in Indian history. It is true India is not mentioned there; but from the time of Bahram Gaur in 420, till the defeat of Feroze in 475, the Persians were waging an internecine war with these Huns, and nothing can be more likely than that the varying fortunes of that struggle should force them to seek the alliance of the then powerful Guptas, to assist them against their common foe.

Precisely the same impression is conveyed by what is said by Ferishta and the Persian historians[698] of the history of that time. Nothing can now, however, be more easily intelligible than the visit of Bahram Gaur to India when first attacked by the White Huns. His marriage with an Indian (? Gupta) princess of Canouge; the tribute or assistance claimed by Feroze and his successors on the Persian throne, are all easily explicable, on the assumption that the two nations were at that time engaged in a struggle against a common enemy. This, too, explains the mention of the Shah in Shahi on Samudra Gupta’s Allahabad inscription.[699] Hence, too, the decided Persian influence on the gold coinage of the Canouge Guptas,[700] and the innumerable Sassanian coins of that period found in all parts of the north of India.[701] In all this the Sassanians seem inseparably mixed with the Guptas. The Persians, however, came eventually victorious out of the war. The great Guptas were struck down at some date between 465-70, or very shortly afterwards. The struggle, however, was apparently continued for some time longer by a subordinate branch of their successors; inasmuch as we learn from an inscription found at Aphsar in Behar,[702] that the fourth of that dynasty, Damodara Gupta, “successfully encountered, at the battle of Maushari, the fierce army of the Western Huns.” This event may have stopped the career of the Huns in India, in which case it could not well have taken place before the year 535, when Cosmas Indicopleustes is supposed to have written his ‘Topographia Christiana;’ but it is by no means clear that he was not describing events that took place when he was himself in India some time previously. But be this as it may, it brings us to the time when the battles of Korûr--of which more hereafter--and Maushari freed India from the Sakas and Hunas, who had long held her in hated subjection. As I shall presently attempt to show, it appears to me hardly doubtful that these two battles were fought between 524 and 544; and they thus fix one of the most important epochs in mediæval Indian history. Indeed, so near each other are these two events in date, that I sometimes feel almost inclined to fancy they may be only different names for the same battle. At all events, they almost certainly represent parts of the same campaign which freed India in that age from the Yavanas; and that it was to commemorate the glories of these struggles that the Vicramaditya Samvat was instituted. This expulsion of the Yavanas was, too, the first serious blow that was struck at Buddhist supremacy, and from the effects of which it never afterwards completely recovered.

BALLABHI DYNASTY.

DATES ON INSCRIPTIONS. A.D. Bhatarka Senapati -- 465 or 470 Dharasena “ -- -- Dronasinha -- -- Dhruvasena Maharaja -- } Cotem. Vicramaditya Dharapatta -- } Dynasty Grihasena -- } of Ujjain, Sridhara Sena -- } 470 to 550. Siladitya I. -- -- Charagriha I. -- -- Sridhara Sena II. 272 591 Dhruvasena II. -- Cotem. Hiouen Thsang Sridharasena III. -- -- Siladitya II. 356 675 Charagriha II. -- -- Siladitya III. -- -- Siladitya Musalli 400 718

However mistaken Albiruni may be in his dates, there is little doubt that he is quite correct in his statement to the effect that “L’ère de Ballabha est postérieure à celle de Saca de 241 ans. Apparemment Ballabha suivit immédiatement les Gouptas, car l’ère des Gouptas commence aussi 241 de l’ère de Saca.”[703] This we learn also, with the particulars how it happened, from Colonel Watson’s account of the transaction; while Colonel Tod’s celebrated Puttun Somnath inscription makes it also certain that the Ballabhi era commenced A.D. 319.[704] This being so, it seems difficult to understand why the era should have been called that of Ballabhi as well as that of the Guptas, unless it were that it was adopted by the first-named dynasty, and that they dated from it their acts and inscriptions, which are extremely numerous. There may be reasons why this should be otherwise; but, though the point has been generally and fiercely contested by eminent Indian chronologists, I fail to appreciate the arguments brought forward in favour of either the Vicramaditya or Saka eras,[705] and look upon their own era (A.D. 319) as certainly the one from which all the Gupta inscriptions are dated.

My impression is, that this would never have been considered doubtful but for an incautious statement by Colonel Tod that Ballabhi was destroyed by the Parthians A.D. 524,[706] in the reign of a Siladitya, its last king. Its inhabitants were, according to this account, slaughtered with the usual romantic incidents; but after a while a remnant established themselves in Sidhapore, and finally built a new capital, which they called Anhilwarra.

The utter falsity of the information so supplied to Colonel Tod is proved by the fact that when Ballabhi was visited by Hiouen Thsang, 115 years after its reputed destruction, he found it not only standing, and neither Sidhapore nor Anhilwarra thought of, but the old capital still remaining one of the richest and most prosperous cities of India, and its king one of the three greatest kings of northern India. The king’s name was Dhruvapaton, and he was a nephew or grand-nephew of Siladitya of Malwa, and the son-in-law of Siladitya, the reigning king of Canouge.[707] Lastly, we have the dates in copper-plates of a Dhruvasena, one in 310+319=629; the other 322+ 319, or 641,[708] the very year that Hiouen Thsang met him at Allahabad, if we assume them dated from the Ballabhi Samvat.

It would be satisfactory if we could determine the date of the destruction of Ballabhi with precision, as it is one of these events that mark an epoch in Indian history. It was one of the concluding acts of the old drama that closed the mediæval period of Indian history, and ushered in the dark ages which lasted more than two centuries from that time.

The materials for this hardly exist at present, though it may be approximated. We have numerous inscriptions of this dynasty, dated 310, 326, 338, 348, &c.,[709] or A.D. 629, 645, 657, 667 respectively, if the figures are all correctly read, which is not quite clear; and lastly, Mr. Burgess reports one dated 400, or A.D. 719, belonging to the last Siladitya, and consequently approaching very nearly to the event. Two accounts are current as to the mode in which the destruction was effected: one, that it was caused by an earthquake, which may have happened at any time;[710] the other (by Tod), that the city was destroyed by the Parthians. If it was by a foreign foe, it could only have been by the Mahomedans. They were on the Indus in strength in 22 Hegira,[711] or A.D. 644, or before Hiouen Thsang had left India, and no foreigner could have crossed the Indus or attacked Ballabhi after that time, or for some years before it, without being noticed by Mahomedan historians. They remained there in strength till after Mahomed Kasim, 711-715,[712] and it was to him that I was at one time inclined to ascribe the destruction. If, however, Mr. Burgess’s date is correct, his death was three years too early. But I do not think it at all improbable that Ballabhi is one of the cities--Barus and Uzain--said to be plundered by Junaid in A.D. 725 or 726.[713] Barus looks very like Baroach, and Uzain is almost certainly Ujjain--but whether Maliba is Ballabhi, I must leave others to determine.

All the accounts agree that Anhilwarra Puttun was founded Samvat 802, or A.D. 746,[714] which may be correct within a year or two; but from the accounts we have, it is clear that an interval of from twenty to thirty years must have elapsed between the two events, during which the inhabitants of the destroyed city sought refuge at Punchâsur and Sidhapore before they undertook the building of their new capital. If, therefore, we assume 725 as the date of the destruction of Ballabhi, we shall probably not err more than a year or two either way.

The earliest date of this family yet discovered is one on a copperplate of Dharasena II., which has been read by Professor Bhandarkar as 272,[715] or, according to the views here adopted, 591. It is hardly probable that any much earlier will be found; for it must be borne in mind that though the Ballabhis wrested the sovereignty of Gujerat from the Guptas two years before Skanda’s death (_ante_, p. 724), neither the first nor second of the race ventured to assume even the modest title of Raja; they were content to remain Senâpatis, or Generals. The third calls himself Maharaja; but their greatness only culminated in or about A.D. 650, when one of them, Sri Dharasena III., became Maharaja Adhiraja--King of kings or Emperor of Northern India.[716] The reason of this, as we shall presently see, was that the family that really succeeded the Guptas in the place of supreme authority in India was that of Ujjain, the second or third monarch of this race being the celebrated Vicramaditya, whose date, for reasons to be given hereafter, seems almost certainly to have been from 515 to 550. Be this as it may, as we shall presently see, it seems quite certain that a great Brahmanical revival took place in the beginning of the 6th century, which quite overshadowed all the Buddhist dynasties in northern India. For a while these were again eclipsed by a reflex wave of Buddhism, which for a century--A.D. 550-650--again illumined India. It was a last expiring effort, however, and after the last-named date it was only a struggle for existence on the part of the Buddhists, and in another century they are known no longer in those central countries where they had so long reigned supreme.

CHALUKYA DYNASTIES.

WESTERN BRANCH. | EASTERN BRANCH. CAPITAL KALYAN. | CAPITAL RAJMEHENDRI. 1. Jaya Sinha Vijayaditya. | 2. Raja Sinha, Rana Raga, Vishnu Vardhana.| 3. Vijayaditya II. | 4. Pulakesi, A.D. 489? | 5. Kirtti Varma I. | 6. Mangalisa. | 7. Satyasraya began to reign 609. |1. Vishnu Vardhana II., or Kubja | Vishnu Vardhana, conquered | Vengi A.D. 605. 8. Amara. |2. Jaya Sinha I. 9. Aditya. |3. Indra Raja, his brother. 10. Vikramadiya I. |4. Vishnu Vardhana III. 11. Vinayaditya, Yuddha Malla, began |5. Manga Yuva Raja. to reign A.D. 680. |6. Jaya Sinha III. } 12. Vijayaditya III. began to reign A.D. |7. Kokkili. } brothers. 695. |8. Vishnu Vardhana IV. } 13. Vikramaditya II. began to reign |9. Vijayaditya I. A.D. 733. |10. Vishnu Vardhana V. |11. Narendra Mriga Raja. 14. Kirtti Varma II. |12. Vishnu Vardhana VI., or Kali | Vishnu Vardhana. 15. Kirtti Varma III., cousin of the |13. Vijayaditya II., or Guna Gunanka last, A.D. 799. | Vijayaditya, conquered Kalinga. |14. Chalukya Bhima I., his brother. 16. Tailapa. |15. Vijayaditya III., or Kollabhiganda | Vijaya. 17. Bhima Raja. |16. Amma Raja. |17. Vijayaditya IV., or Kaudagachita 18. Ayya, or Kirtti Varma IV. | Vijaya. |18. Talapa. Usurper. 19. Vijayaditya IV. |19. Vikramaditya V., the son of a brother | of Amma Raja I. 20. Taila Bhupa II. or Vikramaditya III., |20. Yuddha Malla. in A.D. 973 restored the monarchy | which had been for some time | usurped by the Ratta Kula. He | died A.D. 997. | 21. Satyasraya II. Irivi Bhujanga Deva, |21. Raja Bhima II. A.D. 997. | 22. Vikramaditya V. began to reign |22. Amma Raja II. about A.D. 1008 (?) | 23. Jaya Sinha Deva, Jagadeka Malla, |23. Dhanarnava. Interregnum of | twenty-seven years. about A.D. 1018 (?) | 24. Someswara Deva I., Trailokya |24. Kirtti Varma, son of Dhanarnava. Malla Ahawa Malla, about A.D. | 1040. | 25. Someswara Deva II., Bhuneka Malla |25. Vimaladitya, his brother. A.D. 1099, expelled by his brother. | 26. Vikramaditya VI., Kali Vikrama, |26. Raja Raja Narendra. Tribhuvana Malla, in A.D. 1076. | 27. Someswara Deva III., Bhuloka |27. Rajendra Chola. Malla, A.D. 1127. | 28. Jagadeka Malla, A.D. 1138. |28. Vikrama Deva Kulottunga Chola. 29. Tailapa Deva III., Trailokya Malla, |29. Raja Raja Chola, viceroy for one year. A.D. 1150. | 30. Someswara Deva IV., Tribhuvana |30. Vira Deva Kulottunga Chola, or Malla, A.D. 1182. Dethroned by | Saptama Vishnu Vardhana. Viceroy Bijjala Deva of the Kalabhuriya line. | from A.D. 1079 to 1135. After this the southern part of these |After Vira Deva Kulottunga Chola the country dominions fell under the sway of the |fell under the sway of the Kakatya dynasty of Hoisala Bellalas, whose rise in the | Worangul, of whom Pratapa Rudra was the chief Mysore dates from A.D. 984; their |(A.D. 1162). The latest of their inscriptions is destruction by the Mahomedans in 1310. |dated A.D. 1336.

The two lists in the preceding page are among the most interesting and most important of those we possess, inasmuch as they contain the backbone of all we know regarding the Chalukyas, and are, in fact, what justify us, historically, in erecting their style into a separate division, different from the other forms of architecture known in India.

What we know of these dynasties is almost wholly due to the intelligent zeal of Sir Walter Elliot, who, during his residence in India, made a collection of 595 inscriptions from various parts of the Dekhan. From these he abstracted the lists he first published in the fourth volume of the Royal Asiatic Society; but afterwards much more in detail in the ‘Madras Journal,’ in 1858, from which these lists are copied verbatim.[717] Some of the inscriptions were translated and published with those papers, and others by Major--now General--- Le Grand Jacob, in the Bombay Journal (vol. iii. p. 206, _et seqq._), and other notices of them are found among Mr. Wathen’s inscriptions in various volumes of the ‘Journal of the Royal Asiatic Society.’ But we shall not know more than a fraction of what we ought to, and might know, till Sir Walter Elliot’s inscriptions are translated and published.[718] When this is done, and the architecture of the Nizam’s territory explored, the Chalukyan style will take its place worthily between the Dravidian and Indo-Aryan styles, and will, if I mistake not, be found equal to either, both in importance and in artistic merit.

* * * * *

Fortunately there is no mistake or doubt about the era from which the Chalukyan inscriptions are dated: the Ballabhi branch succeeding to the possessions of the Guptas in Gujerat, naturally adopted their era, but the southern branch being entirely detached from any such association, adopted the Saka era (A.D. 79), which was then, so far as is known, the only other era at that time in use in India. What is equally important is, that there seems only one doubtful date among all those quoted in the lists--that of 411 Saka (A.D. 490), attached to the name of Pulakesi I. In his first paper,[719] Sir Walter Elliot thought it so improbable, that he rejected it altogether; and Professor Eggeling tells me he has strong reasons for suspecting the copperplate on which it is found to be a forgery.

As an initial date it does not appear impossible, if my views are correct, though certainly improbable. If Bhatarka Senapati wrested Gujerat from Skanda Gupta two years before his death, or in 463 or 468, it is by no means impossible that the fourth from him may have been reigning in A.D. 490, but the difficulty is the other way. There seems no doubt, from Mr. Burgess’s Badami inscriptions,[720] that Mangalisa succeeded his brother Kirtti Varma in 567, and it does seem impossible that he should have been the son of one who was reigning in 490, especially if he continued to reign till 609. If Mangalisa was the son of Pulakesi, which there seems no reason for doubting, it is evident that the central figure of his date must be altered to a higher number; but to what extent we shall not know till it is ascertained whether Vijaya was the son or grandson of Bhatarka Senapati. In the meanwhile, however, if we, as an hypothesis, add fifty years to the date of 411, and make it 461, or A.D. 540, it will allow Pulakesi a reign of twenty-seven years before the accession of Mangalisa in 567, which will bring the whole within the limits of probability, and seems perfectly consistent with the context.

With the seventh king we tread on surer ground. He was the king who, when bearing his grandfather’s name, Pulakesi, Hiouen Thsang visited in 640,[721] and was, as his inscriptions tell us,[722] the hero of those wars with Harsha Verddhana, or Siladitya of Malwa, which Ma-twan-lin so graphically describes as occurring in 618 to 627. From that time the dynasty seems to have flourished till the death of Vicramaditya II. He ascended the throne 733, and died about 750, or twenty-five years more or less after the destruction of the Ballabhi branch. After this, as Sir Walter Elliot expresses it, “the power of the Chalukyas was alienated for a time, or had suffered a partial obscuration, till the time of Teila, who is described as restoring the monarchy in 973.”[723] After this it enjoyed two centuries of prosperity, till it was finally extinguished--their northern possessions passing to the Kalabhuryas--their southern to the Hoisala Bellalas of Dwarasamudra or Hullabîd.

The history of the younger branch of this family will be more interesting to some future historian of Indian architecture than it is to us at the present day. Their possessions lay principally below the Eastern Ghâts, on the shores of the Bay of Bengal, in what are generally known as the three Circars, extending from Gangam--in their day I believe--to Mahavellipuram; but of their architecture we know nothing. No traveller educated in architectural matters has yet visited that country; and though it sounds like a paradox to say so, what we do know of it we learn from buildings not erected by them, and in a country they never seem to have possessed. It is only from the buildings of Pratapa Rudra at Worangul and elsewhere above the Ghâts that we can appreciate the perfection to which they had brought their style.

From the meagre extracts from the inscriptions of Pulakesi I., which Sir Walter Elliot gives in his first essay on this subject,[724] there seems little doubt that he was the king who, 100 years before Hiouen Thsang’s time, harried the monastery at Amravati,[725] and abolished Buddhism in those parts. It seems also more than probable, as he conquered the Chola, and burnt Conjeveram, that he also expelled the Pallavas, and commenced the works at Mahavellipur. If the rock-cut monastery mentioned by Fa Hian and Hiouen Thsang, and so often referred to above, existed at all, it was in his territories, and may still exist in the Nizam’s. If it did so, nothing seems more probable than that he should seek to mark the boundary of his southern conquest by similar works. Knowing all this, we see also why there should be so much similarity between Mangalisa’s cave at Badami, and the nearly contemporary caves at Mahavellipur. We know, too, that there is a vast tract of country in Central India, extending east and west from shore to shore, and north and south from Sadras to Ellora, which is covered with buildings of great beauty and interest, but which nobody cares to explore. We know also that there exists in the Asiatic Society’s rooms a volume which contains their history, and that of the dynasties who built them, but which nobody cares to read. Knowing how easily all this could be remedied, it is tantalising to close this history with so meagre a sketch of the Chalukyan style as that contained in the preceding pages, but as the principles of the Indian Council seem fixed, its description must in all probability be relegated to a subsequent generation.

UJJAIN AND CANOUGE DYNASTIES. ------------------------------------+-------+------------ |REIGN. | DATE. +-------+-------- Vasu Deva | | Vicramaditya I. of Ujjain | 25 | 470? Sri Harsha | 20 | 495? Vicramaditya II. the Great | 35 | 515 Siladitya I. of Malwa | 30 | 550 Prabhukara | 25 | 580 Raja Verddhana | 5 | 605 Siladitya II. of Canouge | 40 | 610 Died and troubles commenced | -- | 648-650 ------------------------------------+-------+------------

Although the Ballabhis wrested the province of Gujerat from the failing hands of Skanda, the last of the Great Guptas, two years before his death, in or about 470, they remained long in a subordinate position. Their earliest inscription yet found dates only in 593, and their one Emperor or Raja Adhiraja, Sri Dharasena III., only ascended the throne after the Canouge dynasty were struck down in 648-50.

The interval between these two events we are now happily able to fill up with two of the most illustrious dynasties of India--the first including the reign of the great Vicramaditya of Ujjain, who is to the Hindus what Solomon is to the Jews, or Asoka to the Buddhists. The last-named religion, as mentioned above, was becoming effete about the middle of the 5th century, and the Guptas were introducing the modern Brahmanical faith in its place. What, however, they were only feebly attempting, the Ujjain dynasty accomplished with a brilliancy that has eclipsed everything that happened before or since in India, in the eyes of the Hindus at least. All that is great in science, or in poetry, or the arts, shone forth around his wonderful throne--the exact counterpart of Solomon’s--and all that subsequently took place in India bears the stamp of his greatness. It seems, however, to have been too bright to last. The four succeeding monarchs were Buddhists--of a singularly tolerant type it is true--but still certainly favourers of that religion. The last of them, Siladitya, was the king at whose court Hiouen Thsang sojourned in 636, and afterwards in 642, and where he witnessed the festival of the distribution of alms so often alluded to above. Hiouen Thsang gives the date of his death categorically, 650, and adds, though in the form of a prophecy, that after that, “l’Inde entière sera en proie à des troubles affreux--et des hommes pervers se feront une guerre acharnée.”[726] This is more than confirmed by Ma-twan-lin, but with an apparent discrepancy of date, to the extent, it may be, of two years.[727] It was in fact the commencement of those troubles which extinguished Buddhism, then in Central India, and a century later abolished it wholly, except in some remote corners of the land.

Whether he died in 648 or 650, there is no doubt, from the numerous incidents our Chinese traveller recounts, that this Siladitya ascended the throne 610, one year after his great rival, Pulakesi II., of Kalyan, who, as pointed out above, began to reign in 609, and fought with him with varying success in 618-627.

For the chronology of the four preceding reigns we have nothing but the assertion of Hiouen Thsang, that “suivant la tradition”[728]--and in another place, “on lit dans l’histoire de ce royaume,[729] que le trône était occupé il y a soixante ans par un roi nommé Siladitya;” and further, that he reigned fifty years, which would carry us back to 530 for the accession of this king, supposing the passage was written in 640.

Notwithstanding the confidence with which it is stated, I have no hesitation in rejecting as excessive 110 for the length of the reign of three kings, two of whom were brothers. I do so with the more confidence, as our author, though so exact a geographer, and recorder of things he saw, is in no one instance to be depended upon for his dates. He resided, for instance, for five years at Nalanda, and must have had access to its records, yet he tells us that the convent existed for 700 years,[730] and then gives the names of the five kings by whom the various parts were built from that time to his day, but sees no absurdity in representing these in all instances as the son of the one next named previously. Each, according to his account, must have reigned more than 100 years! To what extent this date of the accession of Siladitya must be curtailed can only be ascertained from subsequent discoveries or investigations. For the present it will suffice to abridge it by twenty years, which will bring it in accord with all that we at present know from other sources.[731]

When we turn to the other end of our list, we have certainly three--probably four kings--for whom we must find room in eighty years and one of the three, the great Vicramaditya, must have had a long reign. Professor Wilson ascribes to him thirty-five years,[732] and I know of no authority better than his, especially for the history or chronology of this period. The Hindus themselves, with their usual carelessness, have forgotten to record it; and though there are certain dates in the Puranas and elsewhere, there are no means of testing their accuracy; for his accession, however, there are one or two that are worth recording. Thus, Wilford reports that this Vicramaditya ascended the throne of Malwa 441,[733] reckoning from the first of Salivahana, or, 520; or, according to the Agni Purana, 437 years after the same epoch, or 516,[734] which, I believe, may be the exact year; and there are several other dates which might be used to confirm this assumption, but there are no means of testing the genuineness.

Assuming this for the present, it leaves only forty-five years for the two or three preceding reigns; and it seems hardly sufficient for the purpose, for, as we shall presently see from the ‘Raja Tarangini,’ there were nine descents between Pratapaditya, the friend of the first Vicramaditya, and Matrigupta, the protégé of the second. Of course there may be considerable overlapping among the first and last of these nine kings, but it seems impossible to compress the whole within a shorter period than has been allowed.

However the small discrepancies of this dynasty may hereafter be adjusted, it is satisfactory to know that there is probably no date that will admit of a greater correction than say ten years, if so much, and the age of the last king, Hiouen Thsang’s friend, enables us to feel perfectly certain as to the dates of his son-in-law, Dhruvasena, of Ballabhi, of Sasanka, of Pundra Verddhana, of Kumara, of Kamarupa, and of Pulakesi II. of Kalyan. We have thus at least one fixed point in our mediæval history which is quite certain, and from which we can calculate backwards and forwards without difficulty, and is also an interesting one, as its final date, 650, is the beginning of the end which was consummated, as we shall see in the next section, by Laladitya just one century later.

KASHMIR.

Asoka, 276 to 240 B.C. Jaloka. Damodara. Hushka } Jushka } Tartar Princes established Buddhism. Kanishka} Abhimanu, 79 A.D.?

GONARDYA DYNASTY.

Gonarda. _Naga worship restored._ Vibhishana. Indrajita. Ravana. Vibhishana. Nara. Siddha. Utpalaksha. Hiranyaksha. Hiranyakula. Vasukulo. Mihirakula, invaded Ceylon 250? Vaka.

Kshitinanda. Vasunanda. Nara. Aksha. Gopaditya, 330? Gokarna. Narendraditya. Yudhishthira.

ADITYA DYNASTY.

Pratapaditya, kinsman of Vicramaditya I., 390. Jalaukas. Tunjina. Vijaya. Jayendra. Arya Raja.

GONARDYA LINE _restored_.

Meghavahana invaded Ceylon, 472. Pravarasena I. Hiranya } Toramana} Contemporaries of Vicramaditya. Matrigupta, viceroy under Vicramaditya II., 515. Pravarasena II., invaded Siladitya of Gujerat, 560. Yudhishthira II. Nandravat. Ranaditya. Vikramaditya. Baladitya.

NAGA OR KARKOTA DYNASTY.

Durlabhaverddhana, 627. Pratapaditya, 663. Chandrapira, 713. Parapira, 721. Lalataditya, 725; died 761. Conquered Yasoverna of Kanouje, and overran India.

When the ‘Raja Tarangini’ is spoken of, in a real Indian history, it is only in the sense of the French proverb--“Parmi les aveugles les borgnes sont rois.” It may be the best, but it is a very indifferent specimen of its class. Some of the few events it narrates are interesting and important, but they lose much of their value from the chronology to which they are attached being wilfully and systematically falsified. Even they, however, may become more valuable than they now appear, when the work is better edited than it has been hitherto. The earliest and best account we have of it is that of Professor Wilson, in the fifteenth volume of the ‘Asiatic Researches.’ The translation, afterwards published by Troyer in French, is fuller, no doubt, but is made from a less perfect manuscript, and is far less critical. Dr. Geo. Bühler, who is now in the valley, is said to have collected several additional and more complete MSS., from which it is understood he is preparing a new edition of the work. When this is done, we may be able to use it more profitably; meanwhile, for chronological purposes, we can only try and find an initial and final date, and with one or two intermediate synchronisms, try to bring the whole into an intelligible sequence; but so hopelessly is the chronology confused by its author, that this at present can only be effected by the application of a system of averages, which is, and always must be, a most unsatisfactory mode of procedure.

* * * * *

Rejecting at once as worthless or hopelessly lost all those parts of the history before the third century B.C., the first name we come to is the familiar one of Asoka, but here placed 1394 B.C., or more than 1000 years too early. It was in order to recover what was lost by this first error that Kalhana Pandit was forced to falsify all the dates up to the accession of the Karkota dynasty (A.D. 627), when they were known, even in his day, as certain within ten or twenty years. To effect this, he added ten, twenty, or thirty years here and there, as caprice dictated, till at last, losing patience, he gave one king, Ranaditya, in the 6th century, 300 years, instead of a possible thirty, and so made both ends meet! So history is written in the East!

After Asoka’s, the next name we meet in the lists with which we are familiar is that of Kanishka, and he plays so important a part in the history of Kashmir and Gandhara, that it would be of extreme interest if his date could be fixed with even approximate certainty. The ‘Raja Tarangini’ gives us no help in this matter. Generally, it has been assumed, principally on numismatic evidence, that he reigned either immediately before or immediately after the Christian Era;[735] but between him and Asoka our lists afford only two names. If, therefore, we are to apply to this history the same logic the very learned have attempted to apply to dates of the Nirvana in the ‘Mahawanso,’ we must either bring down Asoka to the first century B.C., or take back Kanishka to the third. As neither process is admissible, nothing remains to be done but to admit that the record is imperfect, and that it is only from external evidence that these dates can be fixed with anything like certainty.

Even admitting that Hushka and Jushka were the father and grandfather of Kanishka, which I am inclined to think may be the case, instead of his brothers, as is usually supposed, it will hardly help us much--four reigns of insignificant princes in 200 years is nearly equally inadmissible, and will not help us to fix Kanishka’s date from Asoka’s.

Recently the question has been very much narrowed by the discovery of a number of dated inscriptions at Muttra and elsewhere, in which the name of Kanishka and his successor Huvishka frequently occur--the latter always following, never preceding, the former name. It is this that makes me believe that the Hushka of the chronicle was the father of Kanishka, and nothing in that case is so probable as that his successor should take his grandfather’s name. It is almost impossible he should take his uncle’s, and as the name of Jushka appears nowhere in the inscriptions, it is natural to assume that he had passed away some time before they were written.

Be this as it may, the following table gives the inscriptions as they were found by General Cunningham:[736]--

_In the Indo-Pali Alphabet._

{ KANISHKA. Maharaja Kanishka. Samvat 9. { HUVISHKA. Maharaja Devaputra Huvishka. Samvat 39. { Maharaja Rajatiraja Devaputra Huvishka. Samvat 47. _Mathura._{ Maharaja Huvishka. Samvat 48. { VASUDEVA. Maharaja Rajatiraja Devaputra Vasu (deva). Samvat 44. { Maharaja Vasudeva. Samvat 83. { Maharaja Rajatiraja, Shahi, Vasudeva. Samvat 87. { Raja Vasudeva. Samvat 98.

_In the Bactrian-Pali Alphabet._

{Bahawalpur. Maharaja Rajadiraja Devaputra Kanishka. { Samvat 11, on the 28th of the (Greek) month of Dæsius. _Other {Manikyala Tope. Maharaja Kaneshka, Gushana vasa samvardhaka. localities._{ “Increaser of the dominion of the Gushans” (Kushans). Samvat 18. {Wardak Vase. Maharaja rajatiraja Huveshka. Samvat 51, 15th of Artemisius.

In addition to these Bactrian-Pali inscriptions, we have a record of a king called Moga (Moa?), on a copper plate from Taxila, wherein the Satrap Liako Kusuluko (Kozola?) speaks of the 78th year of the “great king, the great Moga,” on the 5th of the month of Panæmus.

In addition to the inscriptions bearing these names, General Cunningham quotes a great number of others, with dates in the same Samvat era, extending from the year 5 to the year 281, but without any kings’ names in them. Their purport, however, and the form of the characters used, he considers sufficient to show that they form a connected series dating from one and the same era, whatever that may be.

Here, therefore, we have an era, which we may safely assume was established by Kanishka, either from the beginning of his reign, or to mark some important event in it, and which was used after his time for two or three centuries at least. The question is, was that the era since known as that of Vicramaditya, dating from 56 B.C., or was it the Saka era of King Salivahana, dating 135 years after that? General Cunningham unhesitatingly adopts the former; and though it is not a subject to dogmatise upon, I am much more inclined to adopt the latter.

In the first place, because I can find no trace of any such era being in use before the cataclysm in A.D. 750. Bhau Daji states that he knows no inscription dated in it before the 11th century.[737] General Cunningham says it was not used as early as 826,[738] but, in another place, quotes an inscription in 754.[739] I know of none earlier; and can trace no allusion to any king of the name of Vicramaditya in the first century B.C., and no events that could have given rise to an era in 56 B.C. No trace of it is found in Thibet, in Burmah, or Cambodia, and it never was heard of in Ceylon or Java. In all these countries the Saka era is known and was used, and it seems strange that an era established by so powerful a Buddhist king as Kanishka should have endured for two or three centuries, and then perished, without leaving a trace in any Buddhist country, and then, after the 8th century, been revived and adopted by the Brahmans for their chronology. It may be so; but it is so strange, it seems to require some strong evidence to make it credible, and none such has yet been advanced.

Hitherto Kanishka’s date has been assumed almost wholly on numismatic evidence, but it seems to me without sufficient grounds. In all the lists hitherto published,[740] there are at least a dozen barbarian kings, several of whom, from the extent of their mintages, must have had long and prosperous reigns. To compress the whole into the sixty-four years that elapsed for the destruction of the Bactrian kingdom (120 B.C.), and the era of Vicramaditya (56 B.C.), seems to me a very strong measure, for which I can see no justification. To allow each, on an average, sixteen years’ reign, seems very much more probable, especially as many more names may yet be discovered--and even without them this would take us on to the Saka era (A.D. 79) without difficulty. One of them, Gondophares, as we shall presently see, reigned for twenty-six years at least.

The Roman consular coins found by M. Court, above referred to (_ante_, p. 79), were so worn as to be hardly legible, and though, therefore, they limit the antiquity of his reign certainly to this side of 44 B.C., they by no means prove that he was so early. On the contrary, the coins being worn, seems to prove that they were old before being buried; the probability is that they may have belonged to some pilgrim, or missionary, in the West, and had become sacred relics before they were enshrined. If Kanishka had merely wanted foreign coins, Greek or Roman, he might have had hundreds of perfect ones at his command. There must have been some other and holier motive for their deposit than merely to mark a date.

Every one has heard of the legend of St. Thomas the Apostle visiting the court of Gondophares, and, some add, being beheaded by his order. It may be a legend, and not one word of truth in it, but those who invented it in the second or third century must at least have had the means of knowing what was the name of the king who was on the throne of Gandhara at, or immediately after, the time of the Crucifixion. This name appears frequently on coins and inscriptions, and, from the numismatic evidence, has been placed by all as anterior to Kanishka, and I fancy that no one looking at the coins can well arrive at any other conclusion. If this is so, and he was reigning at any time between A.D. 33 and 50, Kanishka certainly belongs to the latter half of that century.

Against this it must be stated that both General Cunningham and Professor Dowson read an inscription of this king found at Takht-i-Bahi, as dated in his twenty-sixth year--one says in the 103rd,[741] the other 100th,[742] of the same Samvat as the inscription of Kanishka--a date which would answer perfectly for the legend. If this is so, there is an end of the controversy; but the stone is so worn, and the writing so indistinct, that I cannot see in the photographs of it what these gentlemen find there, and others are equally unable to do so; and besides this, it is such a wrench to all numismatic evidence to place the coins of Gondophares 100 years after those of Kanishka, that we must have more evidence than this imperfect inscription affords before we adopt its epochal date. The regnal date seems quite clear.

There is one other point of view from which this question may be regarded, but which it is difficult to express clearly without going to a greater length than our limits will admit of. It is the date of the third convocation, as the northern Buddhists call it--the fourth, according to the southern. It was held certainly under Kanishka’s auspices, and I cannot help fancying about the year 70 or 80 A.D. At that time, at least, Buddhism seems to have made a great stride in Thibet, in Burmah, and the East generally. It was about this time that it was fabled to have been first carried to Java, and about the time when it was first introduced in China.[743] It looks so like one of those outbursts of missionary zeal that followed all the three previous convocations, that I cannot help fancying that this one was held in the latter half of the first century, and that the era of the king who held it was allowed in all Buddhist countries to supersede that of the Nirvana, which, as far as I can see, was the only one that had existed previously in India.

To argue this out fully would require more space than its importance for architectural purposes would justify; but its bearing on the age of the Gandhara monasteries is in some respects considerable. If they are as modern as I suspect them to be, the more modern date for Kanishka would accord better with the known facts than carrying his date up before the Christian era.

* * * * *

Proceeding onward, the next name we come to of any importance is Mahiracula, who is said to have invaded Ceylon. There is, however, no trace of any such invasion at that time, which, by the application of averages would be about 180 A.D., if Kanishka ruled before, and 250 if after, the Christian Era. His date would be interesting if it could be ascertained from his connexion with Baladitya, the king of Magadha, whose story Hiouen Thsang tells in such minute detail.[744]

The Aditya dynasty opens with a king who is said to have been a kinsman of Vicramaditya, and is evidently the grandfather of the great king of that name, who figures prominently in the next dynasty as the patron of Matrigupta. The story of the latter is told in great detail in the ‘Raja Tarangini,’ and is one of the most curious episodes in the history. He was sent to Kashmir four years before the death of Vicramaditya (550), and on hearing of his patron’s decease, resigned his viceroyalty, and retired to Benares, leaving the throne to his successor, Pravarasena.

In speaking of the dynasty of Malwa, only twenty or twenty-five years were allowed for the reign of Sri Harsha, and only eighty for the whole duration, from the fall of the Guptas, 470, to the death of the great Vicramaditya, 550, a period, it seems from the evidence of the ‘Raja Tarangini,’ it is impossible to contract. Pratapaditya, the kinsman of the first, was, we are told, the great-grandfather of Megavahana, the first king of the next dynasty, and then we have one more king before we reach Hiranya, who is said to have been contemporary with the second Vicramaditya. Of course there may have been considerable overlapping at both ends, and the lives of the Kashmiri kings may have been short; but as we have six intermediate kings in the one list between the two Vicramadityas, and only one in the other, it seems that the last could hardly have ascended the throne before 515, if so early.

One of the acts of Pravarasena was to invade Siladitya, the first Ballabhi king of that name ruling in Gujerat. We have not, it is true, any dated coins or inscriptions belonging to him, but we have of his next successor but one, Sri Dharasena II., 593 (_ante_, p. 730), so that any date between 550 and 570 would answer perfectly well for this war, and the fact of its being so is in itself almost sufficient to establish the correctness of the chronology we are now trying to explain.

Since I wrote last on the subject, a passage has been pointed out to me[745] in Rémusat’s ‘Nouveaux Mélanges Asiatiques’ (vol. i. p. 197), which enables us to fix the chronology of the Naga dynasty within a year or two for extreme deviation. It seems that the third king, Chandrapira, applied to the Chinese Emperor for assistance against the Arabs in 713, and that the Emperor conferred the title of King on him in or about 720. As he was on the throne only eight years and eight months, there is no room for deviation in this date, and it carries with it those of his predecessors and followers. It thus becomes clear that Durlabha I. was the king who was on the throne when Hiouen Thsang resided in the valley, 631-633, and also when he passed near it on his return home in 643, all which is perfectly consonant with what we find in his text; and it also fixes the date of Lalitaditya, one of the most important kings in the list, with almost absolute certainty, as 725-762.[746]

Without placing implicit reliance on all that is said in the ‘Raja Tarangini,’ with regard to the exploits of this king, or of his having overrun and conquered all India, from beyond the Himalayas to Cape Comorin, still a sufficient residuum of fact must remain to enable us to see that the troubles which had begun in 650, on the death of Siladitya of Canouge, had laid India prostrate at the feet of any daring adventurer.

From whatever side we approach it, we can hardly fail to perceive that a great revolution took place in India about the year 750. All the old dynasties are then swept away, and for 200 years we have nothing but darkness, and when light again dawns, about two centuries afterwards, the map is re-arranged, and new dynasties and new religions have taken the place of the old.

This reign, too, forms a most appropriate termination to the principal division of our architectural history. The coins of his rival, Yasoverman of Canouge, found in the great Tope at Manikyala, prove the completion of that great Buddhist monument, just 1000 years after the style had been inaugurated by the great Asoka, and in that thousand years all that is important in Buddhist architecture is included. The fact, too, of his being the builder of the great Naga temple at Marttand, the earliest, so far as I know, in Kashmir, marks the commencement of a new architectural era, the fruits of which we see when the curtain again rises. The Jaina religion, with its new style of temples, had entirely replaced Buddhist forms over the greater part of India, and the Vaishnava and Saiva religions reigned supreme everywhere else, in the forms in which we now find them, after the lapse of nearly another 1000 years’ duration. As, however, there are no chronological difficulties with regard to these later dynasties, the discussion of the dates of the kings’ reigns who built them has evidently no place in this Appendix.[747]

ERA OF VICRAMADITYA.

Before concluding this Appendix, I would like to be allowed to explain an hypothesis which, if it can be sustained, not only clears up what has hitherto been a great mystery, but gets rid of a quantity of rubbish which obscures the chronology of the period. It does not, however, alter any date, nor affect them further than, if true, it confirms some, which, if it prove groundless, are deprived of its support.

No one has yet been able to point to the name of Vicramaditya as belonging to any king in the first century B.C., or to any event likely to give rise to an era being dated from it.[748] What, then, was the origin of the era dating from 56 B.C., and how did it arise and obtain its name?

My belief is that the solution of the mystery will be found in a passage in Albiruni, the meaning of which he did not profess to understand, combined with two or three passages in the ‘Raja Tarangini.’

The passage in Albiruni is to the following effect:--“L’ère de Saca, nommée par les Indiens Sacakala, est postérieure à celle de Vicramaditya de 135 ans. Saca est le nom d’un prince qui a régné sur les contrées situées entre l’Indus et la mer (le Golfe du Bengale). Sa résidence était placée au centre de l’Empire (Muttra?), dans la contrée nommée Aryavartha. Les Indiens le font naître dans une classe autre que celle des (Kchatrias?): quelques-uns prétendent qu’il était Soudra et originaire de la ville de Mansoura. Il y en a même qui disent qu’il n’était pas de race indienne, et qu’il tirait son origine des régions occidentales. Les peuples eurent beaucoup à souffrir de son despotisme, jusqu’à ce qu’il leur vînt du secours de l’Orient. Vicramaditya marcha contre lui, mit son armée en déroute, et le tua sur le territoire de Korour, situé entre Moultan et le Château de Louny. Cette époque devint célèbre, à cause de la joie que les peuples ressentirent de la mort de Saca, et on la choisit pour ère, principalement chez les astronomes.”[749]

It seems impossible to apply this narrative to any events happening in the first century B.C., not to mention the inherent absurdity of Vicramaditya establishing an era 56 B.C., and then 135 years afterwards defeating the Saka king on the banks of the Indus. If it meant anything, it might point to the origin of the Saka era, not that of Vicramaditya.

Turning from this to the ‘Raja Tarangini,’ we find the following passages in Troyer’s translation:--

“Ayant fait venir ensuite, d’un autre pays, Pratapaditya, parent du roi Vicramaditya, ils le sacrèrent souverain de l’Empire.

“D’autres induits en erreur ont écrit que ce Vicramaditya fut le même qui combattit les Çakas; mais cette version est rejetée.”[750]

A little further on we have: “Dans le même temps--the death of Hiranya--l’heureux Vicramaditya, appelé d’un autre nom Harcha, réunit comme empereur à Udjdjayini l’Empire de l’Inde sous un seul parasol....

“Employant la fortune comme moyen d’utilité, il fit fleurir les talents: c’est ainsi qu’encore aujourd’hui les hommes de talent se trouvent la tête haute au milieu des riches.

“Ayant d’abord détruit les Çakas, il rendit léger le fardeau de l’œuvre de Hari, qui doit descendre sur la terre pour exterminer les Mletchhas.”[751]

Before going further, it may be as well to point out what appears to be a fair inference from the above. That the first Vicramaditya, the friend of Pratapaditya, was so near in date to the second--he, in fact, appears to have been his grandfather--as to be confounded with him, and to have the name of Sakari applied to him, which in fact belonged to his grandson, the real destroyer of the Sakas.

My conviction is, that these paragraphs refer to one and the same event; and, assuming that the battle of Korûr was fought 544--the year before Vicramaditya sent Matrigupta to be his viceroy in Kashmir--what I believe happened was this: Some time after 750, when the Hindus were remodelling their history and their institutions, so as to mark their victory over the Buddhists, they determined on establishing two eras, which should be older than that of the Buddhists, A.D. 79, and for this purpose instituted one, ten cycles of sixty years each, before the battle of Korûr, and called it by the name of the hero of that battle, the most illustrious of their history; the other ten centuries, or 1000 years before the same date, and called it by the name of his father, Sri Harsha--a title he himself often bore in conjunction with his own name--the first consequently dated for 56 B.C., the second from 456. It need hardly be added that no Sri Harsha existed in the fifth century B.C., any more than a Vicramaditya in the first.

The co-existence of these eras may be gathered from the following passage in Albiruni:--

“On emploie ordinairement les ères de Sri Harscha, de Vicramaditya, de Saca, de Ballabha, et des Gouptas.” “D’après cela, en s’en tenant à l’an 400 de l’ère de Yezderdjed, on se trouve sous l’année 1488 de l’ère de Sri Harscha--l’an 1088 de l’ère de Vicramaditya--l’an 953 de l’ère de Saca--l’an 712 de l’ère de Ballabha, et de celle des Gouptas (A.D. 1032).”--‘Journal Asiatique,’ series iv. vol. iv. pp. 280, 286.

The Sri Harsha era, exactly 400 years before that of Vicramaditya, was avowedly conventional, and seems never to have come into use, and no further mention is made of it afterwards.

* * * * *

If this view of the matter can be sustained, the advantage will be not only that the date of the battle of Korûr, and of the expulsion of the Sakas, Hunas, Yavanas, &c., from India will be fixed with mathematical precision in 544, but that one of the greatest mysteries connected with the history of the period will be cleared up, and the revival of the Hindu religion relegated to a much later period. If, on the other hand, it can be shown that this view of the matter is not tenable, we shall lose these advantages, but it will require a great deal more than that to prove that Vicramaditya, or any Hindu king, reigned in the first century B.C. Buddhism was then in its palmiest state, and there is no trace of the Hindu religion then existing, and the expulsion of Sakas, Yavanas, and Hunas did not take place for long afterwards.

* * * * *

Be this as it may, having now cursorily run through the whole chronology, in so far as it admits of controversy, I feel very confident, on a calm review of the whole, that none of the important dates quoted above can be disturbed to a greater extent than say ten, or at the utmost twenty years--except, perhaps, that of Kanishka. From the Anjana epoch, 691 B.C., to the death of Lalitaditya, A.D. 761, all seems now tolerably clear and fixed, and, with a very little industry, minor blemishes might easily be swept away. If this were done, the chronology of mediæval India for the Buddhist period might be considered as fixed on a secure and immoveable basis of ascertained facts. The advantages of this being done can hardly be over-estimated for improving our knowledge of India generally, while, among other things, it would give a precision and solidity to all our speculations about that country, which, for want of it, have hitherto been generally so vague and unsatisfactory.

APPENDIX B.

The following are the last of the twenty-four Buddhas, beginning with Dipankara I., who appeared to instruct and enlighten mankind, and to whom Sakya Muni succeeds in the present Kalpa:

22. Kakusanda, born at Khémawatinagara. His Bo-tree the Sirisia (_Sirisa accasia_).

23. Kanagamma, born at Sobhawatinagara. His Bo-tree the Udambara (_Ficus glomerata_).

24. Kassyapa, born at Baranasi-nagara, Benares. His Bo-tree the Nigrodha (_Ficus Indica_).

Gautama, born 623 B.C., at Kapilawasta. His Bo-tree Pipphala (Ficus religiosa).[752]

APPENDIX C.

THE TWENTY-FOUR TIRTHANKARAS OF THE JAINS.

INDEX.

Abhayagiri dagoba, 192.

Abu, Mount, ancient Jaina temples on, 234. Temple of Vimala Sah, 235-237.

Adinah mosque, Gaur, 547.

Afghanistan, topes at, 72. Caves, 107.

Agra, 572. The Taje Mehal, 596-599. Akbar’s mosque, 602.

Ahmedabad, temple of Shet Huttising at, 257. Style and character of the architecture, 527.

Aiwulli, old temple at, 218. Plan, 219. View, 220.

Ajmir, temple at, 263. Mosque at, 510. Plan, 512. Great arch, 512.

Ajunta, rock-cut Tee at, 64. Chaitya cave, 122. View of interior, 123. Cross-section, 123. Plan, 124. View of façade, 125. Rock-cut dagoba, 126. Caves at, 145, 146. Viharas, 153-159.

Akbar, architectural glories of, 574-586.

Alexander the Great, pillars ascribed to, 56.

Allahabad, lât or pillar at, 53. Palace at, 583.

Altumsh, tomb of, 509.

Amara Deva, temple erected by, 69.

Ambêr, palace at, 480.

Amoy, pailoo at, 702.

Amravati, tope at, 71, 72. Rail at, 93, 99-101. Dagoba, 102.

Amritsur, golden temple at, 468.

Amwah, Jaina temple at, 250. View of porch, 251.

Ananda, temple at, Pagan, 615.

Andher, topes at, 65.

Andra dynasty, the, 20.

Anuradhapura, ancient capital of Ceylon, 188. The sacred Bo-tree, 189. Foundation and present state of the city, 188, 189. Topes, dagobas, &c., 189-195. Great Brazen Monastery, 195. Pillars, 196. The Maha vihara, 657.

Arch, objection of the Hindus to the, 210. Indian examples, 211. _See_ Gateways.

Architecture, Buddhist, 44. Stambhas, or lâts, 52-56. Stupas, 57-60. Topes, 60-83. Rails, 84-104. Chaitya halls, caves, 105-144. Vihara caves, 144-168. Gandhara monasteries, 169-184. Ceylon, 185-206.

Architecture, Chalukyan, 386. Temples 388-405.

Architecture, Civil: Dravidian, 380. Northern, or Indo-Aryan, 470-475.

Architecture, domestic, in China, 702-710.

Architecture, Dravidian, 319. Rock-cut temples, 326-339. Raths, 328-330. Kylas, 334. Temples, 340. Palaces, 381-385.

Architecture in the Himalayas, 279. Kashmiri temples, 283-318.

Architecture, Indian Saracenic: Ghazni, 494-500. Pathan, 498. Delhi, 500, 510-514. Later Pathan, 514-519. Jaunpore, 520-525. Gujerat, 526-539. Malwa, 540-544. Bengal, 545-551. Kalburgah, 553-556. Bijapur, 557-567. Scinde, 567, 568. Mogul, 569. Wooden, 608-610.

Architecture, Further Indian: Burmah, 611-620. Siam, 631-636. Java, 637-662. Cambodia, 663-684.

Architecture, Indo-Aryan, or Northern, 406. Temples, 411-436. Brahmanical rock-cut temples, 437-447. Temples, 448-464.

Architecture, Jaina, 207. Arches, 210-212. Domes and roofing, 212-218. Plans, 218-221. Sikras, 221-225. Northern: temples, 226-251. Towers, 252-254. Modern: Temples, 255-260. Caves, 261, 262. Converted mosques, 263. Southern Indian: colossal statues, 267, 268.

Aryans, their migration into India, and position among the Brahmans, 9-11. The dominant people before the rise of Buddhism, 48.

Asoka, Buddhist king, his connexion with Indian architecture, 47, 52. His missionaries into Ceylon, 199. His edicts at Girnar, 229. His missionaries into Burmah, 612, _see_ 61, 65.

Atala Musjid, the, 524.

Audience hall at Bijapur, 566.

Aurungabad, mosque at, 602.

Aurungzebe, 602. His copy of the Taje Mehal, 602. His burial-place, 603.

Ava, modern temple at, 659, _note_.

Avantipore, temples at, 291. Fragment of pillar at, 292.

Ayodhya, 631.

Ayuthia, ancient capital of Siam, ruins of pagoda at, 632, 633.

Babylonia, architectural synonyms in Burmah, 618. Ethnographical connexion, 630.

Badami, in Dharwar, Jaina cave, 261. Dravidian and Indo-Aryan temples at, 411. Contrast of style, 411. Caves at, 439-441. Plan and section, 444.

Bagh, cave at, 146. Great vihara, 159. Plan, 160.

Baillûr, in Mysore, great temple at, 393. Plan, 395. View of porch, 396. View of pavilion, 397.

Baion, Cambodia, temples at, 679-681.

Bakeng, Mount, ruined temple at, 682.

Bancorah, Hindu temple at, 14.

Bangkok, Great Tower, 634. Hall of audience, 635.

Barabar, Behar caves at, 108.

Baroach, mosque at, 537.

Barrolli, temple at, 449. View and plan, 450. Ornamented pillar, 451.

Bastian, Dr. Adolphe, Cambodian explorations of, 663.

Bayley, E. C., sculpture brought from Jamalgiri by, 169.

Bedsa, Chaitya cave, 112. Plan and capital of pillar, 113. View on verandah, 114.

Behar caves, 108. Bengal, 138-144.

Benares, view and diagram of temples at, 412, 460. View of balcony at the observatory, 481.

Bengal, caves, 138. Its architecture and local individuality of its style, 491, 545. Type of the modern roof, 546.

Bettu temples, 267.

Bhaja, Chaitya cave, plan, 110. Façade, 111.

Bhaniyar, near Naoshera, temple at, 292. View of court, 293.

Bharhut, rail at, 85-91. Square and oblong cells, from a bas-relief at, 135. Round temple and part of palace, 168.

Bhatgaon, Devi Bhowani temple at, 304. Doorway of Durbar, 307.

Bhilsa Topes, 60-65.

Bhojpur, Topes at, 65.

Bhuvaneswar, great temple at, 420; plan, 421; view of, 422. Great Tower, 423. Raj Rani temple at, 424; doorway in, 425.

Bijanagur, gateway, 211.

Bijapur, 557; its architecture, 558. Jumma Musjid at, 559. Sections, 560. Tomb of Ibrahim, 561. Of Mahmûd, 562. Ancient Hall, 566.

Bimeran, Tope at, 78.

Bindrabun, 462. Plan of temple at, 463. View, 464. Balcony in temple, 465.

Bintonne, relic of Buddha at, 58.

Bombay, number of caves at, 107.

Boondi, palace at, 476.

Boro Buddor, Java, 643. Plan, elevation, and section, 645. Sections of domes, 646. View of central entrance and stairs, 649.

Bo-tree, the sacred, 189. Branch of it in Ceylon, 199. At Buddh Gaya, 656.

Buddha La Monastery, Thibet, 312.

Bowlees or Reservoirs, use and architectural features of, 486.

Brahma, numerous images of, in Cambodia, 680.

Brahmanism, 323.

Brambanam, Java, group of temples at, 651.

Brazen Monastery, Anuradhapura, 195.

Buchropully, 388. View of temple, 389.

Buddh Gaya, stupa, 69, 70. Temple, 70. Rail, 85. Bas-relief from, 111. The Sacred Tree, 199.

Buddha, period of his birth, 14. Apportionment of his remains, 57-59. Relic of, at Bintenne, 58. Colossal statue of, 200, _note_. His tooth, its sanctity, shrines, migrations, 58, 59, 161. Relics of, at Rangûn, 622.

Buddhism, its founder, 15. Secret of his success, 16.

Buddhist architecture, earliest traceable date, 48-50. Religion dominated by it, 49. Classification, 50. Temple in China, 691. Monastery at Pekin, 693. _See_ Architecture.

Bunds, or Dams, 486, 487.

Buribun, sculptures at, 682, _note_.

Burmah, architecture in, 611. Thatún, 612. Prome, 613. Pagan, 614. Circular dagobas, 619-626. Monasteries, 626-630. Non-use of mortar, 660.

Butwa, tomb at, 536.

Cabul, topes near, 72.

Cambay, Jumma Musjid at, 537.

Cambodia, M. Mouhot’s researches in, 663. Labours of Dr. Bastian, 663; of Mr. Thomson, 664; of Captains Doudart de la Grée and Delaporte, 664. Traditions, original immigrants, history, 665, 666. Temple of Nakhon Wat, 666. Temple of Baion, 679; of Ongcor Thom, 680; other temples, 681. Civil architecture, 682. Remarkable evidences of mechanical skill and civilization, 684.

Canara, stambhas at, 263.

Canouge, Jaina temple at, 263. Mosque at, 525.

Canton, pailoo near, 700.

Canton river, “Second Bar Pagoda” on the, 696.

Capitals and columns, Tirhoot and Sankissa, 54. Jamalgiri, 173, 176.

Caves, 106. Geographical distribution of, 107. Ajunta, 122, 127, 153. Badami, 439. Bagh, 146, 159. Barabar, 108. Bedsa, 112. Behar, 108. Bengal, 138-144. Bhaja, 110. Dhumnar, 131. Ellora, 127. Junir, 166. Karli, 116. Kenheri, 129. Kholvi, 132. Salsette, 161. Satapanni, 108. Mode of ornamentation, 133.

Cenotaphs, 470-475.

Ceylon: Buddhist relics in, 58. Its ancient architecture, 185-206. _See_ Anuradhapura.

Chaitya Halls: Buddhist temples, 105. Examples, 109-132. _See_ Caves.

Chalukyans, the, 386. Early identity of the Jains with the, 387. Peculiarity of their style, 387. _See_ Architecture.

Chandragupta, the Sandrocottus of the Greeks, 17.

Chandravati, pillars at, 238. Temple or porch, 448. Plan, 449.

Cheras, territory occupied by the, 322.

Chillambaram, temple at, 350. Plan, 351. Porch of hall, 353. Section of porch, 353. View of ruined temple, or pagoda, 354.

China, deficiency of information, 685. Point of divergence between its people and the Egyptians, 686. Causes of the absence of certain classes of buildings, 686-688. Pagodas: Temple of the Great Dragon, 689. Buddhist temples, 691. Taas, or towers, 695. Tombs, 698. Pailoos, 700. Domestic architecture, 702-710.

Chittore, diagram of Meera Baie’s temple at, 458. Temple of Vriji, 459. Palace of Bhîm and Pudmandi, 476.

Chola, region forming the kingdom of, 322.

Choultrie, Tirumulla Nayak’s, 361.

Chuttries, or Cenotaphs, 470.

Civil Architecture, 470. Cambodia, 682. China, 702.

Cole, Lieut., explorations in Kashmir, 280.

Colombo, Hindu temple at, 332.

Combaconum, 367. Gopura or gate pyramid at, 368.

Confucius, temples of, 689.

Conjeveram, temples and hall, 369.

Court, M., Topes opened by, 79.

Cowie, Rev. A., Kashmirian explorations of, 280.

Crystal Palace, Sydenham, Gill’s copies of Indian frescoes, 158, _note_. Sculptures, 169.

Cunningham, Gen., Eastern Archæological explorations of, 54, _et seqq._

Cuttack, caves at, 140, 143. Nine-storeyed palace, 433. Hindu bridge at, 434.

Dagobas: Amravati, 102. Rock-cut, at Ajunta, 126. Anuradhapura, 190-196. Circular, of Burmah, 619. Kong Madú, 619. Shoëmadu, 621. Rangûn, 623. Mengûn, 624-626.

Dams, or bunds, 486-487.

Darunta, Jelalabad, topes at, 77.

Das Avatar, Buddhist vihara, Ellora, 165.

Dasyus, the slave people, 12. Their architecture, 13.

Deeg, garden palace of, 481. Hall, 482. View from the Central Pavilion, 483.

Deepdans, or lamp pillars in the East, 336, 337.

Dehrwarra vihara, Ellora, plan of, 163.

Delai Lama, worship paid to him, 312.

Delaporte, Captain, explorations in Cambodia, 664.

Delhi, Lât at, 52. Jaina Temple, 259. Palace, 591. Plan, 592. Jumma Musjid, 601.

Delhi, Old, section, arches, and minar of the Kutub, 503-509. Iron pillar, 507. Interior of tomb, 509. View of tomb, 516. Pendentive from mosque, 519.

Deriah Doulut, the pavilion of, at Seringapatam, 604.

Dhar, mosque at, 540.

Dharwar, deepdans in, 337. Brahmanical rock-cut temples, 437.

Dhumnar, caves at, 131, 162.

Dhumnar Lena, Ellora, Rock-cut temple at, 445. Plan, 446.

Diggu Hublum, unfinished Mantapa at, 378.

Dimapur, monoliths at, 309.

Djeing Plateau, Java, group of small temples at, 659.

Do Tal, or Dookya Ghur, Ellora, a Buddhist vihara, 165.

Dolka, mosque at, 537.

Domes: Hindu, 212. Indian Saracenic, 560. Constructive diagram, 565; Boro Buddor, 616.

Domestic architecture, Chinese, 702.

Doorways: Nepal, 305. Hammoncondah, 390. Bhuvaneswar, 425.

Doudart de la Grée, Captain, explorations in Cambodia, 664.

Dravidian style, 319. Its extent, 319. Historical notice, 320. Religious, 323. _See_ Architecture.

Dravidians, the, 11.

Durbar, Bhatgaon, doorway of, 307.

Duttiah, palace at, 477.

Elaala, so-called tomb of, 189.

Ellora, caves at, 127. Viharas, 163. Indra Subha Cave, 262. Kylas at, 334-337. Dhumnar Lena Cave, 445.

Erun, lâts at, 55. Pillar, 317.

Façades: Behar, 109, 111. Burmah, 627.

Feroze Shah, lât rebuilt by, 52.

Futtehpore Sikri, 578. Carved pillars in, 579. Mosque at, 580. Southern gateway, 581.

Gal Vihara sculptures, 200. Ganesa Cave, Cuttack, 140. Pillar in, 140.

Gandhara Topes, 72-76. Monasteries, 169-184.

Ganges, the, and its ghâts, 484.

Gate-pyramid at Combaconum, 368.

Gateways or Torans, 95. Sanchi, 96. Bijanagur, 211. Jaunpore, 522. Gaur, 550. Futtehpore Sikri, 581. Pekin, 693.

Gaudapalen, temple at, Pagan, 617.

Gaur, peculiar form of roof in, 545, 546. Mosques, 547; their defects, 549. Ancient Minar, 550. Gateways, 550.

Gautamiputra Cave, Nassick, rail at, 94. Pillar in, 150.

Ghâts, or landing-places, 484. Ghoosla, Benares, 485.

Ghazni, buildings of Mahmúd and his nobles, 494. Minar at, 495. Ornaments from the tomb of Mahmúd at, 496.

Ghoosla Ghât, the, Benares, 485.

Gill, Major, Oriental drawings by, 158, _note_.

Girnar, the Hill of, shrine of the Jains, 228. Temple of Neminatha, 230.

Gopal Gunge, temple at, 467.

Gopura at Combaconum, 368.

Gualior, temple at, 244. Teli ka Mandir temple, 452. View, 453. Temple of Scindiah’s mother, 461. View, 462. Palace, 479. Tomb of Mahommad Ghaus, 576. View, 577.

Gujerat, 526. Historical account, 526, 527.

Gurusankerry, pavilion at, 274. Stambha, 276.

Gyraspore, temple at, 249.

Hammoncondah, Metropolitan temple of, 389. View of great doorway, 390.

Himalayas, the, architecture in, 279.

Hindu temple at Bancorah, 14

Hiouen Thsang at Amravati, 103; at Assam, 310.

Honan, China, Buddhist temple at, 691.

Hullabîd, in Mysore, temple at, 397. The Kait Iswara, 398. Plan, 399. Restored view of the temple, 400. Its varied design, 401. View of central pavilion, 402. Succession of animal friezes, 403.

Humayun Shah, tomb of, at Old Delhi, 575.

Ibrahim Shah, Mosque of, at Bijapur, 559.

Imambara, the, at Lucknow, 605.

Immigrations, 25.

India, Northern, inducements to the study of its architecture, 4. Its history, 6-29.

India, Southern, unsatisfactory records, 29. Sculptures, 32. Mythology, 35. Statistics, 42.

India, Western, its architecture, 437-447.

India, Central and Northern, 448.

India, Further, 611-684.

Indian Saracenic style, 489. Divisions of styles and their boundaries, 491-493. _See_ Architecture.

Indo-Aryan or Northern style, 406. Reasons for the term, 406. _See_ Architecture.

Iron pillar at Kutub, 507.

Jaina Architecture, 207. Identical with Buddhist, 207. Region dominated by its style, 208. _See_ Architecture.

Jajepur on the Byturni, pillar at, 432.

Jamalgiri, plan of monastery at, 171. Corinthian capitals from, 173.

Jarasandha Ka Baithak tope, 68.

Jaunpore, style adopted at, 520. Plan and view of the Jumma Musjid, 522. The Lall Durwaza Mosque, 523. The Atala Musjid, 524. Tombs and shrines, 525.

Java, 637. Its history, 640. Boro Buddor, 643-650. Mendoet, 650. Brambanam, 651. Tree and Serpent temples, 653-659. Djeing plateau, 659. Suku, 660.

Jehangir, desecration of his tomb, 587.

Jelalabad topes, 77, 79.

Jinjûwarra, gateway, 211.

Juganât, temple of, 430. Tower, 431.

Jumma Musjid, Jaunpore, 521. Section and view, 522. Ahmedabad, 527. Plan and elevation, 528. Malwa, 541. Plan, 542. Courtyard, 543.

Junaghur, tomb of the late Nawab of, 606.

Junir, caves at, 167. Plan and section of circular, 167.

Kait Iswara, temple at Hullabîd, 397. View, 398.

Kakusanda, one of the Buddhas, discovery of a relic of, 622.

Kalburgah, 552. Mosque, 553. Plan, 554. Half elevation and view, 555.

Kallian, in Bombay harbour, Ambernath temple, 457.

Kanaruc, Orissa, Black Pagoda at, 221. Restored elevation, 222. Diagram, plan and section, 223. History, 426.

Kangra, 314. _See_ Kote.

Kantonuggur, 465. View, 467.

Kanwa dynasty, 19.

Kârkala, colossal statue at, 268.

Karli, cave at, 55, 116. Section and plan, 117. View of exterior, 118. View of interior, 120. Lion-pillar, 121.

Karna Chopar Cave, 108.

Kasachiel, temple of Bouddhama at, 302.

Kashmir, its architecture, 279. Writers thereon, 280. Peculiar form of temples and pillars, 283. Starting-point of its architectural history, 285. Temple of Marttand, 285-291. Other examples, 292-297. The ‘Raja Tarangini,’ or native history, 297.

Kasyapa, one of the Buddhas, discovery of a relic of, 622.

Kenheri Cave, the Great, near Bombay, 129. View of rail in front, 130.

Keseriah, Tirhoot, capital of, lât at, 71.

Khajurâho, temples at, 245-248, 452. Kandarya Mahadeo, temple at, 454. View, 455. Plan, 456.

Kholvi, caves at, 132, 162.

Kioums, Burmese, 628.

Kiragrama, 314. _See_ Kote Kangra.

Kirti Stambha at Worangul, 392.

Konagamma, one of the Buddhas, relic of, 622.

Kondooty, near Bombay, chaitya cave, 108, _note_.

Kong Madú Dagoba, details of the, 619. View, 620.

Kosthakar, or Nepalese temple, 303.

Kote Kangra, temples, 313. View of temple at Kiragrama, near, 314.

Kûmûlûlû, rock-cut temple at, 339.

Kutub, the, Old Delhi, 503. Section of colonnade at, 503. Central range of arches, 504. Minar, 505, 506. Iron pillar at, 507.

Kylas at Ellora, 334-337. Pillar in, 443.

Lahore, Jehangir’s buildings at, 587.

Lall Durwaza Mosque, Jaunpore, 523.

Lassa, monastery of Bouddha La at, 312.

Lâts, or Buddhist inscription-pillars, 52. Examples, 53, 54.

Lomas Rishi, Behar cave, 108. Façade and plan, 109.

Lucknow, the Imambara at, 605.

Macao, temple at, 694.

Mackenzie, Col., Indian researches and drawings by, 638.

Madras, temple on the hill of Tripetty at, 378, _note_. Prevailing style in the presidency of, 385.

Mádura, Perumal pagoda at, 331. Plan of Tirumulla Nayak’s choultrie, 361. Pillar in, 361. View of the hall, 363. Great temple, 364. The Jumbúkeswara temple, 365.

Maha vihara, the, Anuradhapura, 657.

Mahavellipore, raths of, 134, 175, 326, 330. Pavilion at, 274. Tiger cave at Saluvan Kuppan, 333.

Mahawanso, or Buddhist history of Ceylon, accounts of Oriental structures in the, 58, 185, 189, 195, 196, 612.

Maheswar, ghât at, 485. Mahmúd Begurra, tomb of, near Kaira, 538.

Mahmúd of Ghazni, temple of Somnath destroyed by, 494.

Mahomedanism, migration into, and dealings with the architecture of India, 380, 526, 527.

Malwa, 540. _See_ Mandú.

Mandalé, monastery at, 629.

Mandú, capital of Malwa, 540. The Jumma Musjid, 541. Palace, 543.

Manikyala topes, 79-83. Relic casket, 80.

Marttand, temple of, 285. Plan, 286. View, 287. Central cell of court, 288. Date, 289. Niche with Naga figure, 290. Soffit of arch, 291.

Masson, Mr., exploration of the Jelalabad topes by, 77-79.

Matjanpontih, serpent-temple at, 659.

Maurya dynasty, 17.

Mechanical skill of the Cambodians, 684.

Mehturi Mehal, “the Gate of the Sweeper,” 567.

Mendoet, Java, temple at, 650.

Mengûn, circular pagoda at, 624. View, 625.

Michie, Mr. A., information derived from, 689, _note_.

Milkmaid’s Cave, Behar, 109.

Minars and minarets: Surkh and Chakri, Cabul, 56. Ghazni, 495. Kutub, 505. Gaur, 550.

Mirzapore, Queen’s mosque at, 529.

Moggalana, relic casket of, 62.

Mogul architecture, 569. Originality of the buildings, 569. Works of Shere Shah, 572. Akbar, 574-586. Jehangir, 587-589. Shah Jehan, 589. Aurungzebe, 602-604. Oude and Mysore, 604-607.

Mohammad Ghaus, tomb of, at Gualior, 576. View, 577.

Monasteries, or viharas, 133. Gandhara, 169. Burmese, 626-630. Thibetan, 312. Pekin, 693.

Monoliths at Dimapur, 309.

Moodbidri, Jaina temple at, 271, 272. Pillar, 273. Tomb of priests, 275.

Mooháfiz Khan, mosque of, 532.

Mortar, non-users of, 660.

Mosques: Adinah, 549. Agra, 596. Ahmedabad, 527. Ajmir, 511. Baroach, 537. Bijapur, 559. Cambay, 537. Canouge, 525. Delhi, 601. Dhar, 540. Dolka, 537. Futtehpore, 581. Gaur, 547. Kala Musjid, 518. Kalburgah, 553. Kashmir, 609. Kutub Minar (Old Delhi), 501. Jaunpore, 521. Mandu, 543. Mirzapore, 529. Mooháfiz Khan, 532. Sirkej, 531.

Mosques, converted, 263, 264.

Mouhot, M., researches in Cambodia, 663.

Muktagiri, Jaina temples at, 240.

Mukteswara, Orissa, temple of, 419.

Mûlot in the Salt Range, temple at, 296.

Mûti Musjid, the, or pearl mosque, 599. View in courtyard, 600.

Muttra, rail at, 91.

Mysore, 392, 604.

Mythology of the Hindus, 35-42.

Nagas, or Snake worshippers, 10. Head-quarters, 297. _See_ Kashmir, Cambodia, Nakhon Wat, Ongcor Thom.

Nahapana vihara, Nassick, 149. Pillar in, 150.

Nakhon Thom (the Great City), Cambodia, 666.

Nakhon Wat (Cambodian temple), its grand dimensions, 667. Plan, 668. Uncemented masonry, 669. Elevation, 670. Corridors, 671, 672. General view, 675. Pillars, 676, 677. Multiplicity of snake-sculptures, 677.

Nankau Pass, archway in, near the Great Wall, China, 709.

Nankin, porcelain tower at, 695.

Nassick, chaitya cave at, 115.

Nat-mandir, or dancing-hall at Bhuvaneswar, 423.

Neminatha, Girnar, temple of, 230.

Nepal, 298. Its architecture, 299. History, 300-302. Stupas or chaityas, 302. Kosthakar, 303. Temples, 304-318.

Nigope Behar cave, 108.

Ongcor Thom, temple of, 679. Tower and sculptured masks, and tradition relating thereto, 680. Palaces and public buildings, 682.

Orissa, fragment of a column from a temple in, 317. History, 415. Architecture, 417. Earliest authentic building, 417. Temples, 418-420.

Ornament, honeysuckle, at Allahabad, 53. From the tomb of Mahmúd at Ghazni, 496.

Oudeypore, cenotaph of Singram Sing, 471. In Maha Sâti at, 473.

Ourtcha, Bundelcund, palace at, 478.

Pagan, ruins of, 614. Ananda temple, 615. Thapinya, plan, 615. Section, 616. Temple of Gaudapalen and Sem Byo Koo, 617.

Pagodas, Hindu, 221, 344. Burmese, 619-626. Siamese, 632. Chinese, 697.

Pailoos, or “Triumphal Gateways” of the Chinese, 700. Near Canton, 701. At Amoy, 702.

Palaces, 475. Allahabad, 583. Ambêr, 480. Chittore, 476. Deeg, 481. Delhi, 591. Duttiah, 477. Gualior, 479. Ourtcha, 478. Pekin, 705.

Palitana, the Sacred Hill of Sutrunjya, near, 227.

Panataram, three-storeyed temple at, 654. View, 655. Serpent temple, 658.

Pandrethan, temple at, 294.

Pandyas, the, 321.

Parasurameswara, Orissa, temple of, 418.

Parisnath, 239.

Patan, temple of Mahadeo and Krishna, 306.

Paten ta Prohm, Cambodia, character of the buildings of, 667. Temple, 681.

Pathan style, 498-513. Later Pathan, 514-519.

Pathans, the, 498. Historical summary, 498. Their architectural glories and career, 499. Examples, 503-519. _See_ Delhi.

Pavilions: Ceylon, 167. Gurusankerry, 274. Baillûr, 397. Vijayanagar, 385. Hullabîd, 403. Sirkej, 532. Pekin, 705.

Payech, Kashmiri temple at, 294. View, 295.

Pegu, Shoëmadu pagoda at, 620.

Pekin, temple of Confucius, 689. Temple of the Great Dragon, 689. Gateway of monastery, 698. Tombs, 700. Summer palace, 705. Winter palace, 707.

Pemiongchi, Nepal, porch of temple at, 313. View, 314.

Pendants to domes, 216. At Vimala Sah, 237.

Pendentive from mosque at Old Delhi, 519. Bijapur, 564.

Peroor, near Coimbatore, date of porch, 370. Compound pillar at, 372.

Perumal pagoda, Mádura, 331.

Pillars: Ajunta, 156. Amravati, 101. Avantipore, 293. Barrolli, 451. Ceylon, 196. Chandravati, 238. Cuttack, 140. Delhi, 507. Ellora, 443. Erun, 317. Futtehpore Sikri, 579. Gautamiputra, 150. Jajepur, 432. Kashmir, 283. Mádura, 361. Moodbidri, 273. Nahapana, 150. Peroor, 372. Srinagar, 284. Vellore, 372. Yadnya Sri, 152.

Pittadkul, plan of temple at, 221. Temple of Papanatha at, 437. View, 438.

Pollonarua, Ceylon, 199. Extent and epoch of its temples, 200. Examples, 201-203.

Poonah, Saiva temple near, 447.

Porches: Anwah, 251. Chillambaram, 351. Delhi, 259. Jaina, 216.

Prome, early capital of Burmah, 613.

Provincial building, Gujerat, 537-539.

Puri, 428. Plan of Juganât, temple at, 430. View of tower, 431.

Purudkul, or Pittadkul, great temple of, 338.

Queen’s mosque, Mirzapore, 529.

Raffles, Sir Stamford, 638.

Rails: Amravati, 93. Bharhut, 86. Buddh Gaya, 85. Dhumnar, 131. Gautamiputra, 94. Kenheri, 130. Kholvi, 132. Muttra, 91. Sanchi, 92.

‘Raja Tarangini,’ the, or native History of Kashmir, 289, 297.

Rajputana, bund of, 486.

Rajsamundra, bund of Lake, 487.

Ramisseram, great temple at, 355. Plan, 356. Its dimensions, 357. Corridors, 358. View of central corridor, 358.

Rangûn, the Shoëdagong pagoda at, 622. View, 623.

Rani Gumpha cave, the, 140.

Rath at Mahavellipore, 134, 175, 326, 328.

Relic worship, Buddhist, origin of, 57. Distribution and depositaries of the relics, 58, 59, 66, 189, 195. Discoveries of, 622.

Reservoirs, or bowlees, scope for architectural display in, 486.

Roads and bridges of the Cambodians, 683.

Rock-cut temples, 437-447.

Roofing, diagrams, 213-215. Modern curved style, 546. Chinese, 703.

Ruanwelli dagoba, Anuradhapura, 190, 191.

Sadri, Khumbo Rana’s temple at, 240. View, 241. External view, beauty of details, &c., 242.

Saftar Jung, tomb of, near the Kutub, 604.

Sakya Muni, founder of Buddhism, 15. His early life and subsequent self-mortification, 15. Result of his appeal to his countrymen, 16.

Salsette, Durbar cave at, 147. Kenheri caves, 161.

Saluvan Kuppan Tiger Cave, 333.

Sanchi, great tope, 61, 63. View, plan, section, and details, 63. Rails at, 92, 93. Gateways, 95-97. Small tope, 98. Torans, 99. Chaitya hall, 105.

Sankissa, capital of a lât at, 54.

Sariputra, relic-casket of, 62.

Sarnath, tope at, 65-68. Vihara, 173.

Satapanni cave, 108.

Satdhara topes, 64.

Sat Ghurba cave, 108.

Scinde, tombs in, 567.

Sculptures, 32-35. In the Gandhara monasteries, 176, 177.

Secundra, Akbar’s tomb at, 583. Plan, 584. Diagram section, 585. View, 586.

Seringham, pillared hall at, 347. View of temple, 349.

Serpent temples, 653.

Serpent-worship, 266.

Shah Dehri, plan of Ionic monastery at, 176. Ionic pillar, 176.

Shah Hamadan, mosque of, Srinugger, 608.

Shah Jehan, 589. Palace at Delhi, 591. Taje Mehal, 595. The Mûti Musjid, 599.

Shepree, near Gualior, Pathan tomb at, 515.

Shere Shah, works of, 572. Tomb, 573.

Shoëdagong Pagoda at Rangûn, 622.

Shoëmadu, Pegu, the Great Pagoda at, 620. View and plan, 621.

Siam, early and present capitals, 631. Ayuthia, 632. Bangkok, 634.

Sikras, or Vimanas, 221-225.

Sirkej, tombs and mosque at, 531. Pavilion, 532.

Sisunaga dynasty, 14.

Siva, serpent of, 41, _note_.

Snake sculptures, 676, 677.

Somnath, Girnar, temple, 232.

Somnathpûr in Mysore, temple at, 393. View, 394.

Sonaghur, Bundelcund, Jaina temple at, 256.

Sonari topes, 64.

Soubramanya, temple at Tanjore, 345.

Sravana Belgula, colossal statue at, 267. Bastis, 269. View, 270.

Sri Allat, tower of, at Chittore, 251. View, 252.

Srinagar, Kashmir, pillar at, 284.

Srinugger, Jumma Musjid at, 608. Shah Hamadan Mosque, 608. View, 609.

Stambhas, 52. At Gurusankerry, 276. They illustrate the rise and progress of Indian architecture, 277. _See_ Lâts.

Statues: Seperawa, 200. Sravana Belgula, 267. Kârkala, 268. Yannûr, 268.

St. Stephen’s, Walbrook, resemblance to Hindu plans, 218.

Stupas, or Topes, 57. _See_ Topes.

Stupas, or Chaityas, Nepal, 302.

Sudama, or Nigope Cave, 108.

Suku, Java, group of temples, 660. Their likeness to contemporary edifices in Yucatan and Mexico, 661.

Sultangunge, near Monghyr, vihara at, 137.

Sultanpore, tope at, 78. Small model found in the tope, 126.

Sunga dynasty, 19.

Surkh Minar, Cabul, 56.

Swayambunath, Nepal, temple, 302.

Taas of the Chinese, 695.

Taje Mehal, the, 595. View, 596. Plan and section, 597. Details, inlayings of precious stones, &c., 598.

Takht-i-Bahi, plan of monastery at, 171.

Takt-i-Suleiman, Kashmir, Hindu temple at, 282.

Tanjore, diagram plan of pagoda at, 343. View of Great Pagoda, 344. Temple of Soubramanya, 345.

Tarputry, temples at, 375. Views of gopura, 376, 377.

Tassiding, doorway of Nepalese temple at, 313.

Tatta, tomb of Nawab Amir Khan near, 568.

Teen Tal, a Buddhist vihara, at Ellora, 165.

Tees in rock-cut temples, 64. At Ajunta, 64.

Tejpala and Vastupala, triple temple at, Girnar, 232.

Temples: Abu, 234. Ahmedabad, 257. Aiwulli, 218. Ajmir, 263. Amritsur, 468. Amwah, 250. Avantipore, 292. Badami, 411. Baillûr, 393. Bakeng (Mount), 682. Bancorah, 14. Barrolli, 449. Benares, 412, 459. Bhanghur, 250. Bhaniyar, 292. Bharput, 168. Bhatgaon, 304. Bhuvaneswar, 418. Bindrabun, 464. Boro Buddor, 643. Brambanam, 651. Buchropully, 389. Cambodia, 666. Canouge, 263. Chandravati, 448. Chillambaram, 350. Chinese, 689, 694. Chittore, 459. Colombo, 332. Combaconum, 367. Delhi, 259. Djeing Plateau, 659. Gaudapalen, 617. Girnar, 230. Gualior, 244, 453, 462. Gyraspore, 249. Hammoncondah, 390. Hullabîd, 397. Java, 650. Kanaruc, 426. Kantonuggur, 467. Khajurâho, 245, 455. Kiragrama, 316. Mádura, 359. Marttand, 285. Mendoet, 650. Moodbidri, 271. Mûlot, 297. Nepal, 302. Pagan, 615. Pandrethan, 294. Patan, 306. Payech, 295. Pemiongchi, 314. Pittadkul, 221, 438. Poonah, 446. Puri (Juganât), 431. Ramisseram, 355. Sadri, 240. Seringham, 347. Sonaghur, 256. Somnathpûr, 394. Sravana Belgula, 270. Suku, 660. Tanjore, 344. Tassiding, 313. Tinnevelly, 366. Tiruvalur, 346. Udaipur, 457. Vellore, 371. Vijayanagar, 375.

Tennent, Sir Emerson, works on Ceylon by, 185, 200.

Thapinya, temple of, at Pagan, 615. Section, 616.

Thatún, pagoda at, 613.

Thibet, exclusion of travellers, number and character of its monasteries, 311. The Delai Lama, and the worship paid to him, 312. _See_ Nepal.

Thomson, Mr. J., his photographs of the Great Temple of Nakhon Wat, 671, 672, 675-677.

Thuparamaya Tope, Buddhist relic-shrine, 192.

Tiger-cave at Cuttack, 143. At Saluvan Kuppan, 333.

Tinnevelly, temple at, 366. Dimensions, details, &c., 367.

Tirhoot, lâts, or inscribed pillars at, 53. Capital, 54.

Tirthankars, Jaina Saints, 208, 331.

Tirumulla Nayak’s choultrie, 361. Dimensions, cost, and ornamentation, 362. View, 363.

Tombs: Bijapur, 561. Butwa, 536. Chinese, 698. Delhi (Old), 509, 516. Gualior, 577. Gujerat, 534. Lucknow, 606. Moodbidri, 275. Secundra, 584. Shepree, near Gualior, 515. Sirkej, 531. Tatta, 568.

Tooth of Buddha, its sanctity, shrines, migrations, &c., 58, 59, 161.

Topes or stupas of the Buddhists, their form and purpose, 58. Bhilsa group, 61. Example at Sanchi, 63. Invariable accompaniments to these structures, 64. Sarnath and Behar, 66-68. The Jarasandha Ka Baithak, 68, 69. Buddh Gaya, 69, 70. Amravati, 71, 72. Gandhara, 72-76. Jelalabad group, 77. Bimeran, 78. Sultanpore, 78. Manikyala, 79-83.

Torans, 95. _See_ Gateways.

Towers: Bangkok, 653. Chittore, 253. Nankin, 695. Ongcor Thom, 680.

Tree and Serpent temples, 653.

Tree-worship, 266.

Trisul emblem at Amravati, 104.

Tung Chow pagoda, 697.

Udaipur, temple at, 456. View, 457.

Udayagiri, Cuttack, caves at, 138.

Ujjain dynasty, 22.

Ulwar, tomb of Rajah Baktawar at, 474.

Umbrella ornaments on topes, 64, 80, 125, 126.

Vellore, near Coimbatore, 370. View of portico of temple, 371. Compound pillar, 372.

Ventura, General, topes opened by, 79, 81.

Victory, Tower of, at Chittore, 253.

Vigne’s travels in Kashmir, 280.

Viharas, or monasteries, 133. Diagram, 134. Nalanda, 136. Sultangunge, 137. Sarnath, 137.

Viharas, 144-147. Nassick, 147-151. Ajunta, 153-159. Bagh, 159, 160. Ellora, 163.

Vijayanagar, 373. Destroyed by the Mahomedans, 374. View of temple of Vitoba at, 375. Garden pavilion, 384. Palace, 385.

Vimala Sah, temple of, 235.

Vimanas, or Sikras, 221-225.

Vishveshwar temple, Benares, 459. View, 460.

Viswakarma Cave, Ellora, 128.

Vitoba, temple of, at Vijayanagar, 375.

Wall, Great, in China, 708.

Waniyat, Kashmir, group of temples at, 293.

Well-holes in temples, 654, 656.

Window at Ahmedabad, 533.

Wooden architecture, Kashmir, 608. Mosques at Srinugger, 608.

Wooden temples, similarity to the wooden architecture of Sweden and Norway, 308.

Worangul, Kirti Stambhas in, 391. View, 392.

Yadnya Sri cave, Nassick, 151. Pillar in, 152.

Yannûr, colossal statue at, 268.

Zein-ul-ab-ud-dín, tomb of, 281.

FOOTNOTES:

[1] ‘History of Architecture in all Countries.’ 2nd ed. Murray, 1867.

[2] ‘History of Architecture,’ vol. ii. pp. 445-756, Woodcuts 966-1163.

[3] A distinguished German professor, Herr Kinkel of Zürich, in his ‘Mosaik zur Kunstgeschichte, Berlin, 1876,’ has lately adopted my views with regard to the age of Stonehenge without any reservation, though arriving at that conclusion by a very different chain of reasoning from that I was led to adopt.

[4] The following brief résumé of the principal events in the ancient history of India has no pretensions to being a complete or exhaustive view of the subject. It is intended only as such a popular sketch as shall enable the general reader to grasp the main features of the story to such an extent as may enable him to understand what follows. In order to make it readable, all references and all proofs of disputed facts have been postponed. They will be found in the body of the work, where they are more appropriate, and the data on which the principal disputed dates are fixed will be found in an Appendix especially devoted to their discussion. Unfortunately no book exists to which the reader could with advantage be referred; and without some such introductory notice of the political history and ethnography the artistic history would be nearly, if not wholly, unintelligible.

[5] ‘Asiatic Researches,’ vol. xv. p. i.

[6] Almost the only person who has of late done anything in this direction is Sir Walter Elliot. His papers in the ‘Journal of the Royal Asiatic Society’ and the ‘Madras Journal’ throw immense light on the subject, but to complete the task we want many workers instead of only one.

[7] All this has been so fully gone into by me in my work on ‘Tree and Serpent Worship,’ pp. 63, _et seqq._, that it will not be necessary to repeat it here.

[8] Dr. Caldwell, the author of the ‘Dravidian Grammar,’ is the greatest and most trustworthy advocate of this view.

[9] ‘Tree and Serpent Worship,’ pp. 244-247.

[10] In Arrian there is a curious passage which seems certainly to refer to this people. “During the space,” he says, “of 6042 years in which the 153 monarchs reigned, the Indians had the liberty of being governed by their own laws only twice, once for about 200 years, and after that for about 120 years.”--‘Indica,’ ch. ix. The Puranas, as may be supposed, do not help us to identify these two periods.

[11] I cannot help fancying that they occupied some part of southern India, and even Ceylon, before the arrival of the Dravidians. It seems difficult otherwise to account for the connection between Behar and Ceylon in early ages, and the spread of Buddhism in that island leaping over the countries which had been Dravidianised.

[12] I cannot help suspecting that the Gonds also belong to this northern race. It is true they speak a language closely allied to the Tamil; but language, though invaluable as a guide, is nearly useless as a test of affinity. The Romans imposed their language on all the diverse nationalities of Italy, France, and Spain. We have imposed ours on the Cornish, and are fast teaching the Irish, Welsh, and Highlanders of Scotland to abandon their tongue for ours, and the process is rapidly going on elsewhere. The manners and customs of the Gonds are all similar to those of the Coles or Khonds, though, it is true, they speak a Dravidian tongue.

[13] The most pleasing of the histories of Buddha, written wholly from a European point of view, is that of Barthélemy St. Hilaire, Paris. Of those partially native, partly European, are those of Bishop Bigandet, from the Burmese legends, and the ‘Romantic History of Buddha,’ translated from the Chinese by the Rev. S. Beal. The ‘Lalita Vistara,’ translated by Foucaud, is more modern than these, and consequently more fabulous and absurd.

[14] There may possibly be an error of forty to sixty years in this date; but, on the whole, that here given is supported by the greatest amount of concurrent testimony, and may, after all, prove to be minutely correct.

[15] ‘Foé Koué Ki,’ xxv. ch. 11; ‘Mahawanso,’ v. p. 20; ‘Journal of the Asiatic Society of Bengal,’ vol. vi. 527.

[16]

One coin at least of the period is well known. It belongs to a king called Kunanda or Krananda, generally assumed to be one of the nine Nandas with whom this dynasty closed. In the centre, on one side, is a Dagoba with the usual Buddhist Trisul emblem over it, and a serpent below it; on the right the Sacred Tree, on the left a Swastica with an altar? on the other side a lady with a lotus (Sri?) with an animal usually called a deer, but from its tail more probably a horse, with two serpents standing on their tails over its head, which have been mistaken for horns. Over the animal is an altar, with an umbrella over it. In fact, a complete epitome of emblems known on the monuments of the period, but savouring much more of Tree and Serpent worship than of Buddhism, as it is now known. ‘Journal of the Koyal Asiatic Society,’ vol. i. (N.S.) p. 447, _et seqq._

[17] All these particulars, it need hardly be said, are taken from the 12th and 15th chapters of the ‘Mahawanso,’ confirmed by the inscriptions themselves and the relics found at Sanchi, to all which reference will be made hereafter.

[18] Wilson’s ‘Hindu Drama,’ vol. xii. p. 151, _et seqq._, edition 1871.

[19] Lassen, it is true, brings these dates down by ten years below where I have placed it. But he overlooks the fact that according to his hypothesis Asoka, in the sixteenth year of his reign, would claim Magas as his ally ten or twelve years after his death, which is improbable.

[20] For complete details of these two monuments and the dates, the reader is referred to my ‘Tree and Serpent Worship,’ which is practically devoted to a description of these two monuments.

[21] ‘Vie et Voyages de Hiouen Thsang,’ i. p. 215. It need hardly be said that all these particulars are taken from the three volumes relating his Indian experiences, translated by Stanislas Julien.

[22] This does not apply to Orissa, which, from its remote situation, and having at that time no resident Buddhist population, seems to have escaped being drawn into the vortex of these troubles.

[23] The best and most accepted account of these events is found in Vivien de St. Martin’s ‘Les Huns blancs,’ Paris, 1849.

[24] Cunningham’s ‘Numismatic Chron.,’ viii. 175; ‘Journal of the Asiatic Society of Bengal,’ vii. 704; Lassen’s ‘Indische Alterth.,’ ii. p. 24.

[25] I wrote a paper stating the evidence in favour of this last view, which I intended should appear in the ‘Journal of the Asiatic Society.’ The evidence being, however, incomplete, it has only been printed for private circulation.

[26] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. viii. p. 28.

[27] Ibid., vol. v. p. 42.

[28] The argument on which these assertions are founded is stated at length in the privately printed pamphlet alluded to on preceding page. It is too long to insert here, but, if not published before this work is complete, an abstract will be inserted in the Appendix.

[29] ‘Journal of the Koyal Asiatic Society,’ vol. iii. p. 202.

[30] For an exhaustive description of this subject see Priaulx, ‘India and Rome,’ London, 1873. My own impressions are, I confess, entirely in favour of the northern origin of the embassy. We are now in a position to prove an intimate connection between the north of India and Rome at that time. With the south it seems to have been only trade, but of this hereafter.

[31] ‘Dravidian Grammar,’ second edition, London, 1875, p. 129, _et seqq._

[32] Sir Walter Elliot and others frequently speak of Buddhist monuments in the south. I have never, however, been able to see a photograph or drawing of any one except at Amravati and its neighbourhood.

[33] In his ‘Elements of South Indian Palæography,’ Mr. Burnell, the last and best authority on the subject, divides the South Indian alphabet into Chera, Chalukya, and Vengi. The first, he states, appears in Mysore in the second half of the 5th century. The oldest specimen of the second he dates from the first half of that century. The third is more modern.

[34] I am, of course, aware of the existence of a so-called Buddhist pagoda at Negapatam. It was, however, utilised by the British--for railway purposes, I believe--before it was photographed, so its history may for ever remain a mystery. On the spot it was apparently known as the Jaina (hence China) pagoda, which it may have been. To me it looks like the gopura of a small Hindu temple, but I have no real knowledge on the subject. See Yule’s ‘Marco Polo,’ vol. ii. p. 320, second edition.

[35] ‘Storia della Scultura, dal suo risorgimento in Italia sino al secolo di Napoleone,’ Venezia, 1813.

[36] “The ritual of the Veda is chiefly, if not wholly, addressed to the elements, particularly to fire.”--H. H. Wilson, ‘Asiatic Researches,’ xvii. p. 194; ibid., p. 614.

[37] A list of the twenty-four Buddhas, with these particulars, is given in the introduction to Tumour’s ‘Mahawanso,’ p. 32. Representations of six or seven of these Bodhi-trees, with the names attached, have been found at Bharhut, showing at least that more than four were recognised in the time of Asoka. If the rail there were entire, it is probable representations of the whole might be found.

[38] Stobæus, ‘Physica,’ Gaisford’s edition, p. 54. See also Priaulx, ‘India and Rome,’ p. 153.

[39] Wilson’s ‘Ariana Antiqua,’ plates 10, 11.

[40] A book has recently been published by the late Mr. Breeks, of the Madras Civil Service, on the primitive tribes of the Nilagiris, which gives a fuller account of these ‘rude stone monuments’ than any other yet given to the public. It can hardly, however, be accepted as a solution of the problem, which requires a wider survey than he was able to make.

[41] The serpent of Siva is always a cobra, or poisonous snake, and used by him as an awe-inspiring weapon, a very different animal from the many-headed tutelary Naga, the guardian angel of mankind, and regarded only with feelings of love and veneration by his votaries. It may also be remarked that no tree is appropriated to Siva, and no trace of tree worship mingled with the various forms of adoration paid to this divinity--a circumstance in itself quite sufficient to distinguish this form of faith from that of the Dasyu group which pervaded the valley of the Ganges.

[42] Page 41. Dr. Cornish, in the introduction to the ‘Madras Statistical Tables,’ p. 67, states this at only 30,000,000--a very considerable difference; but on the whole I am inclined to place faith in Dr. Caldwell’s figures.

[43] ‘Madras Report,’ p. 90.

[44] These remarks must not be taken as applying to sculpture also. It is quite true that no stone sculptures have yet been found in India of an earlier date than the age of Asoka; but, as will be seen in the sequel, the perfection the Indian artists had attained in stone sculpture when they executed the bas-reliefs at Bharhut (B.C. 200), shows a familiarity with the material that could only be attained by long practice.

[45] No mention of temples, or, indeed, of buildings is, I believe, found in the Vedas, and though both are frequently alluded to, and described in the Epic Poems and the Puranas, this hardly helps us; first because, like all verbal descriptions of buildings, they are too vague to be intelligible, and secondly, because there is no proof that the passages containing these descriptions may not have been interpolated after--probably long after--the Christian Era.

[46] I believe I was the first to ascertain these facts from a personal inspection of the monuments themselves. They were communicated to the Royal Asiatic Society in a paper I read on the ‘Rock-cut Temples of India,’ in 1842. Every subsequent research, and every increase of our knowledge, has tended to confirm those views to such an extent that they are not now disputed by any one acquainted with the literature of the subject, though some writers do still indulge in rhapsodies about the primæval antiquity of the caves, and their connection with those of Egypt, &c. Till all this is put on one side, no clear idea can be obtained of the true position of the art in India.

[47] From two Sanscrit words, Dhatu, a relic, and Garbha (Pali, Gabbhan), the womb, receptacle, shrine of a relic. (Turnour, ‘Mahawanso,’ p. 5.) The word Pagoda is probably a corruption of Dagoba.

[48] In Nepal, according to Hodgson, and, I believe, in Thibet, the monuments which are called Stupas in India are there called Chaityas. Etymologically, this is no doubt the correct designation, as Chaitya, like Stupa, means primarily a heap or tumulus, but it also means a place of sacrifice or religious worship--an altar from Chíta, a heap, an assemblage, a multitude, &c. (Monier Williams’ ‘Sanscrit Dictionary’ _sub voce_). Properly speaking, therefore, these caves ought perhaps to be called “halls containing a chaitya,” or “chaitya halls,” and this latter term will consequently be used wherever any ambiguity is likely to arise from the use of the simple term Chaitya.

[49] These inscriptions have been published in various forms and at various times by the Asiatic Societies of Calcutta and London (‘Journal of the Asiatic Society of Bengal,’ vol. vi. p. 566, _et seqq._; ‘Journal of the Royal Asiatic Society,’ vol. xii. p. 153, _et seqq._) and in various other publications, but always mixed up with extraneous matters. It is, however, very much to be regretted that a carefully-edited translation is not issued in some separate form easily accessible to the general public. An absolutely authentic and unaltered body of Buddhist doctrine, as it stood 250 years before the birth of Christ, would be one of the most valuable contributions possible to the religious history of the modern world, and so much has been already done that the task does not seem difficult. Among other things, they explain to us negatively why we have so little history in India in these days. Asoka is only busied about doctrines. He does not even mention his father’s name; and makes no allusion to any historical event, not even those connected with the life of the founder of the religion. Among a people so careless of genealogy, history is impossible.

[50] ‘Journal of the Asiatic Society of Bengal,’ vol. vi. p. 794.

[51] Ibid., plate 40.

[52] Ibid., p. 969, _et seqq._

[53] These dimensions are taken from Capt. Burt’s drawings published in the ‘Journal of the Asiatic Society of Bengal,’ vol. iii. plate 3.

[54] ‘Tree and Serpent Worship,’ plates 9, 10, 10a, _et passim_.

[55] ‘Archæological Reports,’ vol. i. p. 274, plate 46.

[56] ‘Tree and Serpent Worship,’ plates 1 and 5, and plates 89 and 90.

[57] Ibid., plate 42.

[58] In the description accompanying Daniell’s view of this cave he says: “On the pillars to the right, above the capital, is a group of lions, from the centre of which a few years since arose the chacra, or war disk of Vichnou, though not the least appearance of it at present remains.” On the left he remarked a figure of Buddha, which he mistook for Mahadeva, and in another part a row of bulls, and he adds: “The Chacra of Vichnou, the Mahadeva, and the bulls, seem not to favour the opinion of its being a temple of the Bhoods.” He was not aware how inextricably these religions were mixed up at the time when this cave was excavated, about A.D. 400.

[59] Turnour in ‘Journal of the Asiatic Society of Bengal,’ vol. vii., p. 1013.

The fame of this distribution seems to have reached Europe at least as early as the 1st century of the Christian Era, inasmuch as Plutarch (‘Moralia,’ p. 1002, Dübner edition, Paris, 1841) describes a similar partition of the remains of Menander, among eight cities who are said to have desired to possess his remains; but as he does not hint that it was for purposes of worship, the significance of the fact does not seem to have been appreciated.

[60] ‘Mahawanso,’ p. 26, ‘Hiouen Thsang,’ vol. ii. p. 417.

[61] Account of the great bell at Rangoon, Hough, ‘Asiatic Researches,’ vol. xiv. p. 270.

[62] Abstracted from Turnour’s ‘Mahawanso,’ p. 4.

[63] There may be an error in this date to the extent of its being from fifteen to twenty years too early.

[64] The principal particulars of this story are contained in a Cingalese work called the ‘Daladavamsa,’ recently translated by Sir Mutu Comara Swamy. I have collected the further evidence on this subject in a paper I read to the Asiatic Society, and published in their ‘Journal’ (N.S.), vol. iii p. 132, _et. seqq._, and again in ‘Tree and Serpent Worship,’ p. 174, _et. seqq._

[65] The date being given as 245, Samvat has generally been assumed to be dated from the era of Vicramaditya. I am not aware, however, of any inscription of so early an age being dated from that era, nor of any Buddhist inscription in which it is used either then or thereafter.

[66] The same fate had overtaken another tooth relic at Nagrak in northern India. Fa Hian, B.C. 400, describes it as perfect in his 13th chapter. ‘Hiouen Thsang,’ vol. ii. p. 97, describes the stupa as ruined, and the tooth having disappeared.

[67] For a translation, &c., see ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. p. 33. See also Bird, ‘Historical Researches,’ Bombay, 1847.

[68] ‘Foé Koué Ki,’ ch. xii. p. 77.

[69] ‘Hiouen Thsang,’ vol. i. p. 83.

[70] ‘Foé Koué Ki,’ p. 353. A detailed account of its transference from the true Gandhara--Peshawur--to the new Gandhara in Kandahar will be found in a paper by Sir Henry Rawlinson, ‘Journal of the Royal Asiatic Society,’ vol. xi. p. 127.

[71] Among the bas-reliefs of the Bharhut tope is one representing just such a domical roof as this (Woodcut No. 90). It is not, however, quite easy to make out its plan, nor to feel sure whether the object on the altar is a relic, or whether it may not be some other kind of offering.

[72] ‘Bhilsa Topes, or Buddhist Monuments in Central India,’ Smith, Elder, and Co., 1854. One half of my work on ‘Tree and Serpent Worship,’ and forty-five of its plates, besides woodcuts, are devoted to the illustration of the great Tope; and numerous papers have appeared on the same subject in the ‘Journal of the Asiatic Society’ and elsewhere. A cast of the eastern gateway is in the South Kensington Museum.

[73] ‘Mahawanso,’ p. 76. See also ‘Tree and Serpent Worship,’ p. 99, _et seqq._, where all this is more fully set out than is necessary here.

[74] Cunningham, ‘Bhilsa Topes,’ p. 299, _et seqq._

[75] The Chandragupta inscription on the rail near the eastern gateway (‘Journal of the Asiatic Society of Bengal,’ vol. ii. p. 454) is evidently a subsequent addition, and belongs to the year A.D. 400.

[76] These views, plans, &c., are taken from a Memoir by Capt. J. D. Cunningham, ‘Journal of the Asiatic Society of Bengal,’ August, 1847.

[77] As all the particulars regarding all these topes, except the great one and No. 3 of Sanchi, are taken from Gen. Cunningham’s work entitled ‘Bhilsa Topes,’ published by Smith and Elder, in one volume 8vo., in 1854, it has not been thought necessary to repeat the reference at every statement.

[78] These dimensions and details are taken from Gen. Cunningham’s ‘Archæological Reports,’ vol. i. p. 107, _et seqq._

[79] ‘Asiatic Researches,’ vol. ix. p. 203.

[80] See also paper by Vesy Westmacott, ‘Calcutta Review,’ 1874, vol. lix. p. 68.

[81] ‘Archæological Reports,’ vol. i. p. 17.

[82] Ibid., p. 19.

[83] ‘Hiouen Thsang,’ vol. iii. p. 60.

[84] Buchanan Hamilton was told by the priests on the spot, in 1811, that it was planted there 2225 years ago, or B.C. 414, and that the temple was built 126 years afterwards, or in 289. Not a bad guess for Asoka’s age in a locality where Buddhism has been so long forgotten. Montgomery Martin’s ‘Eastern India,’ vol. i. p. 76.

[85] ‘Hiouen Thsang,’ vol. ii. pp. 464-468.

[86] ‘Archæological Reports,’ vol. i. p. 5.

[87] ‘Journal of the Asiatic Society of Bengal,’ 1834, vol. iv. p. 214. See also Cunningham, ‘Archæological Reports,’ vol. i. p. 5, _et seqq._

[88] ‘Hiouen Thsang, Festival of the three Religions at Allahabad in 643,’ vol. i. p. 254.

[89] A view of it is given, ‘Journal of the Asiatic Society of Bengal,’ vol. iv. p. 122.

[90] Beal’s ‘Fa Hian,’ p. 35.

[91] ‘Vie et Voyages de Hiouen Thsang,’ vol. i. p. 83.

[92] De Guigne’s ‘Histoire des Huns,’ vol. ii. p. 40, _et seqq._

[93] ‘Mahawanso,’ p. 71.

[94] ‘Ariana Antiqua,’ p. 43.

[95] ‘Ariana Antiqua,’ plate 10.

[96] Vassilief, ‘Le Bouddhisme, ses Dogmes,’ &c., Paris, 1865, p. 31, _et passim_. He spells the words Makhaiana and Khinaiana.

[97] Beal’s translation, p. 26.

[98] Honigberger, ‘Reise.’

[99] Mr. Masson’s account was communicated to Professor Wilson, and by him published in his ‘Ariana Antiqua,’ with lithographs from Mr. Masson’s sketches which, though not so detailed as we could wish, are still sufficient to render their form and appearance intelligible.

[100] The length of time over which these coins range--more than 200 years--is sufficient to warn us what caution is requisite in fixing the date of buildings from their deposits. A tope cannot be earlier than the coins deposited in it, but, as in this case, it may be one or two hundred years more modern.

[101] ‘Ariana Antiqua,’ p. 109.

[102] Thomas in ‘Prinsep,’ vol. i. p. 144.

[103] ‘Journal of the Asiatic Society of Bengal,’ vol. iii. p. 559.

[104] Thomas in ‘Prinsep,’ p. 148.

[105] ‘Archæological Reports,’ vol. ii. p. 167, plate 65.

[106] Thomas’s ‘Prinsep,’ vol. i. p. 94.

[107] In the text it is certainly printed “three” with a reference to 19 in the plate 21 of vol. iii. The latter is undoubtedly a misprint, and I cannot help believing the former is so also, as only one fragment is figured; and Prinsep complains more than once of the state of the French MS. from which he was compiling his account. I observe that General Cunningham, in his volume just received, adopts the same views. At p. 78, vol. v., he says: “I have a strong suspicion that General Ventura’s record of three Sassanian coins having been found below deposit B may be erroneous.”

[108] ‘Journal of the Asiatic Society of Bengal,’ vol. iii. plate 21, fig. 18.

[109] ‘Foé Koué Ki,’ chap. xiii.

[110] ‘Fa Hian,’ Beal’s translation, p. 32. ‘Hiouen Thsang,’ vol. i. p. 89.

[111] ‘Archæological Reports,’ vol. ii. p. 172.

[112] ‘Tree and Serpent Worship,’ Preface to the First Edition.

[113] ‘Archæological Reports,’ vol. i. p. 10.

[114] ‘Memorandum,’ dated 13th April, 1874, printed by the Bengal Government, but not published.

[115] ‘Voyages dans les Contrées Occidentales,’ vol. i. p. 465.

[116] ‘Archæological Reports,’ vol. i. plates 8 to 11.

[117] For this last determination, see ‘Tree and Serpent Worship,’ p. 99, _et seqq._

[118] It is to be hoped that when Gen. Cunningham publishes the volume he is preparing on the Bharhut Tope, he will add photographs of the pillars of this rail. It would add immensely to the value of his work if it afforded the means of comparing the two. Some illustrations of the sculpture from Major Kittoe’s drawings will be found in ‘Tree and Serpent Worship,’ woodcuts 7, 20, 24. Two of them are reproduced here, the first representing a man on his knees before an altar worshipping a tree, while a flying figure brings a garland to adorn it. The other represents a relic casket, over which a seven-headed Naga spreads his hood, and over him an umbrella of state. There are, besides, two trees in a sacred enclosure, and another casket with three umbrellas (Woodcuts Nos. 25, 26). They are from drawings by Major Kittoe.

[119] ‘Tree and Serpent Worship,’ p. 99, _et seqq._

[120] When I wrote my work on ‘Tree and Serpent Worship’ nothing was practically known as to the age of the jatakas, or the early form in which they were represented; much, therefore, that was then advanced was, or at least appeared to others to be, mere guess work, or daring speculation. It is, consequently, no small satisfaction to me to find that this subsequent discovery of a monument 200 years earlier does not force me to unsay a single word I then said. On the contrary, everything I then advanced is confirmed, and these inscriptions render certain what before their discovery was necessarily sometimes deficient in proof.

[121] The following outline (Woodcut No. 28, on the next page) of one of the bas-reliefs on a pillar at Bharhut may serve to convey an idea of the style of art and of the quaint way in which the stories are there told. On the left, a king with a five-headed snake-hood is represented, kneeling before an altar strewn with flowers, behind which is a tree (_Sirisa Accasia_?) hung with garlands. Behind him is an inscription to this effect, “Erapatra the Naga Raja worships the Divinity (Bhagavat).” Above him is the great five-headed Naga himself, rising from a lake. To its right a man in the robes of a priest standing up to his middle in the water, and above the Naga a female genius, apparently floating in the air. Below is another Naga Raja, with his quintuple snake-hood, and behind him two females with a single snake at the back of their heads--an arrangement which is universal in all Naga sculpture. They are standing up to their waists in water. If we may depend on the inscription below him, this is Erapatra twice over, and the females his two wives. I should, however, rather be inclined to fancy there were two Naga Rajas represented with their two wives.

This bas-relief is further interesting as being an epitome of my work on ‘Tree and Serpent Worship.’ As expressing in the shortest possible compass nearly all that is said there at length, it will also serve to explain much that is advanced in the following pages. As it is 200 years older than anything that was known when that book was written, it is a confirmation of its theories, as satisfactory as it is complete.

[122] ‘Mahawanso,’ Introduction, p. 32.

[123] Outlines of these sculptures are given in General Cunningham’s third volume of his ‘Reports,’ plate 6. I have photographs of the whole, which represent what is omitted in the lithographs.

[124] General Cunningham collected and translated 196 inscriptions from this tope, which will be found in his work on the Bhilsa Topes, p. 235, _et seqq._, plates 16-19.

[125] The details from which these determinations are arrived at will be found in ‘Tree and Serpent Worship,’ p. 98, _et seqq._ It is consequently not necessary to repeat them here.

[126] It is very much to be regretted that when Lieut. Cole had the opportunity he did not take a cast of this one instead of the eastern. It is far more complete, and its sculptures more interesting.

[127] For details of these sculptures and references, I must refer the reader to my work on ‘Tree and Serpent Worship,’ where they are all represented and described in great detail. Sculptures do not, strictly speaking, belong to this work, and, except for historical purposes, are not generally alluded to.

[128] They must certainly have been very common in India, for, though only one representation of them has been detected among the sculptures at Sanchi (‘Tree and Serpent Worship,’ plate 27, fig. 2), at least ten representations of them are found at Amravati, plates 59 (fig. 2), 60 (fig. 1), 63 (fig. 3), 64 (fig. 1), 69, 83 (fig. 2), 85 (figs. 1 and 2), 96 (fig. 3), 98 (fig. 2), and no doubt many more may yet be found.

[129] ‘Tree and Serpent Worship,’ Appendix I. p. 270.

[130] In Burmah at the present day a roll precisely similar to this, formed of coloured muslin, distended by light bamboo hoops, is borne on men’s shoulders in the same manner as shown here, on each side of the procession that accompanies a high priest or other ecclesiastical dignitary to the grave.

[131] For the reasons of the following determination and other particulars, the reader is referred to my work on ‘Tree and Serpent Worship,’ where the whole are set out at length. A short account of the tope will also be found in the ‘Journal of the Royal Asiatic Society,’ vol. iii. (N. S.) p. 132, _et seqq._

[132] ‘Histoire de Hiouen Thsang,’ traduite par Julien, vol. i. p. 188.

[133] It is probable that a tolerably correct idea of the general exterior appearance of the buildings from which these caves were copied may be obtained from the _Raths_ (as they are called) of Mahavellipore (described further on, p. 328). These are monuments of a later date, and belonging to a different religion, but they correspond so nearly in all their parts with the temples and monasteries now under consideration, that we cannot doubt their being, in most respects, close copies of them. Curiously enough, the best illustrations of some of them are to be found among the unpublished sculptures of the Bharhut Tope.

[134] The only buildings in India I know of that gave the least hint of the external forms or construction of these halls are the huts of the Todas on the Nilgiri Hills. In a work recently published by the late Mr. Breeks, of the Madras Civil Service, he gives two photographs of these dwellings, plates 8 and 9. Their roofs have precisely the same elliptical forms as the chaitya with the ridge, giving the ogee form externally, and altogether, whether by accident or design, they are miniature chaitya halls. Externally they are covered with short thatch, neatly laid on. Such forms may have existed in India two thousand years ago, and may have given rise to the peculiarities of the chaitya halls, but it is, of course, impossible to prove it.

[135] ‘Illustrations of the Rock-cut Temples of India,’ 1 vol., text 8vo., with folio plates. Weale, London, 1845.

[136] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. iii. pt. ii. p. 36, _et seqq._, and vol. iv. p. 340, _et seqq._

[137] Cunningham, ‘Archæological Reports,’ vol. i. p. 45.

[138] At Kondooty, near Bombay, there is a chaitya cave of much more modern date, which possesses a circular chamber like this. In the older examples it is probable a relic or some sacred symbol occupied the cell; in the later it may have been an image of Buddha. No plans or details of the Kondooty temple have, so far as I know, been published. I speak from information derived from MS. drawings.

[139] General Cunningham (‘Archæological Reports,’ vol. i. p. 45) and others are in the habit of calling this an Egyptian form. This it certainly is not, as no Egyptian doorway had sloping jambs. Nor can it properly be called Pelasgic. The Pelasgi did use that form, but derived it from stone constructions. The Indians only obtained it from wood.

[140] A very detailed account of all these caves will be found in Gen. Cunningham’s ‘Archæological Report’ for 1861-62.

[141] From a photograph and an unpublished paper by Professor Bhandarkur, read before the Oriental Congress.

[142] From Bhandarkur’s paper, _ubi supra_.

[143] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. p. 55.

[144] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. pp. 152-3.

[145] A few years ago it was reported that this screen was in danger of falling outwards, and I wrote repeatedly to India begging that something might be done to preserve it; but I have never been able to learn if this has been attended to. Only a small portion of the original ribbing of the Bhaja cave now remains. That of the Bedsa cave has been destroyed within the last ten or twelve years (‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. ix. p. 223); and it would be a thousand pities if this, which is the only original screen in India, were allowed to perish when a very small outlay would save it. Like the Iron pillar at Delhi which never rusts, teak wood that does not decay though exposed to the atmosphere for 2000 years, is a phenomenon worth the attention not only of antiquaries, but of natural philosophers.

[146] For further particulars regarding the Ajunta caves, the reader is referred to a paper I wrote in the ‘Journal of the Royal Asiatic Society,’ 1842, and republished afterwards with a folio volume of plates to illustrate it.

[147] These inscriptions are translated in Bhau Dajis’ paper on the Ajunta inscriptions, ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. viii. p. 63, as if found in cave 2. On the accompanying plate they are described as one on cave 10, the other on cave 12.

[148] Kittoe in ‘Journal of the Asiatic Society of Bengal,’ March, 1847, plate 6.

[149] Sir Charles Mallet, in the second volume of the ‘Bombay Literary Transactions,’ quotes a tradition that the Ellora caves were excavated by a Raja Eelu, 1000 years before his day. This might be true if applied to the Brahmanical Kailas, but hardly to any Buddhist cave in the series.

[150] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. p. 14.

[151] Loc. cit. p. 25.

[152] Introduction to ‘Mahawanso,’ p. 30.

[153] See Appendix.

[154] A tolerably correct representation of these sculptures is engraved in Langle’s ‘Hindostan,’ vol. ii. p. 81, after Niebuhr. The curious part of the thing is, that the Buddhist figures of the Karli façade are not copied here also, from which I would infer, as well as from their own intrinsic evidence, that they were more modern than even this cave.

[155] For further particulars regarding this cave, the reader is referred to my work on the ‘Rock-cut Temples of India,’ p. 36, plates 11 and 12.

[156] The plates in Gen. Cunningham’s ‘Archæological Reports,’ vol. ii. pl. 70 and 74, are on too small a scale to be of much use. I have not myself visited these caves.

[157] The particulars of the architecture of these caves are taken from Gen. Cunningham’s report above alluded to. I entirely agree with him as to their age, and am surprised Dr. Impey could be so mistaken regarding them. ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. p. 336, _et seqq._

[158] Throughout this work the term “Vihara” is applied only to monasteries, the abodes of monks or hermits. It was not, however, used in that restricted sense only, in former times, though it has been so by all modern writers. Hiouen Thsang, for instance, calls the Great Tower at Buddh Gaya a vihara, and describes similar towers at Nalanda, 200 and 300 feet high, as viharas. The ‘Mahawanso’ also applies the term indiscriminately to temples of a certain class, and to residences. My impression is that all buildings designed in storeys were called viharas, whether used for the abode of priests or to enshrine relics or images. The name was used to distinguish them from stupas or towers, which were always relic shrines, or erected as memorials of places or events, and never were residences or simulated to be such, or contained images, till the last gasp of the style, as at Kholvi. At present this is only a theory; it may, before long, become a certainty. Strictly speaking, the residences ought probably to be called Sangharamas, but, to avoid multiplication of terms, vihara is used in this work as the synonym of monastery, which is the sense in which it is usually understood by modern authors.

[159] Vol. iv., Woodcuts Nos. 89, 90.

[160] Beal’s ‘Fa Hian,’ p. 139, ‘Hiouen Thsang,’ vol. iii. p. 102.

[161] ‘Journal of the Royal Asiatic Society,’ vol. vi. (N.S.) p. 257, _et seqq._

[162] ‘Archæological Reports,’ vol. i. p. 28, plate 16.

[163] ‘Hiouen Thsang,’ vol. i. p. 151.

[164] ‘Archæological Reports,’ vol. i. pp. 28-36, plate 16.

[165] Now in private hands in Birmingham.

[166] ‘Journal of the Asiatic Society of Bengal,’ vol. xxxiii. p. 360, _et seqq._

[167] Ibid., vol. xxiii. p. 469, _et seqq._

[168] For this and the other Sarnath remains see Cunningham’s ‘Archæological Reports,’ vol. i. p. 114, _et seqq._, plates 32-34.

[169] These dimensions are from plate 42, ‘Journal of the Asiatic Society of Bengal,’ for 1847, by the late Capt. Kittoe.

[170] This inscription first attracted the attention of Stirling, and a plate representing it very imperfectly is given in the 15th volume of the ‘Asiatic Researches.’ It was afterwards copied by Kittoe, and a translation, as far as its imperfection admitted, made by Prinsep, with the assistance of his pundits, and published. ‘Journal of the Asiatic Society of Bengal,’ vol. vi. p. 1080, _et seqq._

[171] ‘Journal of the Asiatic Society of Bengal,’ vol. vi. p. 1073, plate 54.

[172] ‘Journal of the Asiatic Society of Bengal,’ vol. vi. p. 1075.

[173] There is a very faithful drawing of this bas-relief by Kittoe in the ‘Journal of the Asiatic Society of Bengal,’ vol. vii. plate 44. But casts of all these sculptures were taken some three years ago by Mr. Locke, of the School of Design, Calcutta, and photographs of these casts, with others of the caves, are now before me. Reduced copies of some of these were published on plate 100, ‘Tree and Serpent Worship,’ 2nd edition, 1873.

[174] That there were Yavanas in Orissa about this time is abundantly evident, from the native authorities quoted by Stirling--‘Asiatic Researches,’ vol. xv. p. 258, _et seqq._ These represent them as coming from Kashmir, and Babul Des, or Persia, and one account names the invader as Hangsha Deo, which looks very like Hushka, or Huvishka (the brother of Kanishka), whose inscriptions are found at Muttra.--Cunningham, ‘Archæological Reports,’ vol. ii. p. 32, _et seqq._

[175] ‘Journal of the Asiatic Society of Bengal,’ vol. vii. plate 42. ‘Tree and Serpent Worship,’ plate 100.

[176] ‘Tree and Serpent Worship,’ plate 100, p. 105.

[177] There may have been a structural dagoba attached to the series, which may have disappeared.

[178] Wilson, ‘Ariana Antiqua,’ plate 10.

[179] These inscriptions were first published by Lieut. Brett, with translations by Dr. Stevenson, in the fifth volume of the ‘Journal Bombay Branch of the Royal Asiatic Society,’ p. 39, _et seqq._, plates 1 to 16. They were afterwards revised by Messrs. E. W. and A. A. West in the eighth volume of the same journal, p. 37, _et seqq._, and translated by Professor Bhandarkar in a paper not yet published, but to which I have had access. I have also been assisted by manuscript plans and notes by Mr. Burgess; and, though I have not seen the caves myself, I fancy that I can realise all their main features without difficulty.

[180] Professor Bhandarkar, in his paper on these inscriptions, passes over the inscriptions in the interior of the chaitya, without alluding to them in any way. Is it that there is any mistake about them? and that the cave is a century more modern than they would lead us to suppose? The answer is probably to be obtained on the spot, and there only.

[181] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. viii. p. 40.

[182] _Ante_, p. 129. See also plate 11 of my folio work on the ‘Rock-cut Temples,’ where the pillars of the two caves are contrasted as here.

[183] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. ix. p. 16.

[184] ‘Journal Bombay Branch Royal Asiatic Society,’ vol. viii. p. 42.

[185] Ib., vol. v. p. 49.

[186] ‘Tree and Serpent Worship,’ woodcut 12, p. 92.

[187] Ibid., plates 81, 91, 97, _et passim_.

[188] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. v. p. 55.

[189] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. viii. p. 56.

[190] The caves, it may be explained, were numbered consecutively, like houses in a street, beginning at the north end, the first cave there being No. 1, the last accessible cave at the southern end being No. 26.

[191] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. viii. p. 56. See also, ‘Journal of the Asiatic Society of Bengal,’ vol. v. p. 726.

[192] Curiously enough, on the roof of this cave there are four square compartments representing the same scene, in different manners--a king, or very important personage, drinking out of a cup with male and female attendants. What the story is, is not known, but the persons represented are not Indians, but Persians, and the costumes those of the Sassanian period. Copies of these pictures by Mr. Griffith are now exhibited in the India Museum at Kensington.

[193] ‘Rock cut Temples,’ pl. 8.

[194] Eight large lithographic plates illustrating these caves will be found in my work on the ‘Rock-cut Temples of India,’ 1843. In 1864 I published a small volume containing fifty-eight photographic illustrations of the same series. Reductions of some of the more important frescoes, copied by Major Gill, were fortunately published by Mrs. Speir in her ‘Life in Ancient India,’ in 1856; and since then Mr. Griffith, of the School of Arts at Bombay, has been employed to recover, as far as it can now be done, the frescoes destroyed in the Crystal Palace fire. If he is successful, these curious paintings may still be made available for the history of art in India. It is feared, however, that the means taken by Major Gill to heighten their colour before copying them, and the destructive tendencies of British tourists, have rendered the task to a great extent a hopeless one.

[195] _Ante_, p. 59.

[196] I possess a large collection of MS. drawings of these caves, made for Daniell by his assistants in 1795-6.

[197] ‘Voyage en Arabie et d’autres pays circonvoisins,’ 1776-80. Most of the plates referring to these caves were reproduced by Langles in his ‘Monuments d’Hindostan,’ vol. ii., plates 77, _et seqq._

[198] Schlagintweit, ‘Buddhismus in Thibet,’ plate 3.

[199] Plans of these caves, with descriptions and some architectural details, will be found in Gen. Cunningham’s ‘Archæological Reports,’ vol. ii. pp. 270-288, plates 77-84. Those of Dhumnar I have seen myself, but till those of Kholvi are photographed we shall not be able to speak positively regarding them; the General’s drawings are on too small a scale for that purpose.

[200] The Kholvi group is situated more than sixty miles north of Ujjain, that of Dhumnar about twenty-five further north, and deeper into the Central Indian jungles.

[201] ‘Journal of the Royal Asiatic Society,’ vol. iv. pp. 287-291.

[202] There is a representation of this cave in Dr. Bird’s book, plate 16, but so badly done that it requires being told what is intended in order to find it out.

[203] I have for some time possessed photographs of about one hundred objects obtained in these excavations, principally those in the Lahore Museum; and latterly I have received from Gen. Cunningham twenty large photographic plates, representing 165 separate objects recently obtained in a more methodical manner by himself, principally from Jamalgiri. These plates are, as I understand, to form part of the illustrations of a work he intends publishing on the subject. When it is in the hands of the public there will be some data to reason upon. At present there is scarcely anything to which a reference can be made.

[204] When Gen. Cunningham was selecting specimens in the Lahore Museum, to be photographed for the Vienna Exhibition, he complains that he could only ascertain the “find spot” of five or six out of the whole number--500 or 600. It is therefore to be regretted that, when publishing a list with descriptions of the 165 objects discovered by himself (‘Archæological Reports,’ vol. v. pp. 197-202), he does not mention where they came from, and gives the dimensions of a few only.

[205] The mode in which the excavations have recently been conducted by Government has been to send out a party of sappers in the cold weather to dig, but the officer in charge of the party has been the subaltern who happened to be in command of the company at the time. A new officer is consequently appointed every year, and no one has ever been selected because he had any experience in such matters or any taste for such pursuits. What has been done has been done wonderfully well, considering the circumstances under which it was undertaken; but the result on the whole is, as might be expected, painfully disappointing. Quite recently, however, it is understood that Gen. Cunningham has taken charge of the excavations, and we may consequently hope that in future these defects of arrangement will be remedied.

[206] In the fifth volume of his ‘Archæological Reports’ just received, Gen. Cunningham assumes that both these were stupas of the ordinary character. They may have been so, but both having steps up to them would seem to militate against that assumption. The circular one is only 22 ft., the square one 15 ft. in diameter, and there is consequently no room on either for a procession-path round the dome, if it existed; and, if this is so, of what use could the steps be? Lieut. Crompton, who excavated the Jamalgiri monastery, is clearly of opinion that it was a platform--see page 2 of his report, published in the ‘Lahore Gazette,’ 30th August, 1873. To prevent misunderstanding, I may mention that Gen. Cunningham, in his plate No. 14, by mistake, ascribes the plan to Sergt. Wilcher, instead of to Lieut. Crompton.

[207] ‘Embassy to Thibet,’ p. 317.

[208] ‘Journal Bombay Branch of the Royal Asiatic Society,’ vol. vii., No. 21, p. 116, _et seqq._

[209] These have been removed, and are now in Gen. Cunningham’s possession at Simla, I believe. He has sent me photographs of twelve of them.

[210] ‘Tree and Serpent Worship,’ plates 24 (fig. 3) and 36 (fig. 1).

[211] The modillion cornice, though placed on the capital in the photograph, belongs in reality to another part of the building.

[212] ‘Archæological Reports,’ vol. v. pp. 49 and 196.

[213] ‘The Palaces of Nineveh and Persepolis Restored.’ By the Author.