The Migration of Symbols: Chapter VI. The Winged Globe, the Caduceus, and the Trisula: III. The Transformations of the Trisula

---

III. The Transformations of the Trisula.

I have already referred to the importance of the Trident in the symbolism of Hinduism where, under the name trisula (tri three, and sula point, spear, or pale) it occurs amongst the most important attributes of Siva. This emblem exhibits no peculiarity of form here; it might as well figure in the hands of Hades, or of Poseidon. This is not the case, however, with respect to the Buddhist trisula or, at least, the symbol which bears this name amongst the Buddhists.

The trisula of the Buddhists, termed also vardhamana, "a crescent," may be described, in its

Fig. 142. The Trisula.

simplest form, as an omicron surmounted by an omega.

It is, however, rarely met with under such a simple form. The upper arc of the omicron, or rather of the disk, is nearly always flanked by two small circles, or by two horizontal strokes which often assume the appearance of two leaves or small wings (fig. 143 et seq.). The points of the omega change into fleurons; the disk itself rests on a staff or pedestal, and from its lower arc fall two spires similar to serpents’ tails, the ends of which are sometimes curved upwards (fig. 143), and sometimes downwards (fig. 154).

At times the trisula appears to have only an ornamentive significance. Thus we see it crowning balustrades and porticoes, adorning sword-scabbards,

Fig. 143. Trisula of Amaravati.
Journal of the Royal Asiatic Society, vol. xviii. (new series), fig. 1.)

and forming necklace-pendants and ear-rings. 1 But in most cases it unquestionably discharges the function of a symbol, and even of a religious symbol. Engraved on numerous coins by the side of religious emblems and images, it stands at the beginning and end of votive inscriptions in the caves of Western India. 2 The sculptures

Fig. 144. Trisula on Blazing Pillar.
(Fergusson. Tree and Serpent Worship, pl. lxxi.)

of the bas-reliefs exhibit it, in turn, on the staff of banners, on the back of an elephant, on an altar where homage is paid to it, and, lastly, on a pillar from which flames emerge (fig. 144). At

1

The oldest representations of the trisula are found—in conjunction with the principal symbols of Buddhism, the swastika, the stupa, the Sacred Tree, and so forth—on the coins of Krananda, 2 a native sovereign contemporary with Alexander or the earliest Seleucid kings.

Fig. 145. Coin of Krananda.

Nevertheless the trisula was far from being solely used by the Buddhists. In caves it is sometimes placed near symbols of the solar worship, and on the coins of the Indo-Scythic princes it is not only struck beside images of the Hindu god Siva but, what at first surprises us, beside those of Greek divinities, such as Zeus; 3 it is, moreover, possible that it became a mere monetary sign, as often happens with religious symbols used in coinage.

Among the Jains, who have so many affinities

with Buddhism, the twenty-fourth and last of the Tirthankaras or legendary saints of the sect has the vardhamana as a symbol; this personage even bears, according to Colebrook, the surname of Trisula. 1

It may be wondered, at first sight, that the innumerable texts left by Buddhism give us no positive information with regard to the meaning and origin of the trisula. Few symbols have given rise, in our own times, to more varied explanations.

Some have seen therein the monogram of Buddha; 2 others the symbol of the dharma, the Law, which sums up the doctrine of Buddhism; 3 others again a representation of the tri-ratna, the threefold jewel formed by Buddha, his Law, and his Church. 4 There are those who have discovered in the trisula the juxtaposition of the dharma chakra, the "Wheel of the Law" to the ancient letter , y, which itself is said to stand for the mystic formula ye dharma. 5 Some scholars think they recognize in it the combination of five letters symbolizing respectively intelligence (ma) and the four component elements of matter, air (ya), fire (ra), water (va), and earth (la). 6

Eugene Burnouf thought he found therein the vardhamana kaya, "the Propitious One," one of the sixty-five signs which, according to Buddhist

tradition, adorn the impression of Buddha's foot. 1

Finally, according to some writers, we must seek its origins amongst the less abstract images of the naturalistic forms of worship which preceded Buddhism.—M. Kern, laying stress on the actual meaning of vardhamana, the present participle of a verb signifying "to grow or increase," makes it the image of the "horned moon," and sees in the central protuberance of the trisula the nose with which we ourselves sometimes adorn the representation of the lunar crescent . 2 Edward Thomas seeks therein "an ideal combination of the sun and moon;" the alteration of the primitive form would, according to him, be due to a modification in the forms of worship, or to the overthrowing of the sovereigns termed Lunar by a Solar dynasty. 3

Mr. Burgess recognizes therein an image of the Thunderbolt; 4 Sir George Birdwood, a phallic emblem, or else the Tree of Life; 5 Mr. Monier Williams, "the Two Feet of Vishnu, with a star or embossment in the middle." 6

Lastly, M. Beal takes it for the superposition of the Flame on the Lotus-flower, and M. Senart, of the Trident on the Wheel. 7

Among all these more or less contradictory opinions the interpretation of M. Senart is not only the simplest and most rational, but it is also strictly confirmed by the evidence of the monuments. There are many trisulas in which the upper part of the figure is separated from the Disk, some again in which it assumes distinctly angular forms , instead of the rounded shape of the omega, ω; there are others, finally, in which it becomes beyond doubt a Trident, as amongst the sculptures of Buddha Gaya and of Boro-Budur. 1

The Trident superposed upon the Disk is also met with upon the coins of the anonymous prince

Fig. 146. Sculpture of Buddha Gayâ.
(Numismatic Chronicle, vol. xx. (new series), pl. ii., No. 37.)

known by his title of Basileus Sôter Megas, and on those of several native kings. 2

The only point on which I have some doubt is when M. Senart puts forward the Trident as the original feature, and, so to speak, the primitive nucleus of the Hindu trisula,—which would make it, at least in its origin, an essentially Sivait symbol, destined to represent the flash of the lightning.—As for myself, I should be more inclined to seek this nucleus in the Disk, and, consequently, to connect the trisula with the solar symbols.

From the most remote times the worship of the sun was widely spread throughout India, and, as nearly everywhere else, the sun was first of all represented there by a Disk, as may be seen from the sculptures of the ancient caves, and the ingots used for bartering before the introduction of coins properly so-called. 1 At a later date the Disk became a Wheel, and the Buddhists, who applied so many solar images and symbols to their form of worship, made it the Wheel of the Law, "formed of a thousand spokes darting out a thousand rays."

The secondary character of the omega (or Trident) in the trisula plainly follows from certain figures brought into view by Edward Thomas in

his valuable work on the solar symbols of India. These are circles drawn between four omegas. One of these circles exhibits four arrows radiating from the circle between the crescents.

These figures indicate clearly the function of the Trident in the trisula. Doubtlessly, in the hands of Siva—as formerly in the hands of Neptune, and, at a still earlier period, in those of the Assyrian god of the air and the storm—the Trident must symbolize the flash of the lightning. But may it not be questioned whether, considered apart, it should

not be held to be, in a wider sense, the image of a Three-tongued Flame, and, consequently, when coupled with the Disk, an emblem of fire, or of solar radiation?

Among the sculptures of Boro-Budur, in the island of Java, the Trident, which in some religious scenes is exhibited above the Disk, or Rosette, is replaced at times by a Three-pointed Flame. 1 Eug. Burnouf had already noticed, in the coloured representations of the Buddhas of Nepaul, that the headdress of the Master exhibits a ball terminating above in a kind of flame, and that on many Singhalese statues this flame takes the shape of "a kind of lyre or trident." 2 Finally, according to Mr. Beal, the trisula personifies, amongst the Buddhists of the north, the heaven of pure flame superposed upon the heaven of the sun. 3

The trisula is then certainly a Hindu symbol. It seems, however, to have felt, at an early date, the influence of the Caduceus. Perhaps even it was in order to approach the latter in appearance that the primitive Trident of the Hindu symbol assumed the rounded forms of the omega and placed itself in direct contact with the Disk.

Those who have some hesitation in admitting the possibility of discovering in a complex symbol the traces of a double antecedent—as we find in a child the characteristic features of both its parents—need only cast a glance over the following picture whose contents I have taken from the coinage and figured monuments of India.

I wish to call especial attention to the figures d and e. They resemble one another so closely that writers commonly rank the former as a trisula 

when they find it on the coins of certain Indo-Scythic princes. Yet there is no doubt but that it is directly connected with the Caduceus. Moreover,

Fig. 149. Caducei and Trisulas. 1

Fergusson himself has written with respect to the trisula:—"It bears a singular resemblance to the sign of the planet Mercury, or to the Caduceus of the god who bears this name." 2

It is also representations of serpents intertwined, closely allied to the Caduceus, which supply us with the earliest type of the lower appendages, in the shape of spires, observed in the trisulas of Sanchi and Amaravati (fig. 143 and 154).

On the other hand, the most complex forms of the trisula exhibit an unquestionable likeness to

some types of the Winged Globe which have been observed in Asia Minor, Mesopotamia and Persia.

In both figures the centre is occupied by a Disk which is sometimes converted into the Wheel, or the Lotus-flower. Does not the upper part of the trisula, which I have termed the omega, recall the horns of the Mesopotamian emblem, if we only take into consideration the knob between the horns caused by the upper arc of the disk (fig. 150);

Fig. 150. Cylinder of Chalcedon.
(Lajard. Culte de Mithra, pl. lii., fig. 2.)

or else the projection caused by the tuft of plumes which surmounts the disk of certain Winged Globes (fig. 151)?

Fig. 151. Coin of the Satraps of Tarsus.
(Lajard. Culte de Mithra, pl. lxiv.)

The shaft, frequently conical, on which some trisulas rest, takes the place of the fan-shaped tail, and the spires traced on both sides of this support correspond with the lines, ending in a loop, which descend on either side of the tail in the ornithomorphic Disks of Western Asia.

In order to account for this likeness, I will venture on the hypothesis that these forms of the trisula must have been subjected, during their development, to the plastic influence of the ancient Egyptian symbol which had come to India by way of Assyria and Persia.

The time is past when, dazzled by the sudden discovery of Vedic literature, and fascinated also by the verification of our relationship to the Aryan races of Asia, we turned towards India to seek them the universal source of symbols and dogmas, myths and gods. Since we caught a glimpse of the great antiquity of the civilizations which had reached their prime on the borders of the Euphrates and the Nile at a period when the ancestors of the Aryans were still wandering over the table-lands of Central Asia, we are far more inclined to locate in Mesopotamia, or even in Egypt, the earliest artistic centres which shed their light over the ancient world, from the Mediterranean to the Indian Ocean.

On the other hand, India, up to the Mahometan conquest, did not live in the isolated condition to which historians were long pleased to relegate it. Sir George Birdwood goes, perhaps, a little too far when he asserts, in a general manner, that nearly all the symbols of India are of Mesopotamian origin. 1 But it is none the less an established fact now, that the products of western art and symbolism must have made their way to the valley of the Indus before the appearance of the oldest stone monuments on which ancient India has left us a vestige of its beliefs. 2

Without taking into consideration the intercourse which is supposed to have existed between the dwellers on the banks of the Indus and the most ancient empires of the Euphrates and the Nile, and without laying stress either on the factories which the Phœnicians are said to have established in southern India, I will recall to mind that Darius I

had annexed the valley of the Indus, and the actual province of the Punjab, about the end of the seventh century B.C. 1 Such competent authorities as James Fergusson and General Cunningham have shown that India borrowed its earliest style of architecture from the Persians, 2 and in the north-east of the peninsula there have been repeatedly found products of Persian art which date back to Darius and his successors—especially cylinders and coins bearing the Winged Circle. 3 It was in one of these finds that one of the coins of Tarsus containing a Winged Circle so nearly allied to the trisula was found (fig. 151).

It seems to be now generally admitted that the Indian alphabets are of Semitic origin. 4 Why then should religious symbols not have followed the same paths as the symbols of language and the creations of art?

In the centuries following the expedition of Alexander, it was Greek, or rather Græco-Asiatic art which influenced the development of Indian architecture and sculpture. Coins show us first of all sovereigns of Greek origin, who Hellenize Bactria, the Cabul country, and the valley of the Indus as far as the Ganges basin; then Scythic and Parthian princes who maintain, until the second century of our era, the language and traditions of this civilization imported from the West.

1

At the same time, however, there occurs in India a veritable invasion of Iranian divinities. Mr. Percy Gardner and M. James Darmesteter, the former through investigating the coins of the British Museum, and the latter through the study of Persian traditions in the Hindu epic poem, came simultaneously to the somewhat startling conclusion that western India, after being Hellenised under the Indo-Bactrians, had been largely Iranised under the Indo-Scythians. 2 On the coins of the latter not only do Zeus, Pallas, Helios, Selene, Poseidon, Herakles, and Serapis alternate with Mithras, Mao, and Atar, as well as with Siva, Lukshmi, and even Buddha, but also the classic forms of the Thunderbolt, the Caduceus and the Cornucopia occur side by side with the Buddhist trisula and Wheel as well as with the Sivait Trident and bull.

India has always been the favourite home of religious syncretism; but at no period of its history did it open its portals to so many different forms of worship—even under Akbar, that great Mogul who was willing to combine in a single religion the beliefs of the Mahometans, the Hindus, the Parsees, the Jews and the Christians.

How could symbols have escaped an impulse which carried away the gods themselves? At any rate Buddhism would have been unfaithful to the spirit of its whole symbolism if, when acquainted with the emblems by which neighbouring religions represented their great solar divinity, or even their supreme god, it had not sought to appropriate them, either by adopting them as they stood with a new signification, or oftener by assimilating them, through slight linear alterations, with one or other of its favourite symbols.

It is, as we have seen, among the sculptures of Amaravati that the trisula assumes the form most closely allied to the ornithomorphic Disk. Now, nowhere else has the general influence of Græco-Asiatic art over native architecture and sculpture been verified in a more obvious manner. Already in the year 645 of our era the Chinese pilgrim Hiuen Thsang likened the sanctuary of Amaravati to the palaces of the Tahia, i.e., the inhabitants of Bactria. 1 The same fact has been recorded in our own times by James Fergusson: "There is," he writes, "so much of what is Greek, or rather Bactrian, in the architectural details of Amaravati that this monument must belong to a period nearer to the Christian era than the character of the inscriptions would lead one to suppose." And the eminent archæologist adds that the study of these sculptures seems to him destined to elucidate especially the interesting question of the intercourse, and even of the exchange of thought, between east and west. 2

The Winged Circle and the Caduceus are not,

moreover, the only factors which have reacted upon the genesis, or at least the development, of the trisula.

In an interesting paper, read in 1886 before the Royal Asiatic Society, Mr. Robert Sewell was perhaps the first to seek to the west of the Indus, and even in Egypt, the origins of the trisula, which he connects with the Scarab. The birthplace of a symbol is one thing; the origin of the figures which may have influenced its development is something very different. With this reservation, however, I must admit that there is nothing farfetched in this parallel, especially if, following

Fig. 152.
(Journal of the Royal Asiatic Society, vol. xviii. (new series), fig. 13.)

1 by which fact, according to our theory, its resemblance to the trisula is sufficiently explained. It must be observed, moreover, that several of

these Assyrian Scarabs hold between their fore-legs the Disk adorned with the uræi. 1

It may be noticed, by the way, that among the Egyptians the Trident already occurs in conjunction with the Winged Globe, at least in the texts. The inscription of Edfu, which relates how Horus was transformed into a Winged Globe in order to fight the armies of Set, gives him as a weapon a three-pointed spear. 2

On a number of monuments the Disk of the trisula changes into a Rosette, imitating an expanded lotus-blossom, as on the gate of Sanchi. The same monument further exhibits Lotuses at the ends of two stalks which spring from the basis of the central fleuron; finally, the two extreme points of the omega take a shape which suggests the calyx of a flower. We have already shown the solar character of the Lotus in the symbolism of the Hindus. 3 In the trisula the transformation of the Disk into a Lotus is therefore the equivalent, the plastic rendering of the transformation which, among the Buddhists, substituted the padma mani, or "Jewel of the Lotus," for the sûra mani, or "Jewel of the Sun"; whence the well-known formula:—Om! mani padme, "Oh! the Jewel in the Lotus!" 4

Elsewhere the trisula seems to reproduce the Sivait emblem of the lingam between two serpents standing erect. This figure, which when superposed upon the Disk has perhaps a doubly phallic

import, certainly seems quite foreign to the original doctrine of Buddhism. Yet here again the Buddhists have shown themselves to be masters in the art of adapting the symbols of other religions. M. Gustave Lebon, in his Voyage au Népaul, quotes a characteristic instance of the fanciful interpretations with the help of which Buddhism brings about or justifies these adaptations. "The linga," he writes, "is likewise adopted by the Buddhists of Nepaul as an emblem of the Lotus in which Adi-Buddha manifested himself, in the shape of a flame, at the beginning of the world." 1

It ought to be mentioned that, in the opinion of some writers, such as Ch. Lenormant and Baron de Witte, the Caduceus among the Greeks symbolized the combination of the two sexes in the

Fig. 153. Caduceus on an Amphora. 2
(From the Elite des monuments céramographiques, vol. iii., pl. xci.)

same individual, hermaphrodism in fact, 3—and in the valuable publication of those two archæologists upon the Elite des monuments céramographiques de la Grèce we find a variety of the Caduceus where a vertical projection, similar to the representation of the phallus, springs from the centre of the Crescent, which in this case is somewhat separated from the Disk.

On the coins of the Sunga dynasty the lingam placed between the serpents becomes the headdress of Buddha; the Disk stands for the Master's head, and the lateral appendages of the trisula are represented by two projections which stretch out horizontally on either side of this head. 1

The trisula is thus seen to be converted into an anthropoid figure. A transformation of the same kind, but still more accentuated, is observable in the three famous idols of Puri which General Cunningham long ago proved to be three ancient trisulas (figs. 154 and 155). 2

Fig. 154. Trisula of Sanchi.
(Musée des moulages, Brussels.)

These emblems were doubtlessly a great object of popular veneration at the period when Puri was a Buddhist sanctuary. When Brahminism came to establish itself there it contented itself with changing them, by means of a few slight alterations, into the image of Vishnu, or rather Jaganath, and his brother and sister. In thus appropriating the old solar symbol, still discernible in

spite of its successive alterations, Vishnu, moreover, did nothing but recover what belonged to him, since he is, in Hinduism, pre-eminently the solar divinity.

Finally, the trisula, whose plasticity is only equalled by its power of absorption, borrows forms from the vegetable kingdom with the same freedom as from the human physiognomy. M. Rousselet points out the resemblance of the mystic symbol of the Buddhists to the kalpavriksh or Tree of Knowledge which the Jainas represented

FIG. 155. IDOL OF JAGANNATH. (ROUSSELET. L’Inde des Rajahs, p. 517.)

Fig. 155. Idol Of Jagannath.
(Rousselet. L’Inde des Rajahs, p. 517.)

by a stem with three branches on the mitre of the Tîrthankaras sculptured on the caves of Gwalior. 1 A similar combination is observable on coins of the Sunga dynasty, where the upper part of the trisula, forming the head-dress of Buddha, is transformed into a regular crown of branches. 2 On other monuments the stem on which the trisula rests becomes the trunk of a tree whose branches are laden with conventional fruits and interlaced with necklaces of jewels. 3

This vegetalization of the trisula, or, properly

speaking, of the Trident which surmounts it, is nowhere more evident than among the sculptures of Boro-Budur, where it literally merges into the Bô tree through a series of gradual transformations. "The shape of the points of the trisula," M. Ch. Leemans writes in his able commentary upon the Atlas published under the patronage of the Dutch government, "may sometimes

Fig. 156. Trident of Siva.
(Moor. Hindu Pantheon, pl. xlii.)

have been derived from that of a flame, or else from the calyx of a flower, or again from a symbolic tree." 1 The same observation applies

Fig. 157. Egyptian Lotus.

moreover to the Trident of Siva, which at times exhibits the forms of a lotus calyx depicted in the Egyptian manner (fig. 157).

Perhaps other transformations of the trisula might still be found at Boro-Budur. I will restrict myself to pointing out a detail which is not without interest: the same Disk which, when transformed into a most complicated ornament, is.

sometimes crowned by a Trident, is also met with between two serpents—which brings us back to the origin of the Winged Circle—the Globe of Egypt with the uræi.

Fig. 158. Bas-Reliefs of Boro-Budur. 1
(Boro-Boedoer, Atlas, pl. cccxvi. and ccclxx.)

Moreover this ornament, between which and certain forms of the trisula the transition is easily traced, commonly surmounts the entrance to the pagodas depicted in the bas-reliefs—in exactly the same manner as the Winged Globe adorns the lintel of the temples in Egypt and Phœnicia.

Fig. 159. Bas-Reliefs of Boro-Budur.
(Boro-Boedoer, Atlas, pl. cclxxxiii., No. 105.)

It is proper to point out that in the West there are some figures which exhibit an odd resemblance to the trisula. Such is, in the first place, the Cyprian image to which I have drawn attention (fig. 114), as presenting to our view a combination

of the Sacred Tree and the Winged Globe. The upper volutes with the flowered projection bisecting the centre of the arc recall the three points of the trisula with their central fleuron. The pair of volutes whose lower parts are bent downward on both sides of the base, suggest the ophidian appendages which, in the Buddhist symbol, descend on either side of the pedestal. Lastly in both cases, the middle part of the figures shows two leaves which by their position, as well as by their shape, suggest two small wings.

Mr. William Simpson, an English artist, well known on account of his archæological work in connection with the monuments of India, has

Fig. 160. Thunderbolt of Elis.

pointed out the resemblance of the trisula to some representations of the Thunderbolt graven on coins of Elis which date from the fifth century B.C. 1

Fig. 161. Sculpture of Boro-Budur.

This Thunderbolt especially resembles a somewhat enigmatical figure sculptured among the bas-reliefs of Boro-Budur, where, to all appearance, it plays the part of a trisula. 2

On the whole, there would be nothing remarkable

in these similarities, since the Thunderbolt and the trisula, as we have just seen, are both a development of the Trident.

What is more singular is to find a kind of trisula engraved on the flank of a lamb which adorns the serpent-shaped scroll on a pastoral staff dating from the Middle Ages. 1

Have we here an exchange or a mere coincidence? This is a question which I will not venture to decide, although nothing is wanting in the Christian symbol to make of it a real trisula, neither the Disk, nor the central point imitating a fleur-de-lis, nor the rounded projection of the two lateral points.

 

After all, if we have been able to find the antecedents, and, so to speak, the factors of the trisula, and even its probable signification in the creeds which preceded Buddhism, we have not learned much as to the meaning of this symbol in the religion which made most use of it. The fact is, that here the plastic figures cannot make up for the silence of the written monuments. So long as symbols remain images and are applicable to concrete objects, or physical phenomena, it is not impossible to discover the meaning which they in all probability bore. But when, having entered upon what may be called their derived or secondary phase, they become signs and express abstract ideas—which is nearly always the case in Buddhism—the field of interpretation becomes, so to speak, unlimited for critics, as sometimes also for the faithful.

The proper signification of the trisula remains then in the suppositive stage, although the purpose of its transformations does not always escape us. The issue of some hitherto unpublished text

can alone reveal to us the general and authentic meaning of this symbol, before which millions of our fellow-creatures have bowed down, but whose name we do not even know with any degree of certainty.

My purpose, moreover, was less to solve a problem whose solution has so far escaped the most competent minds, than to trace the transformations of the trisula in the course of its plastic development, and to show once again with what ease symbols of most dissimilar origin merge into one another as soon as, in their form, or in their meaning, there occur points of contact which are sufficient to facilitate this transition.

---

Footnotes

238:1 A. Cunningham. The Stupa of Barhut. London, 1879, pl. xlix., fig. 10; pl. 1., figs. 5 and 6.

238:2 Eug. Burnouf. Le Lotus de la Bonne Loi, p. 626.

239:1 A. Cunningham. The Stupa of Bharhut, and J. Fergusson. Tree and Serpent Worship, 1st vol., with plates. London, 1868, passim.—Cf. engraving of our frontispiece.

239:2 Edward Thomas has maintained that Krananda was identical with the Xandrames of Diodorus (Journal of the Royal Asiatic Society. London, vol. i., new series, p. 477, On the identity of Xandrames and Krananda), which would make this coin earlier than the year 317 B.C.—Wilson, on his side, makes Xandrames to be Chandragupta, the ancestor of Asoka (Introduction to the translation of the Mûdrarakshasa. The Theatre of the Hindus, vol. ii., 131, 132).

239:3 Percy Gardner. Coins in the British Museum. Greek and Scythic Kings of India and Bactria, pp. 106 and 107.

240:1 Colebrook. Observations on the Jainas, in the Asiatic Researches. London, 1809, vol. vii., p. 306.

240:2 J. Fergusson. Description of the Amaravati Tope, in the Journal of the Royal Asiatic Society. London (vol. iii. of new series, p. 162).

240:3 Edw. Thomas, in vol. iv. (new series) of the Numismatic Chronicle, p. 282, foot-note.

240:4 A. Cunningham. The Stupa of Barhut, p. III.

240:5 F. Pincott. The Tri-Ratna, in the Journal of the Royal Asiatic Society. London, vol. xix. (new series), p. 242.

240:6 A. Cunningham. The Topes of Central India, in the Journal of the Royal Asiatic Society. London (vol. xiii., 1st series), p. 114.

241:1 Eug. Burnouf. Le Lotus de la Bonne Loi. Paris, 1852, p. 627.

241:2 Kern. Der Buddhismus, German translation by Jacobi. Leipzig, 1884, vol. ii., pp. 241–242.

241:3 Edw. Thomas. On the identity of Xandrames and Krananda, in the Journal of the Royal Asiatic Society. London, vol. i. (new series), pp. 483–484.

241:4 Burgess. Archæological Report on Elura.

241:5 Sir George Birdwood, in the Journal of the Royal Asiatic Society. London, vol. xviii. (new series), p. 407.

241:6 Quoted by Mr. Greg. Archæologia. London, 1885, vol. xlviii., p. 320.

241:7 S. Beal. A Catena of Buddhist Scriptures from the East. London, 1871, p. 11.—E. Senart, Essai sur la légende du Bouddha in vol. vi. of the Journal Asiatique. Paris, 1875, p. 184.

242:1 Boro-Boedoer op het eiland Java. Leyden, 1873, Atlas, pl. cccxvi.

242:2 Percy Gardner. Op. cit., pl. xxiv., figs. 1–6. See also Senart. Journal Asiatique. Paris, 1875, vol. vi., p. 185.

243:1 Edward Thomas. The earliest Indian Coinage, in vol. iv. (new series) of the Numismatic Chronicle, p. 271. See also his article in vol. xx. of same series, The Indian Swastika.

244:1 Boro-Boudour op het eiland Java. Atlas, pl. cclxxx., fig. 100.

244:2 E. Burnouf. Le Lotus de la Bonne Loi, p. 539

244:3 S. Beal. A Catena of Buddhist Scriptures from the Chinese, p. 11.

245:1 a, Coin of Sophytes (Percy Gardner, pl. i., fig. 3).

b, Ancient ingot (Numismatic Chronicle, vol. iv., new series, pl. xi., fig. 28).

c, Old coin (Numismatic Chronicle, vol. iv., new series, pl. xi., fig. 16).

d, Coin of Azes (Percy Gardner, pl. XX., fig. 2).

e, Coin of Krananda (Journal of the Royal Asiatic Society, vol. i., new series, p. 475).

f, On a sculptured altar at Sanchi (Fergusson. Tree and Serpent Worship, Atlas, pl. xxv., fig. 3).

g, On the pole of a standard at Sanchi (Cunningham. The Bhilsa Topes, pl. xxxii., fig. 8).

h, On the pole of a standard of Sanchi (Fergusson. Tree and Serpent Worship, pl. xxxviii., fig. 1).

i, On the handle of an ivory instrument (Cunningham. Archæological Survey of India, vol. x., pl. ii., fig. 5).

245:2 Tree and Serpent Worship, p. 116.

247:1 Journal of the Royal Asiatic Society. London, 1886, vol. xviii. (new series), p. 407.

247:2 G. Rawlinson. The Five Great Monarchies of the East. London, 1862, vol. i., p. 101.—A. H. Sayce. Religion of the ancient Babylonians. London, 1887, pp. 137–138.

248:1 G. Maspero. Histoire ancienne des peuples de l’Orient. Paris, 1886, p. 618.

248:2 J. Fergusson. Tree and Serpent Worship, p. 94.—A. Cunningham. Archæological Survey of India, vol. v., Append. A.

248:3 Relics of ancient Persia, in the Proceedings of the Asiatic Society of Bengal. Calcutta, 1881, 1st part, p. 151; 1883, 1st part, pp. 64 and 261.

248:4 See upon this question the summing up of Mr. Cust in the Journal of the Royal Asiatic Society. London, 1884, vol. xvi. (new series), p. 325; also an article by M. Halévy in the Journal Asiatique. Paris, 1885, vol. ii.

249:1 Percy Gardner. Coins of Greek and Scythic Kings in India, p. lviii.

249:2 Percy Gardner. Id., § iv.—J. Darmesteter, in the Journal Asiatique. Paris, July-August, 1887.

250:1 For the identification of the Tahia with the Bactrians, see Percy Gardner. Op. cit., p. xxxi.

250:2 Description of the Amaravati Tope, in vol. iii. (new series) of the Journal of the Royal Asiatic Society. London.—The gates of Sanchi seem to date from the first years of our era although the tope itself is older by several centuries (Rousselet, L’Inde des Rajahs, p. 513).

251:1 See Perrot et Chipiez. Histoire de l’Art dans l’antiquité, vol. i., p. 811.

252:1 See Perrot et Chipiez. Histoire de l’Art dans l’antiquité, vol. ii., p. 399.

252:2 H. BRUGSCH. Die Sage von der geflügelten Sonnenscheibe, in the Abhandlungen der königlichen Gesellschaft der Wissenschaften zu Göttingen, vol. xiv. (1868–1869), p. 201.

252:3 S. BEAL. A Catena of Buddhist Scriptures, p. 11.

252:4 It must be observed, however, that the Disk is already found closely associated with the Lotus-flower in the symbolism of Asia Minor. (See the Histoire de l’Art dans l’antiquité, by MM. Perrot and Chipiez, vol. iii., fig. 509.)

253:1 In the Tour du Monde for 1886 (vol. li., p. 266)—At Bôrô-Boudour the Sivait linga has become a representation of the dâgoba (C. Leemans, Boro-Boedoer, p. 452).

253:2 Cf. the form of the trisula on the pillar of the sun at Buddha Gaya (our illustration, fig. 146).

253:3 Elite des monuments céramographiques de la Grèce. Paris, 1868, vol. iii., p. 197.

254:1 A. Rivett Carnac. Coins of the Sunga or Mitra dynasty, in the Proceedings of the Asiatic Society of Bengal, vol. xliv., 1st part, pl. vii. and viii.

254:2 The Topes of Central India, in the Journal of the Royal Asiatic Society. London (vol. iii. of 1st series). General Cunningham adds that these rude figures are likewise used in native calendars to represent Vishnu in his avatar of Buddha.

255:1 Rousselet. L’Inde des Rajahs, p. 370.

255:2 Rivett Carnac. Loc. cit.

255:3 F. Pincott. The Tri-Ratna, vol. xix. (new series) of the Journal of the Royal Asiatic Society. London, p. 243.

256:1 Boro-Boedoer, p. 455 of commentary.

257:1 See also same Atlas, pl. ccxxxvi., 11; cclxxxviii., 114; and especially ccxlviii., 36, where this subject rests on the apex of a triangle which corresponds with the pennated tail of the Mesopotamian globes.

258:1 The Trisula Symbol in the Journal of the Roy. As. Soc. for 1890, vol. xxii. (new series), p. 306.

258:2 Boro-Boedoer, vol. iii., pl. ccxlvi., fig. 35.

259:1 Cf. CAHIER et MARTIN. Mélanges d’archéologie, vol. iv. fig. 58.