^{page 557 note 1} In this, and as regards other points in this paper, I am indebted for advice and assistance to Professor Takakusu, of Tokyo. Respecting the reading of the letters, etc., see also the Dictionary of Buddhism, Bukkyo Dai-ji-ten, compiled by Oda Tokuno, and revised by Takakusu, Nanjio, and others.

^{page 558 note 1} Major-GeneralForlong, J. G. R., Rivers of Life or Faiths of Man, London, 1883, vol. i, p. 235.Google Scholar

^{page 559 note 1} ProfessorMacdonell, A. A., History of Sanscrit Literature, p. 229.Google Scholar

^{page 559 note 2} Ibid., p. 422.

^{page 560 note 1} DrZeller, E., History of Greek Philosophy, trans. Alleyne, Introduction, p. 34.Google Scholar

^{page 560 note 2} ProfessorBurnet, J., Early Greek Philosophy, 2nd ed., p. 56, n. 1, and p. 263.Google Scholar

^{page 560 note 3} Burnet, , op. cit., loc. cit., and p. 389Google Scholar; Zeller, , vol. ii, p. 127.Google Scholar

^{page 561 note 1} Burnet, , p. 340.Google Scholar

^{page 561 note 2} Zeller, , op. cit., i, p. 318Google Scholar, note; see also Burnet, , pp. 335–6.Google Scholar

^{page 561 note 3} Zeller, , i, pp. 436–7.Google Scholar

^{page 561 note 4} Burnet, , p. 342.Google Scholar

^{page 563 note 1} Garbe, R., The Philosophy of Ancient India, Chicago, 1899, 2nd ed., pp. 44–6Google Scholar; Die Sāṃkya Philosophie, 1894, pp. 94–6.Google Scholar

^{page 563 note 2} The Philosophy of Ancient India, p. 43.Google Scholar

^{page 563 note 3} On the migrations of the Greek Philosophy to India, see article by M. M. Satis Chandra Vidyabhusana in JRAS., 10, 1918, pp. 486–8.Google Scholar

^{page 564 note 1} See Fergusson, , History of Indian and Eastern Architecture, revised and edited as to Indian Architecture by Burgess, 1910 ed., pp. 274–5.Google Scholar

^{page 565 note 1} To the same effect see also Hodgson, 's Essays on the Language, Literature, and Religion of Nepal and Tibet, 1874 ed., pp. 30 and 136Google Scholar, note; also Oldfield, H. A., Sketches from Nipal, pp. 253–4 el alibi.Google Scholar

^{page 566 note 1} Examples of votive Stūpas from Bodh Gaya exist in other museums in this country, but invariably, it would seem, having the apex broken off and lost.

It will be noticed that in Fig. 5 the cross-piece, immediately below where the apex has been broken off, takes the form of the āmalaka. This is stated by E. B. Havell (Ancient and Medieval Architecture of India, 1915, p. 63Google Scholar) to be the fruit of the blue nymphæa, which is Vishnu's ensign; and at p. 97 of the same work he gives the figure of a miniature shrine of Vishnu, found at Sarnath, crowned by the āmalaka; on which he remarks that “the finial, water-pot, or kalasha, is missing in most cases”. The appearance of the āmalaka in our illustration (5) of a Buddhist Stūpa from Bodh Gaya is not surprising, where monuments of different cults are to be found in close proximity. Further, Havell remarks that Buddhism seems to have gradually adopted the Vishnu symbol as a distinctive mark of the Mahāyāna school.

As regards the crescent form, both in the ordinary and in the elemental Stūpa, a suggestion may be ventured that, not impossibly, its occurrence may be connected with Saivism. The Stūpa was one of the early symbols of Saivism: and the crescent in the head of Siva may have in this way found a place in the finial of the ordinary Stūpa. If this were so, the section of the sphere in the elemental Stūpa would thus have been influenced to become a crescent, in place, as suggested in this paper, of the occurrence of the crescent in the later ordinary and miniature or votive Stūpas being due to its previous occurrence in the elemental Stūpa as a variation of the section of the sphere. The Japanese Dictionary of Buddhism, Bukkyō Dai-ji-ten, already quoted in note 1, p. 557, gives, however, the section of the sphere as the true representation of this portion of the sotoba.

^{page 567 note 1} Professor H. A. Giles informs me that the figure is of common occurrence in China apart from any Buddhist surroundings.

As regards the occurrence of the Chinese Moral Series in the Buddhist picture, it may be remarked that Nukariya Kaiten, Religion of the Samurai, p. 230Google Scholar, points out that the Five Constant Virtues of Confucianism are intrinsically similar to the Five Buddhist Precepts.

^{page 568 note 1} An interesting parallel to the assignation of Dhyāni Buddhas to the several forms of the elements is to be found in the assignation of Deities to the elements by the Greeks. The usual arrangement is to attach the name of Hera to Earth, of Nestis, said to have been a Sicilian water-goddess, to Water, of Zeus to Fire, and of Idoneus to Air. Nestis is always associated with water, but there is some variation in the distribution of the names of the other three deities among the other three elements. (Burnet, , Early Greek Philosophy, pp. 264–5.)Google Scholar

Originally the conception of the several elements as consisting of geometrical forms was, clearly, by way of a scientific explanation of the nature of those elements; the use of divine names as applied to the elements was, as indicated by Burnet, no doubt of the nature of a poetical fancy. In a similar manner it appears safe to assume that the several forms in the elemental Stūpa were taken, in India, to be the actual representation of the several elements; the assignation of a Dhyāni Buddha to each of those forms was a later fanciful conception.

^{page 569 note 1} Each of the five Dhyāni Buddhas, mentioned above, has as his symbol a letter, as have the Buddhas and Bodhisattvas generally. Thus:—

aḥ, Divyadundubhimeghanirghosa, Japanese Ten-ku-rai-on-Butsu, otherwise Amoghasiddhi, Japanese Fukūjōjyu, a form of Sākyamuni.

hrīḥ, Amitābha.

trāḥ, Ratnasambhava.

huṁ, Aksobhya.

va , Mahāvairoćana.

These symbolic letters, as explained by Monier-Williams, “form the essential part of (their) mantras.”

^{page 569 note 2} Prabhūtarātna, a former Buddha, in the Saddharma Pundarīka Sūtra. See Kern's translation, chap, xi, in SBE., vol. xxi. See also Nanjio, , Twelve Japanese Buddhist Sects, p. 135Google Scholar. With the doctrine here compare that of the Shingon sect (Nanjio, as above, pp. 78, 99, 101, and 102). See also Okakura, K., Ideals of the East, pp. 128 et seq.Google Scholar

^{page 570 note 1} An example of the tarak-tō is given in the illustration (Fig. 9) opposite this page, after a wooden sotoba on a grave at the village of Negishi, near Yokohama.

On the base of a sotoba, in stone, in the grounds of the Zen-shyu monastery, Sho-fuku-ji, near Hyogo, are inscribed the four symbolic letters hrīḥ, trāḥ, huṁ, aḥ, representing four of the Dhyāni Buddhas, as in note 1, p. 569.

^{page 570 note 2} Professor Takakusu furnishes me with three mantras sacred to Dainichi (Mahāvairoćana), the superior, middle, and inferior. The one quoted in the text, a-vi-ra-huṁ-khaṁ, is the middle one. The superior mantra in full is: Namaḥ samanta buddhānām, a-vaṁ-raṁ-haṁ-khaṁ, svāhā!, and has the same purport as the first-named one. The inferior mantra is a-ra-pa-ća-na, being the mystic collective name for the Five Dhyāni Buddhas, each being represented by a letter. (See also Monier-Williams' Dictionary, sub voce Arapaćana.) The first two only of these mantras have reference to the five elements. An example, from Nepāl, of the use of the name forming the third mantra will be found in the Journ. Roy. Asiat. Soc., Oct. 1916, p. 735.

^{page 572 note 1} The analogy is striking between this doctrine and the teaching in the Upanishads, where Brahman is said to dwell in the water, fire, ether, wind, sun, moon, and stars, the regions of the earth, etc., and that the same is the inner self of man. (See Garbe, , Philosophy of Ancient India, pp. 70 et seq.Google Scholar) If we compare the notion, such as was in vogue among the Greek philosophers, of Space, or the Fifth Element, “which embraces all the others,” “permeating” the four elements, or being “inhaled” by them, we may possibly have a reason of the Fifth Element being selected as the representative of Mahāvairoćana, “whose constituents are the five.”

^{page 572 note 2} The form given in Fig. 3 was probably kept in the temple there referred to as a model for those to be placed on graves.

The letters in Fig. 9 are represented, as nearly as possible, as they stood on the sotoba.

Additional Note.—The form which constitutes the apex of the sotoba is often, in Japan, termed the Nyo-i. The Nyo-i is properly the staff in the hands of the Buddhist priesthood,—originally, probably, a club for defence, or offence; and then a mace. If the flame-like terminal of the apex of the sotoba is bent over towards one side, the apex assumes a certain resemblance to this mace or staff. The term Nyo-i () means “according to desire”,—“as you wish”.