A Chinese Buddhist Water Vessel and Its Indian Prototype

•期刊原文

A Chinese Buddhist Water Vessel and Its Indian Prototype

Ananda K. Coomaraswamy & Francis Stewart Dershaw

Artibus Asiae

MCMxxviii/xxiv No.2/3

pp.123-143

--------------------------------------------------------------------------------

P.122

LATE IN THE VIITH CENTURY A.D., I CHING, ONE OF THE SEVERAL CHINESE pilgrims to the holy places of Buddhism in India, observed in use among the Indian hik.sus a utensil for drinking water which he describes as follows: "The vessel should be made so that the tip always connects with [i.e.is permanent- y fixed to, or is contimuous with] the mouth [of the vessel]; the tip is pointed above a flange andshould be two fingers [i.e. finger-breadths] high; in it a hole as small as a bronze chopstick is made. "Fresh water should be kept in such a vessel. At the side of the vessel is another round opening; it has a pipe [i.e. funnel] rising two fingers high and is as large as a small 'cash'. This opening is used for pouring water in; two or three sheng may be put in. A small vessel is not useful."

He has already described the material of which the essel is made as a hard earthenware **; and a litter *** indicates the method of drinking from it in these words "..to hold the tip in the mouth and pour water, is improper, and (makes it) difficult to distinguish the clean from the touched."****
────────────
*See I-Tsing: A Record of the Buddhist Religion as Practised in India and the Malay Archipelago, Oxford, 1896, p.28 for Dr. Takakasu's translation of the passage.

The Chinese text as more closely ranslated above seems comparatively clear, though Dr. Takakasu recognizes the vessel as a ku.n.di. Cf. Daizookyoo, (Chinese Buddhist Tripi.taka), Kyooto, 1905; XXIX, 10 p.95 recto. Genmyoo Ono. Bukkyoo ijutsu Gairon, s.v; sobyoo, rightly distinguishes ku.n.dikaa, a vessel with two openings, from kala'sa, a vessel with only one. ** Op. Cit. P.27, Chapter VI, line 3. *** Op. Cit., p.28, line 19. ****Cf. Sutta Vibhanga, Paatimokkha (S.B.E., Vol. XIII, p.65), " Not with a hand soiled with food will I take hold of the waterjar." I Ching, however, has in mind the rule that remains of food or of drink, are "touched", -that is, unfit for consumption - and is implying a method of drinking from a ku.n.di which leaves both the contents and the tip of the vessel literally untouched and therefore ncontaminated. It is perhaps significant that ku.n.da means not only a vessel, but also a well, spring, or especially a sacred source, and that one who drinks from a ku.n.da of earthenware on the manner here described still actually drinks from a flowing stream. Can we infer that it had once been a rule of forest- dwelling sages to drink only from running water?

P.123

Evidently we have here a vessel of the form illustrated in Figs. 1 to 9. It is, as Dr. Takakasu glossed it, a ku.n.di (ku.n.dikaa), an essential utensil in the proper equipment of a Buddhist monk when I Ching was in India. Its peculiarities are hat it has a funnel on the shoulder through which it can be filled, and that the neck or handle ends in a tip or nipple through which it may be emptied. It was devised to hold drinking water which was to be consumed with due regard to cleanliness in the fashion exaggerated in Fig. 2.If we may judge from I Ching's habit of referring to customs in China which resemble or differ from those he found prevalent among bhik.sus in India, this form of combined water container and drinking vessel as new to him; for he makes no reference to anything similar in his native land. His description of the vessel, written for Chinese readers, indicates that he believed it equally

P.124

strange to them. We may guess, accordingly, that this form of vessel was not known in China previous to 671 A.D., the year when I Ching began his pious pilgrimage to India; and we may certainly infer that when he wrote the description (probably in the year 692 A.D.) he believed that he was imparing information.
Testimony of the same negative character may be derived from two other sources, -Chinese representa- tions of the equipment of monks, and actual drinking vessels of the kind known, or supposed, to behinese, Japanese, or Korean in origin. No proper ku.n.di appears in any Chinese carving or painting the date of which is earlier than T'ang. Very robably the kind of water bottle used by Chinese monks of pre-T'ang times is shown in a scene carved on a stele dated 551 A.D.* it is a flask-shapedbottle, with-out funnel or tip, and near it is hanging what may be the carrying bag prescribed for t in the Vinaya rule as indicated by I ching. ** Its shape is precisely like the pao p'ing (to be mentioned later) carried so often by Bodhisattvas, occasionally by bhik.sus***, and in a few instances by attendants who appear to be laymen****. Similar negative testimony is yielded by the monk's drinking vessels found or reported in China, Japan, or Korea: none of them antedate the VIII the century. The earliest one of which the date can be fairly ascertained is in the Shoosooin. It isigured on plate 262 of the six volume edition (in English) of the Toyei Shukoo (1909), and on plate 168 of the three volume revised edition (also in nglish) published in 1910), In both editions the vessel is called "jug made of white copper". It does not appear on the list of the original donation to Todaiji made on the 21st day of the 6th month, 756 .D., nor in avaliable lists of the giftssubsequently deposited in the Shoosooin. The last of these was made in 758 A.D. it will be remembered that the shoosooin, the storehouse of Todaiji, ontained some objects belonging to the temple when the first Inperial gift was dopposited there. The "jug of white copper" may be one of earlier lot; quite as likely it is one of the numerous unlisted objects given latter, but
────────────
*See Siren, O.: Chinese Sculpture from the fifth to the fourteenth Century, 1925; Vol. II, Pl. 171. **See I-Tsing: op. Cit., p.28f.*** See M.F.A. Bulletin, Vol. XXIV, No. 141, and Siren, op. Cit. Vol. II, Pl 198; Vol. III, Pls. 295, 298, 319, 323, 327. ****See M.F.A.Bulletin, Vol. XIII, No. 78, Fig. . Here the flaskbearers are in attendance upon the four chief donors of the stele. See also Siren: op. Cit., Vol. II, Pls. 144 and 172: Vol. III, Pl. 317(the stone coffer, M. F. A. 16. 287).

P.125

whenever it was housed in the Shoosooin, there is little room to doubt that it has been there ever since 758 A.D.

Two similar bottles* are among the treasures of Hooryuuji. One of them is very like the Shoosooin jug", having a similar human mask at the base of the funnel. The tip is horizontally striated like the tips of Figs. 2,3 and 4. Neither the history nor the dare of this specimen is known, but we mayreasonably group it with the Shoosooin "jug" and the other Hooryuuji, bottle. The latter, which has a hinged cover on the funnel, is "possibly referred to in a n inventory of the Tempyo period" (722-748 .D.) and may therefore be a little older than the Shoosooin "jug". A third example, said to have once belonged to Hooryuuji, is in the collection of Baron Masuda***.in this specimen, the tip has been shortened and the cover of the funnel is lost. Examples of the monk's drinking vessel may be seen in two painting widely separared in time. The earlier **** is a portrait of the Chinese priest known in Japan as Ji-on Daishi. On his right stands a ku.n.di with hexagonal tip and covered funnel. As the painting is supposed to date from the Heian time (from very late VIII□ through half of the IX□ entury), it is the earliest instance yet found which shows our drinking vessel as part of the proper equipment of a monk.

The other painting is one the treasure of ooshodajii.+ It depicts the life of another Chinese priest, Kanjin, as the Japanese call him, who reached Japan in 759 A.D. One of the scenes shows him ready to embark with several disciples, one of whom carries a ku.n.di. Since this painting dates rom the Kamakura period, it indicates that the tradition of the ku.n.di as a necessary utensil in the equipment of a T'sang Buddhist monk, was alive in Japan at the end of the VIII□ century. vidently the vessels depicted in these paintings treasured in the Yakushiji and the Tooshodaiji espectively, are to be regarded as Chinese. The origin of the bronze ku.n.di remaining in Japan is not so certain. Though they were perhaps made in Japan, the
────────────
* See Nara Hooryuji Okagami, part XXXIII, 7th plate, and Part LV, 11th plate. ** See op. cit., Part LV, text. *** See Catalogue of an exhibition of a portion of the Masuda Collection, held in Tokyoo in 1926, no pl., n.d. * See Yakushiji Okagami, PartIII, 1st and 4th plates.+ See Toshodaiji Ookagami, Part VI, 18th plate. In this reproduction, the unnel of the ku.n.di is not unmistakably in evidence but there is no doubt about the kind of bottle the painter intended to draw.

P.126

model for them came along with Buddhism from the Continent of Asia. What the model was like may be gathered from the Chinese specimens which follow. he pottery vessel outlined in Fig. I is probably contemporary with the Shoosooin "jug", and is certainly Chinese. It is of the fine, white, low-fired clay covered with green lead glaze, amiliar in Chinese pottery of the T'ang Dynasty. The absorbent property of the quite unvitrified clay and the fact that the interior was not glazed, indicate that this vessel was not made to be practicable but rather for some symbolic use onnected with burial of the dead. It had abundant traces of the earth of burial when it came into the possession of the Museum of Fine Arts.*Fig.3 represents a thoroughly practicable vessel hich is in the Freer Gallery at Washington. It is slightly damaged, the mouth of the funnel on the shoulder being broken ington. It is slightly damaged, the mouth of the funnel on the shoulder being broken off. It is made of a white clay, high-fired, impervious to water, covered with a thin, transparent, highly vitreous glaze of very faint tea-color.

Probably it is not much later in date than the VIII □ century, and in details of its design is significantly similar to Fig. 2: the setting of the neck (which is also the handle of these vessels) upon the body, is alike in both vessels, there are faint transverse gouges on the tip of the Freer specimen of the kind strongly arked on the tip of the Boston vessel, the extremities of the tips are alike, and the flaring bases are of the same model.

Fig. 4 is in the Peters Collection, New York. It has been ascribed to the Sung Dynasty and described as f " white porcellanous ware with creamy glaze"**. In other words, it is a usable vessel like Fig. 3, of later date and clearly different in the modelling of the fot, of the extremity of the tip, and of the flange. Fig.5 represents a Chinese drawing published as Ch. 00145, P1.XCVII, VOL.IV of Stein's Serindia. In Vol. I, p.967 of that work, it is mistakenly described as a "stoppered vase with ovoid body, and narrow neck, spreading mouth and foot, and applique handle or ear on shoulder". Obviously it is another of our drinking vessels, unique among them in the construction of the foot, the proportions of the ncek and
───────────
*M.F.A. 26.62. ** See Met. Mus. Of Art: Catalogue of an Exhibition of Early Chinese Pottery and Sculpture. By S.C. Bosch reitz; N.Y.. 1916, p.9, No.33. The word "creamy" as used here, is a color designation: in reality this glaze is thin, transparent, weak-tea colored, showing warm white over the white clay of the body.

P.127

of the flange, and in the curious rendering of the funnel at the shoulder. In the original drawing, it stands at the proper right of a Chinese Buddhist monk,* thus reminding us that I Ching found it among the few things essential for a bhik.su to possess. Since that drawing was made earlier than 1035 A.D., ** we may assume the X□ or the very early XI□ entury as the date of of the flange, and in thecurious rendering of the funnel at the shoulder. In the the vessel it of the flange, and in the curious rendering of the funnel at the shoulder. In the represents.

In the His Ch'ing Ku Chien ***, chuan XVIII, p.16 recto, is engraved a ku.n.di of proper form, with lidded funnel, labelled T'ang su p'ing 唐素瓶 or unornamented p' ing of the T'ang Dynasty. On pages 17,18 and 19 three other water jars are grouped as T'ang su p'ing. Each has a funnel with hinged cover, ut neither flange nor tip. A similar tip-less jar with lidded funnel is ascribed to the Han Dynasty (sic) by the compilers of the Hsi Ch'ing Hsu Chien**** and juxtaposed to two precisely similarvessels which they label T'ang. These p'ing and the one like them in the Ning Shou Chien Ku+, are probably imperfect ku.n.di: the lidded funnel on the shoulder of each is a useless adjunct to a jar with a mouth practicable for receiving liquids as well as pouring them out.

With these examples from China, we must be satisfied; for with one exception －－－ it is of pottery and like Fig.1 except in color of the glaze －－ no more are available. On the other hand, examples from Korea are comparatively abundant. Theyare represented in Figs. 6,7 and 8. Figs. 6++ and 7+++ are of the usual high-fired gray clay covered with the thin, translucent, green-toned glaze, recognizable as typical of the pottery made under the dominance of the Korai Kingdom. The extremity of the tip in Fig. 6, is a diminished version of the shape observable in Chinese vessels of the kind; but the tip itself, like that of Fig. 8, is octagonal in orizontal section. The tip of Fig.7 is not significant －－－ it is a metal replacement of the roken original －－－ but the funnel on the shoulder offers an element not previously observed on pottery ku.n.di, though familiar on those of bronze treasured in Japan; it is closed with a inged cap. Such a cap is often found on Korean vessels of this kind, especially on those made of bronze. Fig.8++++ is typical of the latter. They occur in a variety of sizes,
───────────
*Cf. Yakushiji Ookagami, loc. Cit. **See Stein, op. it., Vol.II,p827. *** 西清古鑑 **** 西清續鑑 Chuan VIII, pp.30,35,36.+ 寧壽鑑古 Chuan VII, p. 30, recto. ++Museum of Fine Arts, 27. 328.+++ Freer Gallery of Art, Washington, D.C. ++++ Prince Li's Museum, Seoul.

p.128

some of them so diminutive that they must have been intended for symbols only. In date, these Korean rinking vessels are all ascribed to the Korai Period (936-1399 A.D.). They have all been found in graves of that period, and since none of later date have been turned up, we may infer that none were in ater use. This inference applies equally to China: all the known specimens are reasonably ascribed to the three and a half centuries beginning with the III □. In Japan, the limit of date is even narrower. In none of these countries does it appear that this form of drinking vessel was a common domestic utensil; on the contrary, it seems to have een employed exclusively in the ritual of burial, and by the religious who followed strict Buddhist practice. Whether or not it was customarily used in temple ceremonies is quite uncertain.

The names by which the vessel has been designated in Japan and China do not tend to clarify the question f its use in either country or in Korea. I Ching calls its Indian prototype by the unspecific term p'ing 瓶, that is, vase, jug, jar, bottle; but he is very particular in describing it and its proper useby Buddhist monks (bhik.sus) in their strictly prescribed daily lives. The vessel itself is unfamiliar to the several Chinese to whom it has been referred; they unite in calling it bu 壺, which is even less descriptive than p'ing and not at all in conformity with the classification employed in the Chinese books on bronzes. As we have seen, thecopmilers of the Toyei Shukoo call it a "jug". One Korean specimen* is called simply 水瓶 shui p'ing(Japanese, sui bin), or water jar, and sui bin is the name given to the specimens treasured at Hooryuuji** and in the Masuda Collection. We get no further light by turning the pages of the Hsi Ch'ing Ku Chien which illustrate the varieties of p'ing,nor by referring to the confusing types classified as p'ing in the His Ch'ing Hsu chien and the Ning Shou Chien Ku. From all these references the inference is clear that the nature of the p'ing or ku.n.di which I Ching describes, and the use of it, have long been unknown or forgotten in China and Japan. To this unfamiliarity confusion has been added. In the His Ch'ing Ku Chien***, the monk's drinking vessel is explained as a kan lu p'ing 甘露瓶, or sweet dew vase,
───────────
* See Chosen Koseki Zufu, Vol. VII, P.1. 859, Fig. 3145. ** See Nara Hooryuuji Ookagami, loc. Cit. ***Chuan XVIII, p.16 verso.

P.129

used for offerings to Buddha. Our Fig. 8, which is No. 176 in the Catalogue of Prince Li's Museum at Seoul, is there called hsien chan p'ing 仙盞瓶 (Japanese, sen san bin), which may be interpreted as jar for ambrosia* and in that sense as equivalent to kan lu p'ing. It happens that No. 175 in that Catalogue is also called hsin chan p'ing; in reality, it is a proper kan lu p'ing and should be so designated. Evidently these designations, hsien chan and kan lu, have been applied to korean and Chinese versio ns of the Buddhist monk;s combined water-bottle and drinking vessel, owing to their resemblance to what, for instance, Foucher had in mind speaking of "la 'gourde d'immortalite' des Chinois."** To clear up this confusion, we must go back to the writings of another Chinese pilgrim.*** In his account of what he saw at Bodhgayaa, Hsuan sang refers to the finial crowning the temple to the east of the Bodhi-tree as a "gilt copper" (orbronze) o-mo-lo-kia-kuo (or o-mo-lo-kia kuo) 阿摩落迦果; the Japanese amarakak-wa (or amaraka kwa) more nearly represents the T'ang pronunciation. Before discussing this observation (of which rather confusing interpretations have been given by Beal and by Watters ), let us first remark that the normal crowning elements of the tower ('sikhar) of a northern Indian (naagara) shrine are (I) a ibbed stone++ called aamalaka or aamalaka-'silaa, myrobalanstone" and (2) above this an ambrosia vessel variously called am.rta-kala'sa, -gha.ta, puur.na-kala'sa, etc.; innumerable examples exist of the kind illustrated in Fig. 24. Now Hsuan Tsan's text embodies two early glosses, one pao p'ing 寶瓶, "precious vessel", "sweet dew vase", etc.+++ (an object already familiar to the Chinese as an attribute held
───────────
*"Jar for ambrosia", "nectar vase", "immortal dish",
"immortal's cup", etc., are secondary meaning. Primarily, sen san bin means mountain man's (that s, hermit's or recluse's) cup or drinking vessel. Now the "mountain man" is one who, through enchantment or other faery process, has received the gift of extraordinary longevity. He is an "immortal" in the Taoist sense, whose sustaining drink is "sweet" or "heavenly" dew (also, in China, a Taoist conception originally). Buddhist "immortals" are another kind o f being, inherently divine. As a name for a Buddhist utensil, sen san bin seems inept. **

L'Art Greco-Bouddhique du Gandbara, 1905; p.96, note (2). ***See Beal, S., Buddhist Records of the Western World, London, n. dl, Vol. II, p.118, and p.136, note 2.+ Beal, S., Buddhist Records of the Western World (Si yu ki), II, p.137; Watters, On Yuan Chwang,II, p.49. ++ Aamalaka, the fruit of the myrobalan, Emblica officinalis, Gaertn.

Burgess has suggested Amala-'silaa, "pure stone", as an emendation (loc. Cit. Infra. ) . In the A'sokaavadaana, where A'soka makes a donation of a half-aamalaka fruit, the word is transliterated an-lo, an-lo-mo-lei, or an-mo-lei. +++ Giles (Chinese Dictionary) perhaps had this gloss in mind when he explained pao p'ing as " the Buddhist name for aamalaka fruit".

P.130

by various divinities); the other, pao t'ai 賽臺, "precious pedestal", or "stand". These two glosses reflect the fact that Hsuan Tsang's word or words have been interpreted in two ways, (1) as a transliteration of Sanskrit amara-karka, "immortals' vessel"* and (2) as a transliteration of aamalaka plus the Chinese kuo, "fruit". Hsuan Tsang's description " gilt copper" would suggest that the first interpretation is the correct one; for while the am.rta-kala'sa may be of earthenware, metal, or stone, the aama laka is always of brick or stone, generally, as the full name aamalaka-'silaa implies, the latter. But, as Professor pelliot has pointed out to us, kuo is a purely Chinese word never usedin transliteration, and therefore the second interpretation must be accepted, although it gives no sense. In either case does Hsuan Tsang's observation have any bearing on the problem of our drinking vessel; for neither the Indian am.rta-kala'sa nor the Chinese pao p'ing are drinking vessels,
───────────
* The form amara-karka is possible (cf. amara-sarit, the Ganges), but karka in the sense of water vessel used by Brahmanical ascetics occurs in the ahabhaarata; karkata is found in the Pa~ncatantra.
Burgess (imperial gazetteer, II, p.181) gives karaka s a synonym of kala'sa. The only Indian paralled for "sweet dew vessel" is the var.sa-sthala,
"rain-vessel", by which name the finial of a stuupa s designated in the Divyavadaana.The Chinese regarded dew as the drink of the Immortals and thus
the phrase kk an lu, "sweet dew", was adopted by hinese Buddhists as the equivalent of the Sanskrit am.rta. the characteristically Buddhist term dhamma-karaka occurs frequently in Pali texts, where it seems to have the general significance of a "vessel authorised to be used by bhik.sus for water". In Cullavagga V,13, Dhammapada Attbhaathaa XVI, 9 = Burlingame III, 290, and Faataka, I, 395 it is a vessel used for strained water, and perhaps fitted with a strainer larger than an ordinary parissavana. In Cullavagga, VI, 21 and XII, 2 dhamma-karaka are listed amongst osther parikkhaaras (requisite properties of a bhik.su) without any ndication of the exact meaning (almsbowls and water-strainers are separately mentioned) . In Faataka VI, 331 we have a simile " as easily as water flow from a dhamma-karaka"; here the meaning ku.n.dikaa is possible. The dhamma-karaka of Milindapa~nha, 68, is more mysterious, as the water s said to be supported by the atmosphere; but this becomes comprehensible if we suppose that Naagasena
took a ku.n.dikaa full of water, stopped the funel with his thumb, and held it upside down, in which case the water would not flow out of the nipple, and ould be in fact supported by the pressure of the atmosphere, as in a siphon. Here, then, there is good reason to think that dharmmakaraka = ku.n.dikaa.

* In Mahaava^msa, XXII, 68, monks attending a festival where they receive rice-milk in their bowls, are expected to lay aside their dhamma-karakas before entering the hall where the meal is served. These "regulation vessels" are here described as ekachattayutam, literally, "having (only) one umbrella", i. e., one handle broadening above, thus, a vessel with one opening, not a ku.n.dikaa: but ekachattayutam may be otherwise interpreted, not as qualifying dhamma-karaka, but as meaning " and their one-handled umbrellas", in which case we learn nothing about the shape of the dhamma-karakas.

Karaka-.tanks of Cullavagga, V, 16, 2, literally, pot-ridge", is probably a "Persian" water-wheel, i.e., with pots tied on to it, raising water as it revolves. Karaka-vassa, said to mean "hail", looks more like "cloud-burst".

P.131

nor does either resemble our drinking vessel in form. They lack, indeed, one essential of the latter, viz., the funnel on the shoulder through which it can be filled; and this lack is perfectly ogical, for the puur.na-kala'sa and pao p'ing are in themselves a never-failing fount of blessing, and never need filling. Some further remarks on the am.rta-kala'sa are postponed to the close of this essay. Confusion of the two vessels arose in the Far Eastern countries, when, after many centuries of oblivion, specimens of the drinking utensil were ecovered, classed with the familiar vessel they most resembled, and called by its name. It is possible that in this formal classification and nomenclature there is an idea that our drinking vessel had a ritual use as a pao p'ing for dispensing lustral water (hardly "sweet dew" ) to the faithful. But that remains to be proved.

It is clear that there is no more descriptive hinese or Japanese name for our drinking vessel than p'ing or bin. Judging by the known specimens inapan or from China and Korea, this p'ing had some vogue in those countries (probably altogeher among the religious of Buddhism) within the period stretching from the VIII□ century into the XIV□, and within that period only. Apparently I Chingrought news of it and its proper use from India to China. In India, then, and not in China, or Japan, or Korea, and c ertainly not through the medium of Hsuan Tsang, must we seek for earlier use of the vesse.

That the same type occurs already in India at a much earlier date has not hitherto been generally ecognized. We shall therefore in the paragraphs immediately following, cite the Indian examples, whole or fragmentary, which have been found on Buddhist sites. Although fragments of our type are commonly found on Buddhist sites in India, only three perfect examples are known. One of these, the original of Fig. 10, was excavated at Sahe.th-Mahe.th, from the foundations of structure D, which are older than those of structures B and C,which are assigned to the first or second century A.D.; thus the present example may be called early- or pre-ku.saana. It is described (by Marshall, in A.S.I., A.R., 1910-11, p.6, with a good reproduction on PL.III, 8) as follows: "The goblet is I'high and s provided with a tall neck pierced with a minute hole for the admission of air and with a spout at the side", and it is added in a footnote that " vessels of this

P.132

shape are found in abundance on the sites of Buddhist monasteries", and "This may have been the ordinary drinking bottle of the monks". The so-called spout is, of course, the filling place, and the neck with its narrow opening is effectively the spout. Another perfect example, fig. 11, found at Bhii.taa, is of Ku.saana date. It is described in A.S.I., A.R., 1911 － 12 p.83, no.52, as follows: " Water-bottle, 8" high － the life-breathing vessel f the Chinese pilgrims(?). High top with projegting rim, pierced wi th small hole. Short plain spout turned upwards. * Round bottom. Fine buff clay with red paint." It is also stated that there is a similar jar, also complete, in the Lucknow Museum; but no illustration of it is available. The oldest Indian example is a fragment from the Bhi.r mound, Taxila, of early Maurya and pre-maurya date. Only the top, showing the handle (mouth) and funnel is preserved the illustration (Fig.12) offers a quite tentative restoration. The height of the part preserved is five and a half inches; the fragment is described as follows in A.S.I., A.R., 1911 － 12pp.41 － 42:" Top of earthen jar. Elongated neck pierced with a minute hole; upturned spout
───────────
* Here, too, the so-called spout is really the funnel. The reference to the "life-breathing vessel f the Chinese pilgrims" is meaningless. I Ching mentions the use of a life-preserving vessel fang sheng chi 放生器, but this was a jar into which the insects caught in the strainer were released, so that their lives might be preserved

P.133

on one side, with small perforations below. Traces of red wash." Numerous necks of similar earthen ku.n.dikaas have been found at Saa~ncii (Hamid, M., et al., Cat. Mus. Arch. At Sanchi, 1922, pp.52－53). The rim of flange at the top of the neck "is invariably surmounted by a tapering nipple which in ome cased is plain and in others ornamented with round markings"; there is usually a red color wash and glaze; in some cases it is evident that the neck was separately made. The term suraahi (by which name vess els with a long neck, but no funnel at the side, and used for carrying Ganges water, are designated) is incorrect. The dates of the Saa~ncii examples range from the third century B.C. to the first century A.D.

A fragmentary example from the Raamatiirtham site is illustrated in A.S.i., A.R., 1910－11, PL. XLIV, 56. An example from Chaarsada (Gandhaara), dating from the third or fourth century A.D. is illustrated in Fig. 14; this is described in A.S.I., A.R.,1902－03, p.182, as the "top of a lidless jug", and it is emarked that the form appears to be based on a metal prototype; on the Plate facing p.182, ih., (Fig. 24, No.32), this object is misnumbered 33. Another specimen, found at Basaa.rh and probably of Gupta date, is illustrated in Fig. 13 it is described in A.S.I., A.R.,1903 － 04, p.95, as "a straight handle, + 3/4 long, glaxed in a red brown tint".Of Gupta or earlier date are the many fragments recovered form the site of the Buddhist monastery at Sa^nkaram, published in A.S.I., A.R., 1907 － 08, p.177, PI. LVII, figs. 6 － 12 and perhaps 24. They are incorrectly described as umbrellas of votive daagabas. The fragments range in size from 3" to 7 /2" in height, and were collected to a total number of 646, a quantity quite intelligible as epresenting an accumulation of broken drinking vessels during a monastic occupation of several centuries.

Actually, few Indian vessels of this type have been met with in bronze: the best example (early mediaeval) is a "Buddhist monk's bottle of bronze" now in the Saarnaath Museum,illustated in A.S.I., A.R., 1924－25, P1. XXXVI, b ( see accompanying illustration). It may

P.134

be assumed that this vessel once possessed the characteristic filling hole, as there is a large break at one side in the position which it may have occupied. No evidence seems to exist for the actual use of our ku.n.dikaa, as described above, after the mediaeval period. If, as seems probable, it went out of use, this can hardly be ascribed to the decline of Buddhism, for it is not used by Buddhist monks in Ceylon at the present day, nor so far as I know, in any Buddhist country*. In Ceylon, the ordinary monk's drinking vessel (kotalaya) is the same as that in general use; it is of earthenware or metal, with a wide mouth through which it is filled, and an ordinar y spout at the side from which the water is poured into the mouth, and thus altogether nlike our ku.n.dikaa.** In Buddhist texts, the monastic drinking vessel or water bottle is usually called a ku.n.di (-kaa), but in one place (Faataka, II, 73), the work kama.n.dalu is used as a synonym. In Faataka II, 317, a thirsty king expects to find a u.n.dikaa in a Brahmanical hermit's (trida.n.dika) hut; *** ib., V, 390, Brahmanical hermits fetch rinking water in their ku.n.dikaas (which are distinguished from cooking vesels, ukkali), and use it to mix with their porridge: ib., I, 8 and 9, and II, 73, ku.n.dikaa is also a hermit's water-pot. In Dhammapada Atthakathaa, I, 92, a Buddhist monk offers to another elder water from his own ku.n.kikaa: ib., III, 448, a Brahman "sprinkled the ground with water from his kama.n.dalu". In one place, Culla-vagga, V, 9, bhik.sus are forbidden to carry their water-pots on their begging rounds: here the words used are tumba-ka.taaha, "gourd", and gha.ti-ka.taaha, "pot".

This injunction seems to be verlooked in Mrcchaka.tika, Act. V, verse 20 (third century A.D.), where a Buddhist monk app ears with his "staff, water-pot (ku.n.di) and begging bowl". In Brahmanical literature, Vedic, Epic, and Classic, kama.n.dalu and ku.n.daka or ku.n.dikaa are regular designations of water-vessels carried by asceticw or students. In the words
────────────
*With Brahmanical ascetics, likewise, the custom of carrying a special water-pot is now obsolete: some Puraa.nas and Sm.rtis even speak of it as forbiddenin the Kali age, see S.B.E., XIV. P.163, note 2. **
See illustration, Knox, R., Historical relation of Ceylon, ed. Glasgow, 1911, plate facing p.140; as nox remarks "when they drink, they touch not the Pot with their mouths, but hold it at a distance, and pour it in". *** In the Uttararaamacarita, Act. IV, verse 15(seventh century), King Janaka, on a vi sit to a hemitage, sprinkles the fainting Kau'salyaa with water from a kama.n.dalu. in the Buddhacarita, VII, 51 a Brahmanical ascetic is described as ku.n.dodahasta.h(v.e. ku.n.devahasta.h, etc) ; Bohtlingk's conjectural rendering "the back of whose hand was like a ku.n.da" is of course mistaken, the sense being "having in his hand a ku.n.da of water".

P.135

of Baudhaayana, Dharma Suutra, 1, 3, 5 (S.B.E., XIV, 160), Now (those who know the law) prescribe the carrying of a water-pot (kama.n.dalu)". So in Manu, VI, 32, the Brahmanical ascetic must carry almsbowl, staff and water-pot. This vessel was of arthenware,* as is shown by the facts that it was liable to be broken, and that it had to be purified by heating on a fire. The contained water was used not only for drinking, but also for personal ablutions, but not for oblations to the manes, gods, or fire. A straining cloth ( pavitra ) was another requisite. It may thus be assumed, and is at least probable, that our special form had been in use by ascetics long before the establishment of the
────────────
*Hence "gourd", while not always wrong, is not the best general meaning for kama.n.dalu; kaama.n.dalava^m is " the trade of a potter"; see lso above. The earlist representation of a hermit with a kama.n.dalu is probably the Uda Jaataka scene at Bharhut (Cunningham, Stupa of Bharhut, PL. XLVI, scene 2); but as the form of the vessel is not quite explicitly shown, we have not reproduced it.

P.136

Buddhist sa^ngha; it is in fact well known that Buddhist monasticism was modelled on already existing forms, and took over older customs, which were in fact Indian, and not specifically Buddhist. In saadhanaas descriptive of Brahmaa, the water-vessel is designated either as a kama.n.dalu (Matsya Puraa.na, Ch, 360, 40－44, and Vi.s.nudharmottara) , or as ku.n.dikaa (Agni Puraa.na, Ch, 49, 14－15). This brings us to the uestion of the use of the kama.n.dalu or ku.n.dikaa in iconography, as an attribute held by.r.sis and by various divinities.

As we might naturally expect, the vessel is thus carried only by divinities whose ascetic character is much in evidence, and is generally associated with the carrying of a rosary in the right hand, and the wearing of matted locks. It is thus especially characteristic for Brahmaa, * 'Siva, and Avalokite'svara, and not for Indra or Vi.s.nu; all the examples which we are able to illustrate are taken from representations of one or other of the first three divinities (Figs. 15-23). The vessel as an attribute is held in two different ways, each subject to some variation, (1) more or less as actually carried for use (Figs. 15-17 and 21-23), nd (2) according to the formal methods generally employed for the holding of attributes as such, regardless oftheir character or actual use (Figs. 18-20). As regards the form of the ku.n.dikaa as seen in iconographic usage, it may be remarked that he vessel is rarely decorated; it is sometimes with, and sometimes without a base; and that the true spout, the extension above the flange of the neck, is reduced to the dimensions of a short nipple, although occasionally (Fig. 17) a ringed and more elongated point is to be seen. This shortness of the nipple is also characteristic of the Indian examples of complete or fragmentary vessels already alluded to, the long point being found only in the case of metal examples or those which seem, like ig. 14, to reflect a metal prototype: while the extension of the true spout above the level of the flange of the neck is more pronounced in the Far Eastern types. Of considerable interest is the figure of a Guru or .R.si, probably Agastya, found in
────────────
* So also Braahmaanii may carry the rosary and ku.n.dikaa, see Avalon, Serpent Power, 1919, II, p, 21:or as Sarasvatii, likewise, see Vi.s.nudharmottara,III, 64.

P.137

Borneo*. Here the bearded Brahmanical ascetic carries a rosary in his right hand, and a kama.n.dalu or ku.n.dikaa in the left (see ccopmanying figure). The kama.n.dalu is h eld indeed, as though it were an am.rta flask, but the characteristic funnel is clearly indicated, and can be seen immediately below and in contact with the little finger. More curious is the flattening of the form, a peculiarity which has not been observed in any Indian or Javanese examples; and which, as Dr. Bosch ha justly remarked, belongs to a whole series of iconographic faults and misunderstandings characteristic of the Borneo sculptures. This peculiar form, flattened brought nearly to a point below, and held as described, may be regarded as affording a satisfactory explanation of certain flattened, pointed objects provided with a handle and suggesting a basket, purse, or padlock, held an attribute by some Chinese odhisattvas; examples from Lung Men, and one in the Gardner Museum, Boston, are illustrated by Siren, Chinese sculptur e, Pls. 90,91, 180, etc. The position of the fingers in the Borneo relief could easily have suggested the existence of a handle, while the flattened and pointed vessel, peculiar as it is, is beyond all doubt a kama.n.dalu or ku.n.dikaa. Mistakes andisunderstandings of the details of Indian iconography are by no means unfamiliar in Chinese uddhist art, and certainly the Chinese had no clear idea of the distinctions between an am.rta flask and a ku.n.dikaa when held as an attribute. Thus it would appear that the basket-like object held by he Chinese Bodhisattvas must be regarded as representing a ku.n.dikaa, or ascetic's water- bottle.**

There exists one late Vedic passage ('Satapatha Braahma.na, XIV, 1, 2, S.B.E., XLIV, pp.453, 454), describing the making of the Mahaaviira earthen pots used in the Pravargya ceremony, of interest in connection with our ku.n.dikaas. The Mahaavira pot s theoretically a drinking vessel: in it is prepared the hot draught of milk and ghii, and this is offered from it to
────────────
* Bosch, F.D.K. Oudbeden van Koetei, Oudheidkundig Verslag, 1925, P1. 31 b. ** Cf. an early Cambodian example of a ku.n.dikaa, held by 'Siva or Loke'svara, published in Arts et archeologie khmers, I, p.207 and fig. 55. For related Indian types see Gangoly, O.C.,The cult of Agastya, O.J.Myth. Soc. Bangalore, XVIL, 3, 1926 and in Ruupam, No.26, 1926.* In A rock-cut cave in Travancore, Modern Review, Oct. 1923, Mr.A.S. Ramanaatha Ayyar describes a igure of Agastya who "holds a jug-like vessel with an oval body, a longstoppered neck,and a short spout. The shape of this vessel is peculiar, unlike that of its modern counterpart, the gindi". Mr. Poerbatjaraka, citing this passage (Agastya in dem Archipel, 1926, p.105 ) adds the Javanese and old Javanese equivalents, ke.n.di and ku.n.di.

P.138

the A'svins as a drink (the sacrificer drinks the -remains, but not from the vessel itself). The Mahaaviira pot, it is true, has apparently no lateral funnel or other side opening by which it can be filled, so that the hole at the top must be usedboth for filling and pouring out: but the neck, etc., is described as exactly like that of our ku.n.dikaas, and the text provides both the names of the various parts and their proportions. The neck, through which there runs a hole of the size of a reed, is to be a span in length, contracted in the middle, where it is held: at the top it is to be drawn out to form a spout (mukha) of three thumbs' breadth in height above the flange (mekhalaa, "belt"), thereby providing it with a nose, So far as the neck is concerned, then, the form of our u.n.dikaa as identical with that of the Mahaaviira pot, and for this form a very considerable, certainly pre-Buddhist, antiquity is thus demonstrated.

It only remains for the sake of clarity to distinguish from the ki.n.dikaa another Indian essel which is carried as an attribute by various deities and, as already mentioned, is in architectural use as a finial. This am.rta kala'sa is a jar containing ambrosia, the water of life which is the source of the immortality of the gods. This am.rta is often used by Indra to restore the dead to life (e.g., in the 'Syaama Faataka); it is in fact Indra who most usually carries the vessel (Fig. 21) *; but it appropri ate to any deity in his capacity as an immortal. It is often carried by Naagas, and is characteristic for the Bodhisattva Maitreya (Figs. 26－28) who is thus distinguished from Avalokite'svara who properly carries the ku.n.dikaa); and even by 'Siva (Fig. 29). It is a simply jar or goblet with a short neck and wide mouth, often decorated, but always devoid of any pout or funnel: it is almost invariably held in the left hand in the manner shown in Figs.25－29,that is to say, with the neck between the first a nd second fingers and the thumb over the mouth.

The finial vase, so familiar a feature of Indian architecture, is of the same character, but the form and terminology are somewhat more varied: the simple form is illustrated in Fig. 24, and the various names include am.rta-kala'sa, -gha.ta, -kumbha, puur.na-kala'sa, etc., and, once only, var.sa sthaala. This jar, at first simple, later of the puur.na-kala'sa or -gha.ta type with vegetative (lotus) elements rising from the mouth and hanging over the rim, is
────────────
* When in Mahaava^msa, VII, 8. Indra carries a ku.n.dikaa, it is in his temporary disguise as a wandering ascetic (paribaajaka).

P.139

found not only as a finial, but forming the base or capital of columns and pilasters; the same type is also commonly met with as an auspicious symbol and as an offering vessel carried by worshippers. The greatest interest attaches to this form, which is the analogue of the brimming vases of early Mesopotamian art and, like these, "etait comme Le Saint Graal",* and, on the other hand, gives rise to all the later forms of the vase and plant motif,including those of Persian 'vase-carpets': but a detailed s tudy of this theme would lead us too far from the immediate subject of our essay. It need only be remarked that as a finial the vase (kala'sa, kumbha) occurs (1) in the simple form of Fig. 24, (2) with a very long projecting nipple, very often with an added symbol-banner (dhvaja), supported by a rod for which the hole in the neck forms the socket, and (3) as a puur.na-kala'sa or 'brimming vessel', with over-falling lotus foliage.
None of these forms have any connection with our ku.n.dikaa, which is of the form already described and illustrated, and is either (1) in actual use, the water bottle and drinking vessel used by ascetics, or (2) is carried as an attribute by Indian deities of an ascetic type, forming with the rosary and matted locks the insignia of an ascetic vocation.

LIST OF ILLUSTRATIONS

Fig. 1. Chinese pottery drinking vessel, Museum of Fine Arts, 26.62, H..251 m.D..139m.
Fig. 2. Original drawing: the method (slightly exaggerated) of using the drinking vessel.
Fig. 3. Chinese pottery drinking vessel, Freer Gallery of Art, Washington, D.C.
Fig. 4. Chinese pottery drinking vessel, Peters Collection, New York, H. 8 in., D. 4 in.
Fig. 5. Copy of Chinese drawing from Tun-huang, Serindia, Vol. IV, PL. XCVII.
Fig. 6. Korean pottery drinking vessel, Museum of Fine Arts, 27.328, H.. 354 m, D..170 m.
Fig. 7. Korean pottery drinking vessel,Freer Gallery.
Fig. 8. Korean bronze drinking vessel,after No.176, Cat. Prince Li's Museum, Seoul.
Fig. 9. Detail of Chinese painting from Tun-huang, Serindia, Vol. LV, PL. LXXVIII.
────────────
* Heuzey, L., Origines orientales de l'art.

P.140

Fig. 10. Ku.n.dikaa, complete example from Sahe.th- Mahe.th. I'. Early or pre-Ku.saana. see A.S.I., A.R., 1910-11, p.6 and PL. III, 8.Fig. 11. Ku.n.dikaa, complete example from Bhii.taa. 8". Ku.saana.See A.S.I.,A.R.,1911-12, P.83 and PI. XXX, No.52.
Fig. 12. Ku.n.dikaa, fragment from the Bhi.Mound, Taxila.5". Maurya or earlier. Tentatively restored. See A.S.I., A.R., 1912-13, p.42 and P1. XXXIX, g, i.
Fig. 13. Ku.n.dikaa, fragment from Basaa.nh: 4 ". Gupta. See A.S.I., A.R., 1903-04, PL. XXXVIII, 5.
Fig. 14. Ku.n.dikaa,fragment from Chaarsada: Graeco- Buddhist, see Marshall and Vogel, Excava- tions at Chaarsada, in A.S.I., A.R.,1902-03, Plate facing p. 182, Fig. 24, No.32 (misn- umbered 33).
Fig. 15－23. Ku.n.dikaas, held as an attribute by various deities.
Fig. 15. Ku.n.dikaa, held as an attribute by a seated Naaga (or Naagaarjuna) from Naalandaa (there is a rosary in the right hand): see A.S.I., A.R., 1919-20, P1. XX, b, or Vogel, Indian Serpent-lore, P1. XIV.
Fig. 16. Ku.n.dikaa, held as an attribute, lower right hand of a copper figure of Amoghapaa'sa (Avalokite'svara), now in the Musee guimet. Ninth century.Fig. 17. Ku.n.dikaa held as an attribute by 'Siva, mukha-li^ngam from Unakoti, tripura State,mediaeval. See A.S.I., A.R., 1921-22, P1.XXIX.
Fig. 18. Ki.n.dikaa held as an attribute by
Brahmaa, relief from Kumbakonam (Rao, Elements of Hindu Iconography, II, P1. CXL VII).
Fig. 19. Ku.n.dikaa, held as an attribute by Brahmaa, relief from Tiruvaa.di ( Rao, loc. Cit., P1.CXLIX).Fig. 20. Ku.n.dikaa,HELD AS AN ATTRIBUTE BY Brahmaa:
Cola, relief in the possession of Mr. C. T. Loo.Fig. 21. Ku.n.dikaa, held as an attribute by 'Siva: panel of the Gupta temple at Deoga.rh.Fig. 22. Ku.n.dikaa,held as an attribute by Brahmaa: panel of the Gupta temple at Deoga.rh.Fig. 23. Ku.n.dikaa, held as an attribute by Avalokite'svara: Gupta relief at Baagh.Fig. 24. Typical am.rta kaala'sa and aamalaka 'sila forming the crowning elements of the naagara 'sikhara, the spire of a northern Indian temple. The actual example from the Mahaadeva temple at Nohta, Damoh District.

P.141

Fig.25-29. The am.rta-kala'sa as carried by various deities. Fig. 25. By Indra. Saa~ncii, east gate, right pillar, front. Ca. 100 B.C. Fig. 26. By Maireya, Mathuraa: Ku.saana, A 68 in the Marthuraa Museum, see Vogel, Catalogue.., and in A.S.I., A.R., 1909-10, P1. XXIV, b. Fig. 27. by Maitreya, Yuusufzai: Graeco-Buddhist,see Burgess, Ancient Monuments, P1.118. Fig. 28. by Maitreya, Takht-I-Bahi: Graeco-Buddhist, see A.S.I., A.R., 1907-08, P1.XLVI, c. Fig. 29. by 'Siva, Kosaam. Gupta (A.D. 45 8/9), see A.S.I., A.R., 1913-14, P1. LXX, b.

EDUARD ERKES:
A NEOLITHIC CHINESE IDOL ?
WHEN I CONTRIBUTED THE ARTICLE ON " IDOLS IN PRE-BUDDHIST CHINA". TI ARTIBUS ASIAE, I was not yet aware of a specimen which has come to light among the Neolithic excavations made by J.G.Andersson at ha-kuo-t'un in Fengt'ien, ("The Cave-Deposit at Sha Kuo t'un in Fengtien", Peking 1923, P1. VIII. Fig. 19), and which looks as if it had been an image of
religious significance.
The specimen which is reproduced in Andersson's publication ** is the figure of an animal made of marble, about 3,3 cm high. It is carved with evident car; the head is distinctly set off; the eyes are indicated by two little round grooves the depth of which Andersson gives as about 2 mm, well carved and clearly distinguishable. The fore-paws are only
indicated by the respective parts of the body being a little thickened; but the hind-legs are well done, and the anal opening is carefully represented by an other little round hole. A larger hole transversely perforating the body shows the object to have been ────────────
* Artibus Asiae 1928, pp. 5 － 12. ** J.G. Andersson, The Cave-Deposit at Sha Kuo T'un in Fengtien (Peking, 1923), p1. VIII, fig. 19, a － e
Description on pp. 18/19.

P.142

intended for being worn on a string. It was found attached to a human bone, perhaps belonging to the same man who had once worn it.
The figure seems to have been an amulet, like similar decorations which occur among modern primitives and are generally meant for such a purpose. As far as we may judge, it represented a deity which was thought to protect the owner. The careful execution of the eyes, the ears and the anus hows that the creature was imagined to see, to hear and, probably, to eat and to digest, if not some kind of "magic of the openings of the body" was intended by the last-mentioned device. It must therefore have been conce ived as a live being which had to watch and to defend its possessor and was erhaps fed, i.e. sacrificed to. That the paws are not better executed, is probably due to technical difficulties, for the artist has certainly tried to
bring them out.
Dr. Andersson thinks that the sculpture was intended to represent "most likely a cat", but I do not think this very probable. Certainly the wild cat belonged to the deities venerated by the ancient Chinese, but it is only very rarely mentioned. The only passage n ancient literature referring to cat-worship seems to be the note in the Li-ki, that at the great cha-sacrifice the cats were sacrificed to, because they devoured the mice which did damage to the fields.* Probably out of the same reason the cats ar mentioned, together with the tigers, in the Shi-king among the traits that made the land of Han so lovely.** But there is no indication that the wild cat stood in a closer relation to man, nor that it was conceived as an animal which gave individual
protection or was ever represented in art. The domestic cat did not yet exist in ancient China, it appeared not before the han time, and probably only later, and it is not even sure whether it developed out of the native wild cat or was introduced from
India. ***
Therefore I think it far more probable that we have an image of a tiger before us. For the connexions between man and tiger were far more intimate; the tiger played a rather conspicuous role in cultic worship and was represented in art from the earliest ────────────
*Li-ki 9, 2, 10 (Kiao-t'eh-sheng, Couvreur, Li-ki,
I, 5 9 5). ** Shi, III, 3, VII, 5. This explanation is already given by the Sung commentator Fan Ch'u-I who seems to allude to the Li-ki (quoted by Legge, Ch. Cl. IV, 550). ***F. Hirth, Die Geschichte der Hauskatze in China, Verhandlungen der Berliner Gesellschaft fur Anthropologie 1890, pp. 140-151. － The li 狸 which is mentioned in writings of the later antiquity was probably not the domestic cat but the weasel.

P.143

times. It was not only revered for hunting the wild boars which devastared the fields,* but also thought to chase away demons, ** an idea which prevails still to-day in China.*** That it goes back to prehistoric times, may be gathered from the following fact: The heads of little children are shaved and only two little tufts left, so that the babies are made to look like tigers and will thereby be protected from the attacks of evil spirits who are afraid of tigers. Now this custom is not restricted to Chin a, but also found among Indochinese tribes of further India who are related to the Chinese.*** Therefore it must go back to a period which preceded the separation of the Indochinese group. The tiger was also figuratively represented. In its character as the deity of the west, it was carved in jade, and these tiger-images received cultic worship.+ They were also put into the graves of the dead, probably ++ with the intention to protect the inhabitant of the grave, which was also sometimes believed to be watched by a living tiger.+++
It seems therefore quite possible that the little marble figure bears testimony to the antiquity of these conceptions, and that it served a Neloithic chinese, probably in life as an amulet which he wore at his girdle, and in death as the protective genius of his tomb. I may add that the shape of the image also somewhat recalls the Kuwen forms of the haracter 虎 hu, "tiger". +++ Especially the characteristic bend of the back is represented, and the incisions which are visible on the back as seen in the side-vi ew and perhaps identical with the "fine elongated grooves" mentioned by Andersson, might well indicate the stripes which are never omitted in the old characters.
────────────
*Li-ki, 9, 2, 10 (Kiao-t'eh-sheng, Couvreur, Li-ki, I, 595). ** Feng-suh-t'ung 8, 6b, quoted by Laufer, Jade, p.182. *** Grube, Religion und Kultus der hinesen, pp. 123, 17+ 176, 177. ***Conrady, China (Pffugk-Harttung's Weltgeschichte, vol. III), p.495. Chou-li 18, 41(Biot, Tcheou-li, I, 434); comp. Laufer, jade, pp. 174-184. Chou-li 20, 43 (Biot, I, 490).+++ Yueh-tsueh-shu 2, 3a, translated by de Groot, The Religious System of China, II, 396.++++LO Chen-yuh, Yin-hu-wen-tze-lei-pien, 5, 8a.