THE LOCAL TRADITIONS OF JIANGNAN
AND THE SHIFT FROM WAIDAN TO NEIDAN IN CHINESE ALCHEMY

Fabrizio Pregadio
Institute of Philosophy, Technical University of Berlin

In the last decade, after some important pioneering studies, Chinese alchemy has received much attention by scholars working on the history of Daoism and the history of Chinese science. Thanks to the published research in these and closely related areas, we are now in a better position than we were before to understand the history, theories, and practices of both waidan or "external alchemy" and neidan or "inner alchemy". However, due in part to the alleged closer relevance of waidan to the history of science and of neidan to Daoism, the two branches of alchemy have usually been studied independently of each other. The ways in which waidan and neidan interacted, and the reasons for the transition from the former to the latter, are still largely unknown.

The present paper examines some facets of this transition. My purpose is to show that the first stages of the shift from waidan to neidan took place during the Six Dynasties against the background of the cultural and religious traditions of Jiangnan, the region corresponding to present-day southern Jiangsu. As we shall see, various features of these traditions are reflected in the Zhouyi cantong qi (Token for the Agreement of the Three According to the Book of Changes), the main scripture of Chinese alchemy and the only one to be cherished within both of its branches. I first briefly describe the two main subtraditions within waidan, then review their history during the Six Dynasties, and conclude with some remarks on the transition from waidan to neidan.

1. Subtraditions within waidan

The history of Chinese alchemy is distinguished by a progressive shift of concern from the world of gods and demons to abstract cosmological speculation. These two views of the alchemical work are distinctive of the two main subtraditions within waidan. The shift from the former to the latter culminated around the 6th-7th centuries, and not only affected the history of waidan but also paved the way for the rise of neidan.[NOTE 1]

[NOTE 1]For a survey of the two traditions within waidan and the respective doctrines and practices see Pregadio (forthcoming). The way the shift towards cosmological speculation in waidan influenced the rise of neidan is discussed in Pregadio and Skar (forthcoming).

The most convenient overview of the first subtradition is found in Ge Hong's (283-343) Baopu zi neipian (Inner Chapters of the Book of the Master who Embraces Spontaneous Nature; ca. 317). In chapter 4 of his work, Ge Hong draws from three texts that formed the core of the early Taiqing (Great Clarity) tradition of waidan: the Taiqing jing (Scripture of Great Clarity), the Jiudan jing (Scripture of the Nine Elixirs), and the Jinye jing (Scripture of the Golden Liquor). Both Ge Hong's quotations from these works and their extant versions in the Daoist Canon show that the purposes of elixir compounding and ingestion in early waidan are to obtain long life and immortality, receive protection from divinities, send off demons and other harmful entities--especially those that cause illnesses--and acquire magical powers. Consistently with this religious background, the alchemical process described in the Taiqing sources consists in a sequence of ritual actions marked by invocations and offerings to divine beings. The main stages are the transmission of the method and the oral instructions from master to disciple, the establishment of the ritual area, the choice of an auspicious time to start the compounding, the actual compounding of the elixir, the offering of the elixir to the gods, and the ingestion of the final product.[NOTE 2]

[NOTE 2]See Pregadio (forthcoming) and, for more details, Pregadio 1991, 584-92.

This emphasis on the religious and ritual features of elixir compounding and ingestion is closely related to one of the main features apparent in the Taiqing sources. None of them describe the alchemical process using the patterns, imagery, and language of Chinese cosmology and its system of correspondences. The few instances of methods related to simple cosmological configurations--i.e., those based on sets of five or eight ingredients--are not prevalent in the tradition as a whole, most of whose recipes are characterized by a large number of ingredients with no obvious correspondence to cosmological principles. More importantly, these methods do not allow the interaction of different cosmological planes which is typical of later waidan texts and of the whole neidan tradition.

This interaction, on the other hand, is the main peculiarity of the waidan sources related to the Zhouyi cantong qi, which give an entirely different view of the alchemical process. In these sources, the system of correlative cosmology acquires primary importance. With a marked innovation in the language and the conceptual system of alchemy, elixirs and their ingredients are explicitly used as emblems of cosmological principles. This feature becomes so important that two Tang texts related to the Cantong qi go so far as to state that "compounding the Great Elixir is not a matter of substances and ingredients, but always of the Five Agents", and that "you do not use substances, you use the Five Agents".[NOTE 3]

[NOTE 3]Danlun jue zhixin jian (Instructions on the Alchemical Treatises, a Mirror Pointing to the Heart; CT 935), 3a; Huandan zhouhou jue (Practical Instructions on the Elixir of Return; CT 915), 2.4b. (The abbreviation "CT" precedes the numbers assigned to texts of the Daozang [Daoist Canon] in the catalogue by Kristofer Schipper, Concordance du Tao-tsang: Titres des ouvrages [Paris: École Française d'Extrême-Orient, 1975].)

The description of the alchemical process in the waidan sources related to the Cantong qi is based on three main sets of emblems: the lines, trigrams, and hexagrams of the Yijing (Book of Changes); the Five Agents (or Five Phases, wuxing); and alchemical symbols proper (especially lead and mercury). These and other sets of emblems allow waidan cosmologists to establish parallels between facets of the alchemical process and facets of the cosmological system. Religious and ritual features are largely disregarded, and alchemy becomes a tool for abstract speculation on the nature of the cosmos. The model of alchemical theory and practice described in the Cantong qi and the related sources was a requisite for the development of neidan, which shares portions of its system with waidan and borrows from the waidan vocabulary.

2. The Cantong qi and the historical background of early waidan

When we look at the two subtraditions of waidan against the background of the historical development of early Chinese alchemy, we gain a better understanding of their relation and their role in the development of neidan. According to Ge Hong's testimony, the three main Taiqing texts were first brought from the northeastern region of Shandong to Jiangnan at the end of the 2nd century. Here the imported alchemical disciplines found a receptive milieu in the local religious and medical practices, which involved the ingestion of drugs based on plant products for exorcistic, therapeutic, and magical purposes. At the same time, Ge Hong maintains that alchemy grants access to higher spiritual realms and is therefore superior to healing, exorcism, and other practices such as breathing, daoyin (gymnastics), or sexual techniques.

About fifty years after Ge Hong, social changes in Jiangnan deeply modified the religious and intellectual background of the region. For the history of alchemy, the main consequence of these changes was the establishment of the Shangqing (Highest Clarity) school of Daoism, which developed from the latter half of the 4th century.[NOTE 4] The Shangqing attitude towards waidan is complex. While some waidan texts of the earlier Taiqing tradition were incorporated into the corpus of Shanqing revealed writings, adepts of this school performed the alchemical process mainly as a meditation practice.[NOTE 5] Accordingly, in the hierarchical arrangement of the southern religious customs and of their historical or legendary representatives, Shangqing assigned waidan a place higher than that of ritual and exorcistic practices--in a way similar to Ge Hong's--but lower compared to the techniques privileged by Shangqing, especially meditation.[NOTE 6] This emphasis on the inner aspects of the alchemical work is an anticipation of traits that were to characterize neidan later.

[NOTE 4]Robinet (forthcoming) provides an excellent survey of the history, texts, doctrines, and practices of this school.

[NOTE 5]Strickmann 1979, 169-78; Robinet 1984, I: 176-80.

[NOTE 6]Robinet 1984, 1: 35-48.

It is around this time--near the turn of the 4th and the 5th centuries--that the Cantong qi enters the history of Chinese alchemy. According to tradition, the text was written in the 2nd century AD by Wei Boyang, a legendary character said to come from the Kuaiji area of present-day northern Zhejiang (about 250 km southeast of Maoshan, the early seat of the Shangqing school). A great many studies have discussed the date of this work and its relation to waidan and neidan. Most scholars seem to accept the traditional Han date of the Cantong qi, and consider it as a text related to the Yijing and the apocrypha (weishu) maintaining that it dealt either with waidan or with neidan since that time. Other scholars suggest that the original Han text of the Cantong qi was lost during the Six Dynasties; in the Tang period, unknown authors wrote an identically-titled work that incorporates parts of the cosmological system of the Yijing but deals either with waidan, or with neidan, or with both.[NOTE 7]

[NOTE 7]The two main representatives of the latter view are Chen Guofu and Fukui Kôjun; see Chen 1983, 352-55, and Fukui 1974, 29-30.

These two views, and the subtle distinctions operated within each of them, are not entirely supported by internal and historical evidence. Based on comparison with the earliest known waidan sources--i.e., those of the Taiqing tradition and those incorporated into the Shangqing corpus--the alchemical content of the Cantong qi does not reflect traditions as early as the Han period: the first extant waidan texts related to the Cantong qi date from the Tang. Moreover, attempts to define the present text of the Cantong qi as concerned with waidan come up against its explicit rejection of waidan practices. For example, a passage of zhang 70 says of those who devote themselves to practices deemed to be erroneous:[NOTE 8]

[NOTE 8]References to the Cantong qi are to the number of section (zhang) in Peng Xiao's (?-955) Zhouyi cantong qi fenzhang tong zhenyi (Understanding the Authentic Meaning of the Cantong qi, with an Arrangement into Paragraphs; CT 1002), completed in AD 947.
They look for renowned medicines far and wide
--how utterly opposite to the Way.

The final part of zhang 89 adds:

Dispose of realgar,
get rid of the eight minerals!

If, on the other hand, the present text of the Cantong qi is the same as the original version, and if the Cantong qi is defined as a neidan treatise, this would lead (and sometimes has led) to the unhistorical conclusion that a fully developed neidan tradition capable of producing a scripture of this sort existed already in Han times.

If one dismisses both the traditional date of the Cantong qi and its alleged relation to waidan or neidan since Han times, any attempt to establish the role of this text in the early history of Chinese alchemy should account for two main facts. First, much evidence is available to suggest that a version of the Cantong qi did exist in Han times and was one of several works belonging to the tradition of the "studies on the Book of Changes" (yixue).[NOTE 9] Second, the alchemical portions of the received text speak the language of Tang and later waidan and neidan sources. One can therefore assume that the original Han text of the Cantong qi was submitted to revisions and additions before it reached its present shape. In this revised form, it was used first within waidan milieux, and later provided momentum for the development of neidan. While this suggestion may sound plausible, a major challenge to it comes from the claim that the Cantong qi was lost during the Six Dynasties and was therefore unrelated to the development of alchemy during that time. To evaluate whether this claim corresponds to historical reality, one should examine some details of the history of the text.

[NOTE 9]See especially Wang Ming 1984.

3. The Cantong qi in the Six Dynasties

Although the almost complete silence that surrounds the Cantong qi in the Six Dynasties has prompted some scholars to suggest that its present version was fabricated after the original text was lost, there does not seem to be any reason to assume that any major break in transmission actually took place. The few citations of the Cantong qi before the Tang are found in sources of diverse nature, but acquire significance and yield details on its history in this period if they are examined together.

About two centuries after Ge Hong, Jiang Yan (444-505) mentions the Cantong qi in a poem devoted to the immortal Qin Gao.[NOTE 10] Two lines of the poem read:

[NOTE 10]Jiang Wentong jihui zhu (Collected Works of Jiang Yan, with Annotations), 3.111 (Beijing: Zhonghua Shuju, 1984); cf. Waley 1930, 8.
He proved the truth of the Cantong qi
In a golden furnace he refined the Divine Elixir

Writing in the same years as Jiang Yan, his younger contemporary Tao Hongjing quotes a passage from the Cantong qi in his Zhen'gao (Declarations of the Perfected; ca. 499), which may have come from one of its early versions or from a lost preface.[NOTE 11] About one century later, Yan Zhitui (531-91) reports in his Yanshi jiaxun (Family Instructions for the Yan Clan; ca. 589) that "in the Cantong qi, the character zao is considered to be composed of ren and gao." These words allude to a cryptogram found in the final section of the Cantong qi.[NOTE 12]

[NOTE 11]Zhen'gao (CT 1016), 12.8a-b.

[NOTE 12]Yanshi jiaxun, 2.20a (Sibu congkan ed.); see trl. Teng 1968, 185. The sentence reads.

The three sources mentioned above provide important information. Jiang Yan was from Jiankang (modern Nanjing); Tao Hongjing was born and lived in the same area; and Yan Zhitui came from Shandong but spent part of his life at the court of the Liang. The three authors who mention the Cantong qi before the 7th century came from or lived in Jiangnan, showing that the Cantong qi circulated in southeastern China during the Six Dynasties.

The development of the cosmological tradition provides a plausible background for the transmission of the Cantong qi to Jiangnan after the fall of the Han. The last great representative of the Han "studies on the Changes," Yu Fan (AD 164-233), is also the first author whose work shows acquaintance with the Cantong qi. A gloss on the character yi (change) in the Jingdian shiwen (Lexicon of Classical Texts; early 7th century) attributes Yu Fan with a reference to a sentence of the Cantong qi ("Sun and moon make change") found in zhang 9 of the received text. In one of two possible interpretations, the gloss reads:

According to Yu Fan's commentary [to the Yijing], the Cantong qi says that this character is formed by the graph for "sun" with the graph for "moon" below it.

A slightly different punctuation alters the meaning of the sentence:

Yu Fan's commentary on the Cantong qi says that this character is formed by the graph for "sun" with the graph for "moon" below it.

The punctuation that yields the second reading fits the pattern of quotations in the Jingdian shiwen. The wording of the gloss, therefore, leaves room for the possibility that Yu Fan wrote the earliest known commentary to the Cantong qi. Bibliographic sources do not mention any work by Yu Fan on the Cantong qi, but one of the two Tang commentaries to the scripture supports the indication gathered from the Jingdian shiwen.[NOTE 13]

[NOTE 13]See the commentary ascribed to Yin Changsheng, Zhouyi cantong qi (CT 999), 3.11a.

Yu Fan's relation to the Cantong qi goes further than this, for he is associated with two cosmological patterns used in the text.[NOTE 14] The first pattern is the Contained Stem (najia), which essentially consists of coordinating the eight trigrams to the ten Stems of Heaven (tiangan). While the original form of this arrangement is associated with Jing Fang (77-37 BC), of whose lineage Yu Fan is a later representative, its main development is ascribed to Yu Fan himself, who applied the najia arrangement to a representation of the moon cycle. In the Cantong qi, this pattern is the subject of zhang 13-15 and 46-48, where six trigrams are associated to nodal days in the waxing and waning of the moon. The second pattern is the Twelve-stage Ebb and Flow (shier xiaoxi). This arrangement represents cyclical change through the twelve "sovereign hexagrams" (bigua), whose unbroken lines flow first upwards and then downwards. This pattern, mentioned twice by Yu Fan in fragments of his commentary to the Yijing, is described in zhang 49-60 of the Cantong qi; it played a major role in the later practices of both waidan and neidan as the basis of the system of the "fire times" (huohou).[NOTE 15]

[NOTE 14]For more details on the two patterns outlined here and their use in the Cantong qi see Suzuki 1963, 629-35; Wang Ming 1984, 252-57; Pregadio 1995, 162-63.

[NOTE 15]Ho 1972, 31-34, provides a detailed analysis of this section of the Cantong qi; see also the summary of his discussion in Needham 1976, 60. On the "fire times" in waidan and neidan see Sivin 1980, 266-79, and Robinet 1995, 120-31, respectively. For Yu Fan's references to the Twelve-stage Ebb and Flow see Li Dingzuo's Zhouyi jijie (Collected Explications of the Book of Changes; Tang), 14.350 (Congshu jicheng ed.).

Yu Fan came from Kuaiji, the traditional birthplace of Wei Boyang. More importantly, his lineage was one of those that preserved the traditions of the "studies on the Changes" in Jiangnan during the early Six Dynasties.[NOTE 16] His followers, therefore, may have continued to transmit the original text of the Cantong qi in southern China after the Han until it reached the alchemical lineages of that region, which rewrote the original text into an alchemical scripture.

[NOTE 16]Strickmann 1981, 100-2.

4. The works of Hugang zi

As shown by Jiang Yan's poem quoted above, the Cantong qi was used in connection with elixir compounding by the end of the 5th century. What were the traditions associated with the Cantong qi during the Six Dynasties? Since the two main substances, or emblems, in the alchemical discourse of the Cantong qi are lead and mercury, to answer this question one should look for sources that describe the compounding of elixirs based on those two metals. The little known body of texts associated with Hugang zi provides some relevant materials. Quotations in the commentary to the Jiudan jing and references in bibliographic works show that the now fragmentary corpus going under his name took shape during the Six Dynasties. Traditions associating Hugang zi with Wei Boyang and with Ge Hong's line of transmission suggest a southern origin for these texts, and, in the first instance, point to links with the Cantong qi.[NOTE 17]

[NOTE 17]See Chen 1983, 303-9; Zhao 1985; Pregadio 1991, 567-68.

The writings ascribed to Hugang zi are of remarkable historical value as they included the earliest waidan texts largely based on metals, and specifically on lead and mercury. The best example is in fragments of the Fu xuanzhu jue (Instructions for Fixing the Mysterious Pearl), a work that contained recipes for the separate refining of lead and mercury culminating in a method for their conjunction.[NOTE 18] The final part of the process is quoted in the commentary to the Jiudan jing (Scripture of the Nine Elixirs) as "Method of the Nine[-Cycled] Essence of Elixir-Lead and the Mysterious Pearl" :

[NOTE 18]Huangdi jiuding shendan jingjue (Instructions on the Scripture of the Divine Elixirs of the Nine Tripods of the Yellow Emperor; CT 885), 11.6a-b for the refining of mercury, 12.3a for the refining of lead, and 11.7a-b and 12.3a-b for their conjunction (the last method of which is given twice). The Fu xuanzhu jue is not listed in bibliographic works.

Heat two pounds of Mysterious Pearl and twelve ounces of Nine[-Cycled] Essence of Elixir-Lead in Vinegar of the Yellow and the White (huangbai zuowei), for seven days and seven nights. When the compound coagulates and becomes white and stabilized, let it dry and remove its toxicity. [...] It is used as "blanket and mat," and no elixir will form in its absence.[NOTE 19]

[NOTE 19]"Blanket and mat" (fuji) alludes to the use of the compound as the highest and lowest layers in the crucible together with the other ingredients. A method for making the Vinegar of the Yellow and White in Three Cycles is quoted in the Huangdi jiuding shendan jingjue, 17.5a-b, from the Wanjin jue (Instructions Worth Ten Thousand), another work attributed to Hugang zi. The meaning of the expression "Three Cycles" is explained in 17.7a-8b. The Wanjin jue is listed in Sui shu (History of the Sui Dynasty), 34.1048, which indicates Ge Hong as its author.

The alleged relation between Hugang zi and Wei Boyang suggests that the lead-mercury processes were practiced within a tradition that counted the Cantong qi among its sources. We know next to nothing about the history of this tradition during the Six Dynasties, but the scarcity of contemporary related sources suggests that what may have started as a minor local lineage progressively acquired influence, and later came to affect the whole development of Chinese alchemy.

5. The Cantong qi and the Huangting jing

In Jiangnan, the Cantong qi also came in touch with the traditions of the Shangqing school of Daoism, whose role in the shift to neidan was mentioned above. One of the earlier scriptures incorporated by Shangqing into its canonical literature is the Huangting jing (Scripture of the Yellow Court), a short work in verses originally dating from the 3rd century which describes the human being as the seat of a multitude of divine beings. The Cantong qi shares a number of terms and phrases with this work. An obvious example is the description of the elixir in the Cantong qi, which matches the description of the center of the human being in the Huangting jing:[NOTE 20]

[NOTE 20]References to the Huangting jing are to section and verse of the Neijing (Inner Effulgences) version edited in Schipper 1975.

Altogether, the Cantong qi and the Huangting jing share no less than fifty or sixty terms and phrases. These include the following:

Both texts, moreover, quote or allude to one sentence each from the Laozi and the Yijing:

(Cf. Laozi 28: "Know the White, keep to the Black, and you will be a model to all-under-Heaven")

(Cf. Yijing, Xici A.5: "Heaven and Earth establish their positions, and change takes place between them")

Not all the shared terms and expressions, however, are used with the same or a similar meaning in the Huangting jing and the Cantong qi. In fact, the Cantong qi criticizes several practices mentioned in the Huangting jing:

The last few passages come from zhang 27 of the Cantong qi, which mentions several religious and ritual practices and criticizes them as inadequate for spiritual realization. This suggests that the authors of the alchemical version of the Cantong qi did know the Huangting jing, but borrowed some of its terminology to express their claim of the superiority of alchemy over the southern meditation practices.

6. Conclusion

As we have seen, the texts of the Taiqing tradition of waidan are based on ritual and are mainly concerned with the supernatural world of gods and demons. In the waidan writings related to the Cantong qi, instead, alchemy becomes a tool for abstract speculation on the cosmological system. These two traditions did not develop at the same time: while the first Taiqing sources date from no later than 300 AD (or, according to Ge Hong, from one century before), no waidan text uses the cosmological language and imagery of the Cantong qi before the Tang.

The original text of the Cantong qi was related to apocrypha and to the Han exegetical tradition of the Yijing. The earliest mention of this scripture is by the late Han commentator of the Yijing, Yu Fan, who is also credited with a commentary on it, and whose lineage, based in Jiangnan, may have played an important role in the preservation of its original text during the early Six Dynasties. In Jiangnan, the Cantong qi came in touch with the local Taiqing tradition of waidan and with the Shangqing meditational traditions. As shown by Jiang Yan's poem, the Cantong qi was used in Jiangnan as a scripture providing the foundation of waidan practices by AD 500. The passages shared with the Huangting jing show in turn that the Cantong qi was rewritten as an alchemical text by anonymous authors who were acquainted with Shangqing Daoism and its practices.

We know very little about the waidan traditions related to the Cantong qi during the Six Dynasties, but materials for the study of the early cosmological tradition in waidan are found in the fragmentary corpus of texts attributed to the Hugang zi. These writings are the first known alchemical texts to favor metals--especially lead and mercury--over minerals. The corpus attributed to Hugang zi developed during the late Six Dynasties; not surprisingly, hagiographic accounts that describe Hugang zi as the disciple of Wei Boyang and as the teacher of Ge Hong imply that his lineage originated in Jiangnan.

The Cantong qi profoundly changed the doctrines, methods, and language of waidan. Through an obscure and metaphorical language, it describes facets of the alchemical process resorting to various cosmological configurations. This let the whole array of emblems and patters of correlative cosmology enter the language and imagery of waidan for the first time, allowing the interaction of the notions of yin and yang, those of the Five Agents, the trigrams and hexagrams of the Yijing, and several other sets of emblems in a way that was impossible for the methods of the earlier Taiqing tradition. Moreover, the shift to processes based on lead and mercury allowed the emergence of neidan, whose alchemical imagery is based on these two metals and on the same set of cosmological emblems and patterns. The terms and expressions shared by the Cantong qi and the Huangting jing show the role played by the Shangqing tradition of Daoism in the early development of neidan. Shangqing had transformed the waidan practices of the earlier Taiqing tradition into meditational practices. The encounter between these traditions and the Cantong qi was a decisive factor in the rise and growth of neidan.

WORKS QUOTED

Chen Guofu. 1983. Daozang yuanliu xukao [Further studies on the origins and development of the Taoist Canon]. Taipei: Mingwen Shuju.

Fukui Kôjun. 1974. "A Study of Chou-i Ts'an-t'ung-ch'i." Acta Asiatica 27:19-32.

Ho Peng Yoke. 1972. "The System of the Book of Changes and Chinese Science." Japanese Studies in the History of Science 11:23-39.

Needham, Joseph. 1976. Science and Civilisation in China. Vol. 5: Chemistry and Chemical Technology. Part III: Spagyrical Discovery and Invention: Historical Survey, from Cinnabar Elixirs to Synthetic Insulin. With the collaboration of Ho Ping-Yü and Lu Gwei-Djen. Cambridge: Cambridge University Press.

Pregadio, Fabrizio. 1991. "The Book of the Nine Elixirs and Its Tradition." In Chûgoku kodai kagakushi ron [Studies on the history of ancient Chinese science], ed. Yamada Keiji and Tanaka Tan, II, 543-639. 2 vols. Kyoto: Kyôto Daigaku Jinbun Kagaku Kenkyûjo, 1989, 1991.

------. 1995. "The Representation of Time in the Zhouyi cantong qi." Cahiers d'Extrême-Asie 8:155-73.

------. Forthcoming. "The Elixirs of Immortality." In Handbook of Daoism, ed. Livia Kohn. Leiden: E.J. Brill.

Pregadio, Fabrizio, and Lowell Skar. Forthcoming. "Inner Alchemy (neidan)." In Handbook of Daoism, ed. Livia Kohn. Leiden: E.J. Brill.

Robinet, Isabelle. 1984. La révélation du Shangqing dans l'histoire du taoïsme. 2 vols. Paris: École Française d'Extrême-Orient.

------.1995. Introduction à l'alchimie intérieure taoïste: De l'unité et de la multiplicité. Avec une traduction commentée des Versets de l'éveil à la Vérité. Paris: Les Éditions du Cerf.

------. Forthcoming. "Shangqing: Highest Clarity." In Handbook of Daoism, ed. Livia Kohn. Leiden: E.J. Brill.

Schipper, Kristofer. 1975. Concordance du Houang-t'ing king: Nei-king et Wai-king. Paris: École Française d'Extrême-Orient.

Sivin, Nathan. 1980. "The Theoretical Background of Elixir Alchemy." In Joseph Needham, Science and Civilisation in China. Vol. V: Chemistry and Chemical Technology. Part 4: Apparatus, Theories and Gifts, 210-305. Cambridge: Cambridge University Press.

Strickmann, Michel. 1979. "On the Alchemy of T'ao Hung-ching." In Facets of Taoism: Essays in Chinese Religion, ed. Holmes Welch and Anna Seidel, 123-92. New Haven and London: Yale University Press.

------. 1981. Le Taoïsme du Mao Chan: Chronique d'une révélation. Paris: Collège de France, Institut des Hautes Études Chinoises.

Suzuki Yoshijirô. 1963. Kan Eki kenkyû [A study of the Book of Changes in the Han period]. Second revised ed. Tokyo: Meitoku Shuppansha.

Teng Ssu-yü. 1968. Family Instructions for the Yen Clan. Leiden: E.J. Brill.

Waley, Arthur. 1930. "Notes on Chinese Alchemy (Supplementary to Johnson's A Study of Chinese Alchemy)." Bulletin of the School of Oriental Studies 6.1:1-24.

Wang Ming. 1984. "Zhouyi cantong qi kaozheng" [An examination of the Cantong qi].In Daojia he daojiao sixiang yanjiu [Studies on Daoist thought], 241-92. Beijing: Zhongguo Shehui Kexue Chubanshe. Originally published in Guoli Zhongyang Yanjiuyuan Lishi Yuyan Yanjiusuo jikan 19 (1947).

Zhao Kuanghua. 1985. "Hugang zi ji qi dui Zhongguo gudai huaxue de zhuoyue gongxian" [Hugang zi and his great contribution to ancient Chinese chemistry]. In Zhongguo gudai huaxue shi yanjiu [Studies on the history of ancient Chinese chemistry], ed. Zhao Kuanghua, 184-210. Beijing: Beijing Daxue Chubanshe.