John Dee and the alchemists: Practising and promoting English alchemy in the Holy Roman Empire

This paper investigates John Dee’s relationship with two kinds of alchemist: the authorities whose works he read, and the contemporary practitioners with whom he exchanged texts and ideas. Both strands coincide in the reception of works attributed to the famous English alchemist, George Ripley (d. c. 1490). Dee’s keen interest in Ripley appears from the number of transcriptions he made of ‘Ripleian’ writings, including the Bosome book, a manuscript discovered in 1574 and believed to have been written in Ripley’s own hand. In 1583, Dee and his associate Edward Kelley left England for East Central Europe, taking with them a proportion of Dee’s vast library, including alchemical books—the contents of which would soon pique the interest of continental practitioners. Kelley used Ripley’s works, including the Bosome book, not only as sources of practical information, but as a means of furthering his own relationships with colleagues and patrons: transactions that in turn influenced Ripley’s posthumous continental reception. The resulting circulation of texts allows us to trace, with unusual precision, the spread of English alchemical ideas in the Holy Roman Empire from the late sixteenth century.


Introduction
For a large part of his life John Dee was engaged in reading about and practising alchemy. His alchemical interests have been most closely studied in relation to his own Monas hieroglyphica, in which he set out his alchemical precepts in their most obscure and challenging form. Less has been said, on the other hand, about Dee the reader of alchemical texts (and colleague of other readers), or about Dee the alchemical practitioner (and colleague of other practitioners). This paper focuses on Dee's relationship with two kinds of alchemist: the authorities whose works he read, and the contemporary practitioners he knew, or may have known. Often the two groups come together, for one interest that Dee shared with many of his acquaintances was an active interest in recovering, reading, and putting into practice the works of past adepts.
The earliest evidence for Dee's interest in alchemy is his list of the alchemical books he read (but did not necessarily own) in 1556. 1 This list is largely composed of the classics of Latin medieval alchemy, including works ascribed to Geber, Arnald of Villanova and Ramon Lull, together with a few English works by George Ripley and Thomas Norton. During his travels in Europe during the 1560s, Dee seems to have become acquainted with a wider range of alchemical doctrines, and began to accumulate a substantial library of alchemical literature. These included up to date copies of books by Paracelsus and his followers, and a fine collection of late medieval manuscripts, many retaining Dee's annotations, as detailed in Robert & Watson's important reconstruction of John Dee's library catalogue (1990).
The use of this prodigious library by Dee and his contemporaries has been extensively explored by William Sherman (1995, esp. Ch. 2). More specifically, scholars have pondered the extent to which Dee's alchemical books were read, copied and discussed by others (Bayer, 2005;Keiser, 2007, p. 190;Webster, 1979, p. 323;cf. Sherman, 1995, pp. 44-45). Given the number of duplicate copies of Paracelsian works in his library, Clulee (1998) posits that Dee may have contributed to the dissemination of Paracelsian doctrines in England, perhaps by loaning out copies or using them for teaching. 2 This suggestion is supported by the names of various individuals, labelled 'discipulus', inscribed in some of these books. Such activity would indicate a significant degree of expertise on Dee's part. Yet, as Clulee notes, it is difficult to gain a clear sense of Dee's alchemical thinking, since his writings do not in general deal with alchemy. 3 His most extended treatment of the subject is the Monas hieroglyphica (1564), dedicated to the Holy Roman Emperor Maximilian II: a work that reflects primarily medieval and Neoplatonic rather than Paracelsian influences, and offers at best an obscured view of Dee's practical alchemical interests. 4 For Dee's interest was not confined to the written word-his collections at Mortlake included chemical materials and apparatus, and appended to the house were several outbuildings where he and his assistants practised alchemy. 5 Traces of this activity now survive only in textual form: in manuscript notes of alchemical procedures, practically-oriented marginalia, and a few contemporary recollections. 6 Like the issue of Dee's alchemical influence, the question of how Dee's books related to his practice is one that can only be partially answered, through sifting diffuse and fragmentary sources.
In investigating both questions, I shall focus on a single, littlestudied thread from Dee's seemingly diverse alchemical interests: the late medieval tradition of pseudo-Lullian alchemy, as popularised in works by or attributed to George Ripley (d. c. 1490). In the second half of the sixteenth century, 'Ripleian' writings were well known in his native land, but less so in East Central Europe. This strand therefore stands out in relatively high relief against the rich backdrop of continental alchemy, enabling us more easily to pick out clues to its dissemination. The circulation of Ripleian texts coincides in time with the arrival of Dee and his associate Edward Kelley in Bohemia during the 1580s. Indeed, the spread of Ripley's alchemy in mainland Europe is intimately related to the circulation of Dee's books and Kelley's burgeoning alchemical practice. The travels of the English party therefore leave us clues to the relationship between Dee and two alchemists: the living Kelley and longdead Ripley.

Dee, Ripley and pseudo-Lullian alchemy
From the mid-fourteenth century, a body of alchemical texts posthumously and pseudonymously attributed to the Majorcan philosopher Ramon Lull (1232-1316) became increasingly influential on the theory and practice of western alchemy. Detailed studies by Michaela Pereira (1989Pereira ( , 1992 have identified over 140 works in the corpus, encompassing a host of alchemical applications, from the transmutation of metals to the production of medicinal elixirs and the manufacture of precious stones. Pseudo-Lullian techniques and doctrines proved popular in England from the fifteenth century onwards, and were employed in a number of attempts by English alchemists to obtain the support of noble and royal patrons, including an application to Henry VI (Pereira, 1998), and, a century later, numerous proposals addressed to Elizabeth I and her Lord Treasurer, Lord Burghley. 7 As might be expected given the size of the corpus, pseudo-Lullian alchemical operations are many and varied. However, English commentaries on this tradition frequently invoke two powerful waters, sometimes described as 'fires', with contrary actions. 8 One, the 'fire against nature' (ignis contra naturam) is a powerful and toxic corrosive, valuable for gold-making but deadly if ingested. The second, 'natural fire' (ignis naturae), embodies the vegetative power of nature, and may be used to heal both metals and the human body. Various works, describe the use of these fires in manufacturing the 'mineral ' and 'vegetable' stones. 9 While the former employs corrosive mineral acids in its manufacture, the potable vegetable stone is made using products derived from wine: vinegar, tartar, and spirit of wine (Rampling, 2009(Rampling, , 2010. One basic recipe for the vegetable stone, found in hundreds of permutations in alchemical literature, employs a metallic body called 'sericon' as its prime matter: usually taken to denote minium (red lead). The minium is dissolved in strong wine vinegar, and the resulting solution heated until a thick gum remains in the bottom of the glass. 10 When distilled, the gum yields a white vapour, the fumus albus, which is collected in a receiver and condensed to form a liquid, which is then subjected to further procedures. Owing to the circumstances of its manufacture, the vegetable stone was regarded as combining both mineral and vegetable qualities. In this form, it was regarded as a safe and legitimate product for human consumption, and therefore provided one basis for the medicinal aurum potabile-an objective which cannot have been assisted, in practice, by the use of toxic lead compounds.
I have elsewhere described this lead-and-vinegar method as 'sericonian', to distinguish it from approaches based on different ingredients and processes (Rampling, 2010, pp. 128-129). In England, this approach was commonly associated with the figure of George Ripley, a canon-regular of the Augustinian priory at Bridlington in East Yorkshire who flourished in the 1470s, as attested by the colophons of his best known works, the Compound of alchemy (1471) and Medulla alchimiae (1476). 11 In the Medulla, for instance, Ripley gave instructions for the manufacture of the vegetable stone from sericon, and its subsequent combination with the mineral 'fire against nature' (a corrosive mixture of cinnabar, vitriol and saltpetre) to produce a powerful transmutational elixir-the aqua composita, or compound water (Ripley, 1649, pp. 143-145, 170-172).
Doctrines of both Lull (or 'Raymond') and Ripley would have been familiar to John Dee from the earliest days of his interest in alchemy, as appears from the list of alchemical books he read in 1556. These include 'Quinta essentia Raymundi Lulli', the abbreviated title of the Liber de secretis naturae seu quinta essentia, a key work in the pseudo-Lullian corpus, and the first to employ the 'quintessence' (highly rectified spirit of wine) for both medicine and transmutation. 12 Dee owned at least six copies of this work by 1583, together with many other items from the corpus. 13 The list also includes an unnamed text, labelled 'Ripley, anglice.' 14 By 1556, a variety of Ripleian treatises would have been available in English, although none had as yet been printed. Ripley's famous Compound was composed in English verse, while the Medulla was translated into English in 1552 as the Marrow or Mary of alchemy. Several other Latin texts attributed to Ripley, including the Philorcium alchymistarum and Concordantia Guidonis et Raymundi, were translated around the same time, while two other well known treatises, the Pupilla alchimiae and Accurtations of Raymond, seem to have been originally composed in Middle English. All circulated widely in the second half of the sixteenth century. 15 By the far the most popular of Ripley's works was his Middle English poem, the Compound, or 'Twelve gates.' Dee later owned a mid-sixteenth-century copy of this work which he annotated on 21 December 1595 (Bodleian Library MS e museo 63, fols. 41r-65r). It is likely that he was familiar with the Compound long before this-by the 1560s, it was already one of the best known alchemical works in England. Yet Dee's interest was not confined to Ripley's poetical masterpiece: he also copied out several Ripleian prose works in his own hand, all of which may be distinguished by their practical, 'sericonian' character. These include the Accurtations of Raymond, a theoretical treatise accompanied by a collection of recipes, which was transcribed (at least in part) by Dee, with copious annotations, in Wellcome Library MS 239 (pp. 1-44). 16 Like the pseudo-Lullian waters it describes, this text is a composite, which accreted over time from originally disparate elements (Rampling, 2009, chap. 4;. However, by the second half of the sixteenth century it was generally regarded as having been compiled by George Ripley, an association that doubtless stemmed from the distinctively 'Ripleian' flavour of both the theorica and practica (which include several sericonian recipes), as readers often observed in marginalia. In Dee's copy, for instance, one note compares a passage on fermentation with the Compound: 'Riplay in his 12 Ga[tes] in the Chap. of Ferm[entation]' (MS 239, p. 33).
Dee seems also to have drawn on Ripley when translating his textual studies into practical procedures. Bodleian Library MS Rawlinson D.241 is an alchemical notebook, written partly in Dee's hand, that details experiments carried out between 22 June and 6 October 1581. This account of Dee's daily practices and observations implies that Dee-no armchair alchemist-was willing to get his hands dirty. At one point, he describes how he obtained over 10 oz of 'quick mercury' from a sublimate, 'by my diligence in pressing the soft stuff || betwene my fingers partly: and by washing it | destilled vineger.' The result was, 'a faynt fyne slick or slyme: which I yet | left in an stone Chamber pot in the vyneger, farder | to try what wold comme of it' (fol. 3r-v). In extracting this slimy 'mercury', Dee was guided by an earlier authority: Whereby I perceyued that an [ounce] had byn enough | or 6 [drams] rather, to kepe the water in hand | and not to overcharge it, as Riplay in philortium warnes of. (Ibid.,fol. 3r) The reference is to a passage in the Philorcium, in which Ripley advises his disciples to take care when adding liquid to their work: Draw a water from the same stone using an alembic, and with that water dissolve the stone by alembic, and with that water dissolve the stone infinite. But dissolve it by little and little, and you shall not incontinently suffocate it with water, for if it is correctly regulated you may with one pint of water make, if you wish, an infinite quantity of water. 17 Ripley was clearly regarded-at least by Dee-as having something more than theoretical application: his recipes and recommendations assigned practical value almost 90 years after his death.
Following the arrival of the scryer Edward Kelley in his household in 1582, Dee's alchemical pursuits increasingly relied on spiritual as well as temporal authorities. Within two years of making the notes in MS Rawlinson D.241, Dee had relocated himself, his family, and many of his alchemical books to East Central Europe, where he and Kelley attempted to establish themselves first in Cracow and later in Prague, guided by the advice of their angelic interlocutors (Harkness, 1999;Parry, 2011). Yet Dee's practical interest in alchemy continued. Furthermore, his continental travels coincided with a new phase in the reception of Ripleian alchemy: one that would culminate 40 years after his death with the publication of Ripley's Opera omnia chemica (1649) in Kassel. When compiling the twelve texts that constituted this volume, the editor, Ludwig Combach, was concerned to secure reliable exemplary manuscripts-at least one of which was linked to Dee and Kelley, and their contacts within the Empire. These connections not only shed light on the alchemical activities of Dee and Kelley in Bohemia, but also suggest that these played a role in shaping Ripley's own posthumous reception.

Ripley and the 'Kelley circle'
Dee, Kelley and their families left England on 21 September 1583, in the company of Albrecht Łaski, palatine of Sieradz, whose pretensions to the throne of Poland had received encouragement in the angelic conversations mediated by Kelley at Mortlake. Following an inauspicious start in Cracow and at the imperial capital of Prague, the Englishmen eventually settled in Třeboň (Wittingau) as clients of Vilém of Rožmberk (1535-1592), one of the greatest magnates in Bohemia, and like the Emperor, Rudolf II, a patron of alchemists. 18 In time, Kelley's reputation as an alchemist enabled him to obtain imperial favour and a position at Rudolf's court, where his claims of noble Irish descent were validated by the conferral of a knighthood, and the opportunity to amass considerable property.
Until his death in poorly-documented circumstances, probably on 1 November 1597 (Prinke, 2010), Kelley pursued a spectacular alchemical career in Třeboň and Prague. Indeed, by 1587 his own claims to alchemical expertise had eclipsed those of his erstwhile 'senior partner', Dee, securing the favour of both Rožmberk and Rudolf II and attracting interest from Elizabeth I and Lord Burghley in England. The reasons for his success are as elusive as they are incontrovertible. English and Bohemian sources testify to Kelley's extraordinarily convincing alchemical transmutations, some involving the active participation of the Emperor himself. While the extent to which Kelley himself believed in the authenticity of alchemy remains unclear, it is certain that many of his contemporaries regarded him as a legitimate adept, whose successes were converted into status, wealth, and real estate.
Kelley's rise is mirrored by an increasing interest in the works of George Ripley in East Central Europe, which seem to have made their Bohemian debut at around the same time. For instance, nine Ripleian works were transcribed in Bohemia in 1605-06 by the alchemist Simon Thadeas Budek, including a Latin translation of the Compound. 19 Budek's keen interest is attested by dense annotations and cross-referencing, and by the provision of a 31-folio catalogue listing terms and materials employed in Ripley's works. 20 The manuscript also includes an intriguing reference to Kelley, reporting that the disgraced Englishman committed suicide following a failed attempt to escape from Rudolf's custody (ÖNB 11133, fol. 393v;Hausenblasová & Purš, 2009, p. 78).
Such connections were not lost on Ripley's later editor, Combach, who attributed the Latin translation of both the Compound and two other works (the Epistle to Edward IV and the Clavis aureae portae) to Kelley. 21 For the first two cases at least, this attribution is spurious. The same translations of the Compound and Epistle were already circulating in France by the early 1570s, as the Liber duodecim portarum ('The book of the twelve gates') and the Epistola ad Regem Eduardum respectively. 22 An abbreviated, prose version of the Liber was printed in Frankfurt in 1595, edited by the Paracelsian physician Bernard Georges Penot (c. 1522-1620), while a fuller version was published in Leiden in 1599 by Penot's friend, the Protestant pamphleteer Nicholas Barnaud (c. 1539Barnaud (c. -1604, in his Quadriga aurifera. 23 Combach's own edition, although the most complete to date, was based on the same early translation.
Combach's attributions might therefore be dismissed as an attempt to promote Ripley's authority by linking it to the celebrated Kelley. Yet surviving manuscripts suggest that Kelley's own fascination with Ripleian alchemy may have helped stimulate wider interest in the medieval English alchemist, subsequently contributing to the Liber's appearance in print. Several of the individuals involved in the later dissemination of Ripley's works were staying in Prague at roughly the same time as Kelley, and in circumstances conducive to the sharing of alchemical ideas, or, for that matter, texts.
Penot and Barnaud actually met in Prague some time after 1586, while Barnaud was a guest of another alchemical enthusiast, the emperor's physician, Thaddeus Hájek (1525-1600). During this time they collaborated on a 'Practica alphabetica', which-as Penot later reported-Barnaud 'wanted to be arranged according to the time honoured model.' 24 By this, Barnaud seems to have had in mind an alphabet of the kind found in the pseudo-Lullian Testamentum, in which letters denote ingredients and chemical processes. The alphabet, published in Penot's Apologia (1600, pp. 85-96) as 'Rubri & Albi vini a terra foliata extractio' ('The extraction of red and white wine from foliate earth'), yields a fairly straightforward sericonian recipe, beginning with the dissolution of the calcined 'green lion' in vinegar to create a gum.
There would have been plenty of opportunities for both men to meet the English visitors. Barnaud, like Dee and Kelley, became a client of Vilém of Rožmberk, while his host, Hájek, had housed the Englishmen during their first visit to Prague. Barnaud would eventually dedicate his edition of the Liber to another Bohemian acquaintance: Sir Edward Dyer (1543-1607), an English courtier and poet sent to Prague on at least three occasions between 1588 and 1590, partly in repeated attempts to persuade Kelley either to return to England or to teach him alchemical secrets. 25 Kelley's distinctive presence on the Bohemian alchemical scene is also recorded in Penot's De denario medico (1608) Mai, or Maius, was well connected in both political and alchemical circles. 26 In 1601, Rudolf II appointed him to the position of Appellationsrat, prefect of the silver mines at Joachimsthal-a post previously vacated by another alchemist, Sebald Schwaertzer. From there he corresponded with another alchemical patron, Moritz, Landgraf of Hesse-Kassel. In one letter he apologises for being prevented from visiting Moritz, since the emperor requires his services as an envoy to the Counts of Mansfeld. 27 In lieu of his person, he reveals an approach to alchemical transmutation that, in characteristic sericonian style, begins with lead and vinegar. 28 Mai also maintained links with other alchemical practitioners. Penot dedicated a tract in De denario to him, addressing him as a councillor to the Elector of Brandenburg (1608, p. 127). Mai contributed a eulogistic poem to the De signaturis internis rerum (1609) of the Paracelsian chymist Oswald Croll, and was the addressee of one of the letters collected in the Rerum chymicarum epistola (1595) of the Saxon alchemist and physician, Andreas Libavius. 29 He was also acquainted with both Dee and Kelley. Dee wrote to him on 14 November 1586 (Dee, 1582, fol. 88v), and he visited Clavis aureae portae, Riplaeo tribui video, atque hoc nomine ex Anglico idiomate illum vertit Eduardus Kellaeus ', ibid., p. 225. 22 See, for instance, CRC 9.42 and 9.46-47 in Rampling (2010). 23 Ripley (1595Ripley ( , 1599. On Penot and Barnaud, see Olivier (1992Olivier ( -1996, Kahn (2005Kahn ( , 2007. 24 'Ne huius artis indagatores mirentur, ita nos sub verborum inuolucris artem tradidisse: et enim cum Pragae apud Do: Hegetium Thadeum ageremus praxim alphabeticam, auspicatisumus cum Do: Nicolao Bernado, quam ipse more veterum descriptam ita esse voluit', Penot (1600, p. 96). See also Olivier (1992Olivier ( -1996. 25 Dyer was knighted in 1596. See Sargent (1968). 26 On Mai, see Evans (1997, p. 209-210;216), Olivier (1992Olivier ( -1996, Telle (1991, pp. 176-177), Croll (1998, p. 13 In his verse dedication, Mai complains that while his own age is not without learning, it struggles to regain the wisdom of earlier authorities: One man, since he grasps nothing, condemns the most learned writings of the wise ancients, calling them barbarous writings; another despises chymistry as false, and believes the books of Ripley vain dreams. 32 The Emperor, as patron and guardian of alchemists, of course knows better. Yet, Mai explains, he actually has two reasons for completing the translation: The first is that you understand the arts and are favourable to philosophers; you nurture artificers. The other is that Kelley, than whom no one is more excellent, ordered this work to be turned into Latin verse. 33 The Englishman's influence is underscored by the inclusion of a distichon attributed to Kelley This artful verse, with its pun on the shared meaning of maius and magis ('greater'), suggests that Kelley was able to trade on Ripley's reputation as a means of forging or strengthening relationships with other adepts. Besides encouraging Mai to compose a more elegant verse translation of the Liber, Kelley also compiled a number of Ripleian works for presentation to a Silesian patron. These survive in a Latin manuscript, Ksią _ znica Cieszyń ska SZ DD.vii.33: a compendium of Ripleian treatises transcribed in Prague between May and July 1592 by one 'Jan Kapr.' 34 Jan Kapr of Kaprštejn, also known as John Carpe or Johannes Carpio, was an administrator responsible for Rudolf's vineyards. 35 As Carpio, he is mentioned several times in Dee's diaries as one of Edward Kelley's known associates: a regular visitor to Třeboň , who may have assisted Kelley in his laboratory (Dee, 1998, p. 230). His latinized name appears in several alchemical manuscripts now in the Royal Library of Copenhagen, written in English hands and connected with Kelley (Bäcklund, 2006). Bäcklund argues that these manuscripts 'strongly suggest . . .the production of English alchemical manuscripts stemming from a circle around Dee and Kelley in Prague', a circle that was active from 1588-89, and may have continued even after Kelley's death until the death of Rudolf II in 1612 (ibid., p. 306). The existence of such a circle is further supported by the transcription of SZ DD.vii.33, which suggests that in addition to practical assistance, Kapr had access to Kelley's alchemical books, and may have served him as an amanuensis. At the conclusion of Ripley's Medulla, he has added a note, indicating that the collection was compiled at Kelley's behest: Edward Kelley wrote this book out of kindness and love for his most sincere friend, the noble lord Karl von Biberstein, 2 August in the year 1589: whom he wishes to have known as his adopted philosophical son, and to be esteemed above all other mortals. 36 The recipient of this gift, the nobleman Karl von Biberstein, or Karl of Biberštejn (1528-1593), was a Silesian official and imperial councillor, who twice served as master of the Bohemian mint. 37 Like Mai, Biberstein was therefore well positioned both socially and economically, and given his position might have been expected to have a particular interest in novel metallurgical processes, including alchemy. As Dee recorded in the diary, Biberstein visited Třeboň on 26 March 1587, and 'sent for me to his ynn to make acquayntanse with', returning to Třeboň on 16 September (Dee, 1582, fols. 96r, 100v). Although Dee was silent regarding the substance of their conversations, it was apparently Kelley who made the most lasting impression on the mint master: a relationship that was clearly in good health in August 1589, some months after Dee's own departure from Bohemia.
In adopting Biberstein as his 'philosophical son', Kelley assumed the role of his master and tutor, a relationship embodied in the transfer of alchemical knowledge. It is significant that the knowledge Kelley bestowed on his pupil centred on the work of his English predecessor, Ripley. SZ DD.vii.33 includes seven Ripleian works: the original Latin translation of the Liber duodecim portarum (fols. 28r-31r, 165v-204r), together with the Epistola ad Regem Eduardum (fols. 31v-37r), Liber de mercurio et lapide philosophorum (fols. 37v-44v, dated 17 July 1592), Clavis aureae portae (fols. 45r-62r), Medulla alchimiae (fols. 99r-119r), Pupilla alchimiae (fols. 139v-146v, dated 14 May 1592), and Philorcium alchimistarum (fols. 147r-165r). These are supplemented by other texts, including a treatise attributed to Ripley's major authority, Raymond, and the Latin text of the Work of Dunstan, a theoretical and practical treatise pseudonymously ascribed to the tenth-century archbishop of 30 Bassnett (2006, p. 290 Canterbury, but actually adapted from another work in the Ripley Corpus-the Accurtations of Raymond. 38 Besides its value as a storehouse of authoritative texts, the compendium thus drew attention to English expertise in the theory and practice of alchemy: a tradition to which Kelley was himself an heir. Ripley's authority thus underwrote Kelley's, while his assembled corpus offered a store of intellectual capital that could be disbursed as gifts and deployed as evidence of experience within a long established practical tradition. Yet the benefits of this diachronic relationship were not one-sided. Kelley's interest in Ripley would have a lasting effect on the European reception of the medieval adept, as may be seen from the subsequent history of the Ripley corpus. By 1649, Rudolf II and Moritz of Hesse-Kassel were long dead and their alchemical circles dispersed. The Thirty Years' War had raged through Europe, and a presentation volume of Mai's Liber duodecim portarum, either the Emperor's copy or Rožmberk's, was borne away from Prague as booty by Swedish troops. 39 Another copy, now 4 o MS chem. 68, came to rest in the archives of the princely court of Hesse-Kassel, there to be noticed by Ludwig Combach, physician to the former landgrave, Moritz, and his successor Wilhelm. In the preface to the Opera, Combach expressed his regret at being unable to publish Mai's version of Ripley's poem, since the manuscript was not his own. 40 Fortunately, he still had the earlier translation to fall back on, which he believed to have been translated by Kelley, found in a manuscript previously owned by Mai. 41 This manuscript, now 4°MS chem. 67, is a compendium of Ripleian texts, probably compiled in Prague some time after 1600, densely annotated and amended by several scribes. One of the annotating hands is that of Mai, while the compendium also includes one of Mai's Latin verses, 'AEnigma M. Nicolaii Maii' (fol. 183r). Although Mai's is not the principal hand, we would have to guess that the scribe, too, had some relationship with Kelley. The collection includes a recipe heard from Kelley's own lips ('Ex ore EK'); the 'Deposition of Edward Kelley, Englishman', dated Galway, 10 March 1593 (testifying to Kelley's noble Irish lineage); and several extracts apparently taken from letters (one dated Prague, 20 June 1587). 42 To Combach, these Kelleian connections may have been at least as valuable as the texts they adorned, to judge by the fact that he published the epistolary fragments two years earlier than the Opera, in his Tractatus aliquot (Combach, 1647, pp. 31-33). Their provenance was assured since they came from one of Councillor Mai's own manuscripts, as Combach asserted in the preface to the Tractatus. 43 Later, he returned to the manuscript again when preparing the Opera. It provided the primary exemplar for his edition of the Liber, and is the second of two manuscripts used in preparing the Medulla. 44 In total, MS chem. 67 includes eight of the twelve texts included in the Opera. 45 All eight of these recur in Budek's Bohemian compilation, while six were also included in Kelley's gift to Karl von Biberstein.
Combach may therefore have felt that he had good grounds for adding Kelley's name to Ripley's works. The references to 'E.K.', coupled with a complete copy of the Liber in a manuscript previously owned by Kelley's known associate, Mai, could well have persuaded Combach that one English adept, Kelley, took a hand in the dissemination of another: Ripley. In preparing the Opera, Combach thus built on the activities of a network of practitioners based in Bohemia some 60 years earlier-a circle linked to Kelley, and characterised by its interest in the English Canon of Bridlington.

Ripley in practice
Within the circle of Kelley's alchemical clients and correspondents, John Dee himself has so far played a surprisingly modest role. Although his Monas was discussed and imitated in print and manuscript (including a reference to 'Liber Monadis J. D' in MS chem. 67, fol. 1v), Dee's name appears less frequently than that of Kelley in relation to alchemical receipts. 46 The disparity may be partly explained by Kelley's celebrity as a successful adept-a reputation which naturally placed a premium on recipes associated with his name-and partly by Kelley's own strategy of actively disseminating alchemical texts, ranging from authoritative treatises to his own poems, letters and receipts. 47 What, however, was the source of Kelley's bibliographical knowledge? Little is known of his reading habits prior to his departure from England. On one famous occasion he brought Dee a book: the mysterious 'Book of Dunstan' that he claimed to have discovered, together with a sample of powder, following angelic guidance. 48 In general, however, the evidence suggests that he owed such knowledge to Dee, who lent him books, and translated the Opuscule of Daniel Zeccaire for him out of French (Dee, 1582, fol. 105r). Given the scale of Dee's alchemical interests, manifested both 38 The Work of Dunstan is based on a Latin translation of the Accurtations, with the references to Ripley's chief authorities, Raymond and Guido, removed-presumably in order to strengthen the case for authorship by the tenth-century Archbishop, Dunstan (Rampling, 2009, chap. 4). To complicate this picture further, the Clavis aureae portae is itself a 'doctored' version of the Work of Dunstan, to which some further material, including a firm attribution to Ripley, has been added (Rampling, 2010, p. 149). Given Kelley's relationship to both the Work and the Clavis, it is conceivable that he may have helped compose either or both of these forgeries. One section of the Clavis which duplicates material from the Work is not present in SZ DD.vii.33, which also includes the full version of the Work. On Dee and Kelley's interest in Dunstan, see Kassell (2000). 39 Dudík (1855, p. 228). Dudík suggests that the manuscript may have been part of the Rožmberk collection in Prague (see also Evans, 1997, p. 210, n.1). It eventually reached in his books and his laboratories, it seems probable that some of the information Kelley shared with his 'circle' also originated from Dee's famous library. When Dee set out for Europe, he was accompanied by approximately 800 of his books, some of which are mentioned in his diaries. 49 Although we do not know for certain whether these included any of Ripley's works, we can infer that they did, and that these works were shared within a network of which only vestigial traces now remain. This conjecture is supported by the case of a lesser known Ripleian work, the Viaticum.
The Viaticum seu varia practica Georgii Riplaei ('Viaticum, or various practices of George Ripley') is one of the works included in Ripley's Opera (1649, pp. 337-65). Unlike the widely travelled Liber duodecim portarum, it is possible to set a terminus post quem for this work's appearance in East Central Europe, for the Viaticum includes several processes extracted from a manuscript that was only discovered in 1574, by the English alchemist Samuel Norton (1548-1621).
On 20 July 1577, Norton, son of the Somerset gentleman Sir George Norton, dedicated his alchemical treatise, the Key of alchemie, to Elizabeth I. Here he described his discovery of an old Latin commonplace book, containing Ripley's personal notes and jottings-his daily 'Bosome book': Although it fortuned mee in manner vnloked for, to hitt vpon the secret bosome booke of Riple, wherby the true grounds are discovered, Of which havinge by profe found so many to be true, and little doubtinge of the accomplishment of the rest; I thought it but a point of dutie to reveall and vppen the Secrets heereof vnto your Highnes. (Getty Research Institute MS 18, Vol. 10, Pt 2, p. 7) Norton says nothing of the provenance of this manuscript, which is apparently no longer extant. The Latin text of the original survives in only a single, later copy, now British Library MS Harley 2411. With the intention of presenting its contents to the Queen, however, Norton set about translating the collection of treatises, poems and practical recipes into English. 50 This plan was subsequently amended, as Norton decided instead to incorporate practical information from the Bosome book into his own treatise, the Key.
Several copies of Norton's English translation, dated February 1573 (i.e. 1574), 51 survive, together with various 'hybrid' recipes excerpted and adapted from this translation. Dee was certainly familiar with some of these English redactions: Bodleian Library MS Ashmole 1486 (Part V) contains his own transcriptions of two of them, the Whole work of the stone philosophical (pp. 1-18, here titled 'George Ryppleys bosome booke or Vade mecum') and the Practise by experience of the stone (pp. 19-25). 52 Dee's annotations suggest that he was attempting to make practical sense of the recipes (on p. 1 he interprets 'sericon' as antimony), which he had grouped beneath the well known aphorism, 'Liber librum apperit' ('The book opens the book').
While Norton's English translation seems to have circulated only in England, extracts from the original Latin Book were soon available in continental Europe as components of the Viaticum: a collection of short, practical tracts gleaned from various sources and gathered under sixteen subheadings. Of these, at least eight are items apparently extracted directly from the Book, as appears from comparison with the most complete surviving copy of the latter, MS Harley 2411. The Viaticum also shows evidence of revision, most obviously in 'Oleum verò Solis fiet' ('Oil of the sun [i.e. gold] will truly be made'). Although framed as a single, practical process, this short text is actually a reworking of 31 of the Notable rules from Guido: a collection of 45 aphorisms supposedly gathered by Ripley from the works of Guido de Montanor, and a component of the Bosome book. 53 Two further sub-sections of the Viaticum, 'Elixir vitae' and 'Virtutes huius quintae essentiae hoc modo poterunt probari' ('The virtues of this quintessence may be proved in this way') have been adapted from another Ripleian work, the Elixir vitae, that circulated widely in English in the later sixteenth century. 54 Yet another, 'Sequitur aliud opus', is extracted from the Accurtations of Raymond. The Viaticum is therefore a compendium, prepared by a compiler who was clearly familiar with a variety of Ripleian works, including the original version of the recently discovered Bosome book.
Although a few later manuscripts of the Viaticum survive in English archives, the distribution of surviving copies suggests that the work may actually have been compiled in Bohemia. The earliest manuscript copies date from around the turn of the century, in the compilations of Nicolaus Mai (MS chem. 67) and Simon Budek (ÖNB Codex 11133). The presence of these excerpts in Prague within a few decades of Norton's discovery merits some investigation, since their appearance of course postdates the arrival of Dee and Kelley in Třeboň . A clue is provided by the copy in MS Harley 2411. This early seventeenth-century manuscript includes a recipe entitled 'Magna philosophorum corrosiva' ('The great corrosive of the philosophers'). This begins with the distillation of vitriol with 'salt of sericon', resulting in a curious residue: When our menstrual 'mercury' ascends from the sericon by the violence of the fire, a certain part of it is found cleaving to the side of the flask after the complete distillation and cooling of the glass, like salt and of crystalline appearance. And that crystalline earth is a fixed material, and apt to receive any form whatsoever. Let it therefore be gathered and kept. And the form of this earth is like mercury sublimed, and therefore shines brightly . . . This secret I learned through practice: G[eorge] R[ipley], as God is my witness. 55 Beneath, the scribe has sketched a small picture of a flask and receiver, depicting the crystalline residue as a ring around the inner circumference of the flask. A note adds, 'So in a circle aboue the matter was the cleare matter lyke [mercury]' (MS Harley 2411, fol. 55r).
The picture is dated 1588, while additional information is supplied by a marginal note: I saw the same on 8 February 1588 (new style) in Trebon in Bohemia. From 2 lb. of sericon dissolved in distilled vinegar, and by means of spirit of wine cleansed of much sediment, came 4 oz. of red wine or oil. J:D. E: K. 56 These notes apparently describe a sericonian experiment carried out by Dee and Kelley in Třeboň , together with its results: the red oil and crystalline residue, complete with sketch. Since MS Harley 2411 dates from after 1588, it is likely that the scribe copied both recipes and annotations from a copy of the Bosome book in the possession of Dee or Kelley. 57 This conclusion is supported by Dee's own diary entry for 8 February 1588, which records how, Mr. E.K. at 9 of the clock after none sent for me to his laboratory over the gate: to se[e] how hee | distilled sericon, according as in tyme past & of late he hard [sic] of me out of Riplay. (Dee, 1582, fol. 111r) Evidently Dee, like Samuel Norton, did not regard the Book merely as an antiquity, but as a practical resource. Indeed, this episode provides a rare glimpse of Dee and Kelley's alchemical activity in Třeboň , in which the Ripleian recipe, taken from Dee's copy of the Book, was tested by Kelley, and the results recorded in the margins of the Book itself. At a later point, both the Book and its dated annotations were copied by the scribe of MS Harley 2411.
Kelley's personal interest in the Book seems not to have been confined to replication of its practices. One of the epistolary fragments in MS chem. 67 describes, A certain golden and silvery hermaphoditic water, which, if you will extract it naturally from a perfect body and an imperfect metal, will give you the water of life, the stinking water and the green lion, in which are all colours, ending in two, white and red. It does not matter what earth your substance comes from (as Guido asserts), as long as it is fixed. 58 This short passage draws on two axioms of Guido de Montanor. 59 The first defines three 'species' observed in the work: the white fume (fumus albus), the stinking water (aza foetida), and the green lion (leo viridis). However, Kelley has modified his source by substituting 'water of life' for the 'white fume.' The second axiom, concerning the earth to be used, is another saying of Guido, in this case an exact quote. 60 Although I have been unable to locate these axioms in any of Guido's surviving works, both are found in two Ripleian texts: the Accurtations of Raymond and the Notable rules from Guido. Kelley's usage is slightly closer to the latter, suggesting that the model for his alchemical composition, as for his laboratory practice, was a component of Ripley's Bosome book.
If Dee had a copy of the Book with him in Třeboň , this must be considered a likely source for the extracts that began to appear in Bohemian manuscripts soon after, in the form of the Viaticum: a distillation of noteworthy processes from an authoritative source, in which we can discern echoes of Dee and Kelley's own alchemical practice. This new compilation went on to acquire an afterlife of its own in manuscript and print. The reworked process 'Oleum verò Solis fiet', based on Guido's axioms, may also owe its existence to Dee or Kelley, since MS Harley 2411 includes a copy towards the end of the Book, with the note 'Revised' (fols. 75r-77r, at fol. 77r). Grouped with other items taken from the Book and elsewhere in the Ripley Corpus, this process yielded the Viaticum: a 'new' Ripleian work with a predominantly continental circulation and an unambiguously practical flavour.
That Dee owned a copy of Ripley's lost book makes sense when we consider his own network of alchemical acquaintances in England, several of whom were involved in the Book's dissemination. These included Edward Cradock, the Lady Margaret Professor of Divinity at the University of Oxford, who translated one item from the Book, the Somnium, in June 1582, a year before Dee's departure abroad. 61 A new English translation of the Book was later commissioned by another friend of Dee, Gawin Smith, a prominent engineer in the service of both Elizabeth I and James I. 62 This translation, made twenty years after Norton's, was completed on 24 July 1593 by one Roger Howes, 'for Mr Gawyn Smithe gentleman.' 63 Smith, whom Howes styles 'gentleman Master of her maiesties Engines', had visited Dee for several days in Bremen in October 1589, and petitioned the queen on his behalf in July 1590 (Dee, 1582, fols. 134r, 147r).
Irrespective of whether or not Norton's discovery was an authentic manuscript of George Ripley, these connections suggest that the newly discovered works of England's master alchemist provoked keen interest among Elizabethan and Rudolfine cognoscenti. It is unlikely that Dee, the well-connected bibliophile and alchemical enthusiast, would have remained in ignorance of such a find. Rather, the evidence suggests that he acquired a copy himself, and that he considered it sufficiently valuable to include among the 800 books (a fraction of his vast library) that accompanied him to Bohemia.
Like the original Book, Dee's copy has not survived, although its vestiges remain: in the diary record of Kelley's practical experiments; in the copy in MS Harley 2411; in Kelley's letter; and in Bohemian copies of the Viaticum. Such traces also hint at a vanished network of readers and practitioners, who circulated and transcribed new and authoritative works, translated them into or out of Latin, and even prepared them for presentation to princely patrons. One outcome of such correspondences was the publication of the otherwise obscure Viaticum: a work linked to a great 56 'Ego idem vidi | Anno 1588 feb: | die 8. novo sti|lo Traeboniae | in Bohemia | Ex 2 lib Se|riconis dissol|uti in [vinegar] dis|tillato, et per | spiritum vini pur|ificatum a | multis English authority, translated by Norton for presentation to an English queen, harboured by Dee, and tested and revised by Edward Kelley.

Conclusion
In March 1589, disheartened by his lack of fortune and under pressure by Rožmberk to leave Třeboň , Dee departed from Bohemia with his family, leaving Kelley in possession of the field. Over the previous few years, Dee's preoccupation with the angelic conversations and their millenarian implications seems to have come increasingly into conflict with Kelley's practical interest in transmutation-an interest more congenial to his patrons, Rožmberk and Rudolf II, than the specious pledges of the showstone. From 1587, Kelley's attempts to extricate himself from his role as Dee's scryer were paralleled by his own increasing success as an alchemist (Dee, 1998, p. 210;Parry, 2011, chap. 17).
Although Dee was unable to muster the same level of support for his own projects, he continued to make some contributions to his colleague's practice. In an entry dated 28-29 October 1587, a few days after Kelley's return from a visit to Prague, he noted that 'Jo. Carp. did begyn to make furnaces over the gate &c.: & he vsed of | my rownd bricks' (Dee, 1582, fol. 102v). Kapr was apparently engaged in equipping Kelley's laboratory: we may recall Dee's earlier reference to Kelley's experiment conducted in 'his laboratory over the gate.' A month earlier, on 28 September, Kelley had visited Dee to ask that they split their supply of '[mercury] Animall', bringing weights with him for the purpose (ibid., fol. 101r). The 'Diary' leaves terse records of other alchemical activities, but by this stage Kelley was the dominant practitioner, while Dee, in a reversal of their earlier fortunes, increasingly performed the role of his assistant.
This reverse finds a parallel in Kelley's deft use of another English alchemist, Ripley, whose works offered a form of intellectual capital to be deployed in Kelley's own writings, correspondence and laboratory practice. Yet although works like the Bosome book informed his alchemical activities, the benefits of the association were not one-sided. The connection with a successful and charismatic practitioner, Kelley, attracted new interest in an author upon whom Kelley himself relied, smoothing the passage of Ripley's English works through the courts and presses of the Empire. In this trans-generational conference of philosophers, one authority supported another-yet Combach eventually published Kelley's letters before Ripley's texts.
Even after Dee's return to England, the three Englishmen would remain inseparably bound in print. Ralph Rabbards, magistrate and frustrated engineer, published Ripley's Compound in 1591: the first time that an English vernacular alchemical work had been printed in its original language. 64 In his dedication to Elizabeth I, Rabbards hailed the achievements of English adepts, '& especially M. Doctor Dee in his Monas Hyerogliphica.' 65 His praise was reserved for these adepts' 'depth of learning Theoricall', rather than practical: although he hints at the results that might be obtained if the work 'were yet executed by any experienced practitioner' (Ripley, 1591, sig. [A4]v).
While Rabbards doubtless had his own skills in mind, another such practitioner was surely Edward Kelley, whose absent pres-ence is recorded by the inclusion of a poem by 'Sr. E. K. concerning the Philosophers Stone, written to his especiall good friend, G.S. Gent' (ibid., sig. ⁄ 3r-v). This poem was already available in manuscript: in Copenhagen, Royal Library GKS 242 (fol. 37v), one of a group of documents connected with the Kelley circle in Prague (Bäcklund, 2006). This copy is titled 'The praise of vniti for frendships sake made by astranger to furder his frende his Conceyts. 1589', and signed 'Sir Edward Kelle' (ibid., pp. 300-301). Rabbards' change of title to include 'G.S. Gent.' may perhaps suggest a link with Dee's friend, 'Mr Gawyn Smithe gentleman': the royal projector and reader of the Bosome book. 66 If Smith was indeed the recipient of Kelley's poem, then we might speculate that he and Rabbards, as fellow engineers with a taste for alchemy, also shared an acquaintance-a possible indicator of the route by which Kelley's 1589 poem reached the English press only two years later.
The host of coincidences surrounding the publication of Ripley's works draws attention to the vigorous, scribal transmission of early modern alchemical texts. 67 The first editions of the Compound-by Rabbards, Penot, Barnaud and Combach-were not set in isolation, but lay enmeshed within webs of communication, authority and patronage: a pan-European network in which John Dee, practising alchemist and one of the great bibliophiles of Renaissance Europe, was an enthusiastic and influential participant. This network now survives only in fragmentary form: in friendly dedications, in marginal notes, and in the appearance of particular works in unexpected places. Such clues guide us to the routes by which Ripley's poem attained a level of success that Dee, indifferently successful petitioner to a host of European monarchs, might well have envied: written for an English king, printed for an English queen, and translated for a Holy Roman Emperor.