¹ ASE 7 (1978), 9–60Google Scholar (hereafter ‘Lapidary Traditions, part I’), to which the following additions and corrections should be made: p. 12, Riddle, J. M., Marbode of Rennes' (1035–1123)Google Scholar‘De Lapidibus’, Sudhoffs Archiv, Beiheft 20 (Wiesbaden, 1977)Google Scholar, provides an ‘improved text’ of Marbod's poem, with information about the manuscripts and misinformation about its literary status (‘as a medical treatise’) and relations to other works; p. 22, nn. 6–8: for ‘PL 101’, ‘PL 94’, ‘PL 91’ and ‘col. 471’ read ‘PL 100’, ‘PL 114’, ‘PL 111’ and ‘col. 470’ respectively; p. 28, the Ansileubus glossary was compiled at Corbie in France during Charlemagne's reign (see Lindsay, W. M., ‘Romensis. Callis’, Bull. Du Cange: Archivum Latinitatis Medii Aevi 2 (1925), 81–4Google Scholar; and review of Glossaria Latina 1–111, Bull. Du Cange 3 (1927), 95–100, at 97–8Google Scholar; and now Bishop, T. A. M., ‘The Prototype of Liber glossarum’, Medieval Scribes, Manuscripts and Libraries: Essays Presented to N. R. Ker, ed. Parkes, M. B. and Watson, A. G. (London, 1978), pp. 69–86, esp. 82–3; p. 32Google Scholar, text line 11, italicize initial T of Twelfta; line 19, add accent over second i in piríten; p. 41, sard from Epiphanius: cf. below, pp. 104–5; P. 42 topaz, ‘mystery’: see below, pp. 94–9; p. 50, that mocritum resulted from corruption of a Solinus text had been recognized by Zettersten, A. (‘The Source of *mocritum in Old English’, SN 41 (1969), 375–7); p. 51Google Scholar, Anglo-Saxon familiarity with complex carvings on gems: Dr Simon Keynes notes that a yet more complex intaglio is said by the thirteenth-century chronicler of St Albans to have been given to that monastery by Æthelred II (Annales Monasterii Sancti Albani, ed. Riley, H. T., Rolls Ser. (London, 1871) 11, 332–3)Google Scholar. A drawing of it by Matthew Paris is reproduced Henig, M., A Corpus of Roman Engraved Gemstones from British Sites, 2nd ed., BAR Brit. ser. 8 (Oxford, 1978), 160.Google Scholar

² The two unpublished tracts noted (‘Lapidary Traditions, part I’, p. 22, n. 2 and p. 29, n. 2) as possibly of independent significance are probably not so. The incipit and explicit of the Orléans Diversitates Duodecim Lapidum suggest that it is a slightly rearranged and incomplete version of a tract in Berne, Burgerbibliothek, A. 92. 27, 1r—3v, largely dependent on Bede; ed. Meier, C., ‘Zur Quellenfrage des ‘Himmlischen Jerusalem’: ein neuer Fund’, ZDA 104 (1975), 204–43Google Scholar. It seems not unlikely that Meier's text constitutes Sinner's Berne ‘glossary’. Since my part I there has appeared also Meier, C., Gemma Spiritalis I, Münstersche Mittelalter-Schriften 34.1 (Munich, 1977)Google Scholar, the first of two promised volumes on the allegorical use of precious stones by Christian writers up to the eighteenth century. Meier's massive work attains impressive near-fullness of patristic reference, but its chaotic arrangement (with vol. I unindexed) compounds a lack of historical control (especially in the early period), to which Meier's preferences for considering allegory in isolation from factual information, and exegetic in isolation from non-exegetic writings, inevitably expose her.

³ In ‘Lapidary Traditions, part I’, pp. 19–20, I repeated the accepted view that there is no clear trace of use of the Latin Damigeron in literature before Marbod. That view is false (see my forthcoming article in Ériu): descriptions of four marvellous stones in the tenth-century Irish apocryphon In Tenga Bithnua (‘The Ever-New Tongue’) use material derived unmistakably from Damigeron. This is not evidence that a text of Damigeron reached Ireland: the treatment of the material implies that it entered Irish learned tradition by word of mouth, and before the tenth century. The affinities of the antecedent text were with Pitra's recension, not with the alphabetic recension, which is consonant with the dating proposed for the latter in ‘Lapidary Traditions, part I’ (p. 18). All this to some extent enhances the possibility mooted ibid. pp. 48–9, that material from Damigeron was known to the author of the Old English Lapidary.

⁴ I use the following abbreviations: CCSL = Corpus Christianorum Series Latina; CSEL=Corpus Scriptorum Ecclesiasticorum Latinorum; DDL = De Duodecim Lapidibus; MLWL = Revised Medieval Latin Word-List from British and Irish Sources, ed. Latham, R. E. (Oxford, 1965)Google Scholar; OEL = Old English Lapidary; PW = Pauly's Realencyclopädie der classischen Altertumswissenschaft, ed. Wissowa, G. et al. (Stuttgart, 1894–1972)Google Scholar. I refer to the editions of Pliny, Isidore and Solinus cit. ‘Lapidary Traditions, part I’, pp. 10–11; of Damigeron, cit. ibid. pp. 13–14; of Epiphanius, cit. ibid. p. 20; of medical texts, cit. ibid. p. 55.

⁵ Laistner, M. L. W. and King, H. H., A Hand-List of Bede Manuscripts (Ithaca, N.Y., 1943), p. 25Google Scholar. Explanatio Apocalypsis is ed. J-P. Migne, Patrologia Latina (hereafter PL) 93, cols. 129–206. Bonner, G., Saint Bede in the Tradition of Western Apocalyptic Commentary, Jarrow Lecture 1966 [Jarrow, 1967], p. 2Google Scholar, states that a CCSL edition is in preparation.

⁶ Laistner and King, Hand-List, pp. 26–30, list over seventy manuscripts, of which a dozen are of the late eighth or early ninth century. Cf. Bonner, Saint Bede, p. 9.

⁷ PL 93, col. 203, quoted below.

⁸ And, doubtless for that reason, not set out as separate paragraphs by Migne: beginning ‘In iaspide ergo fidei viriditas …’ and ‘Cuius artifex et conditor Deus …’ (PL 93, col. 202c and end of col. 202D).

⁹ PL 93, col. 203 A–B: ‘I may perhaps seem to have set forth these remarks concerning the precious stones at greater length than belonged to the commaticum mode of interpretation. For it was necessary diligently to expound their natures and provenance, then very carefully to investigate their sacred meaning, not omitting to pay attention to their order and numbers. As touching, indeed, the real profundity of the matter, it seems to myself that I have said very little, and that briefly and superficially. I humbly beseech the reader that if I seem to him to have kept to the right path he will give thanks to God; but if he shall discover me to have ended otherwise than I wished, let him pray to the Lord for pardon for my error. But so much for these matters: let us look also at the remainder.’ (The last three and a half columns of the commentary then follow.)

¹⁰ As in part I, my statements are based on published texts. The analysis below of Bede's sources makes it impossible to suppose that this part of the picture will be significantly altered by any still unpublished.

¹¹ Victorini Episcopi Petavionensis Opera, ed. Haussleiter, J., CSEL 49 (Vienna, 1916), 146–54.Google Scholar

¹² Reworking Victorinus: CSEL 49, 147–53 and 47.

¹³ Whether the fragments ed. Lo Bue, F., The Turin Fragments of Tyconius's Commentary on Revelation, Texts and Stud. n.s. 7 (Cambridge, 1963)Google Scholar, or what can be deduced by the comparative method of Bonner, Saint Bede.

¹⁴ PL 68, cols. 793–936, at 927 and 814.

¹⁵ S. Caesarii Arelatensis Opera Omnia 11: Opera Varia, ed. Morin, G. (Maredsous, 1942), pp. 274 and 219.Google Scholar

¹⁶ Apringius de Béja: Son commentaire de l' Apocalypse, ed. Férotin, M. (Paris, 1900), pp. 76–7 and 27–8.Google Scholar

¹⁷ Beati in Apocalipsin Libri Duodecim, ed. Sanders, H. A. (Rome, 1930), pp. 268–70.Google Scholar

¹⁸ Ein Traktat zur Apokalypse des Apostels Johannes, ed. Hartung, K. (Bamberg, 1904)Google Scholar, which McNally, R. E. (The Bible in the Early Middle Ages (Westminster, Maryland, 1959), pp. 116–17)Google Scholar dates c. 800.

¹⁹ PL 100, cols. 1087–1156; attribution to Alcuin, made in the late eleventh century, denied Wilmart, A., Auteurs spirituels et textes dévots du moyen âge latin (Paris, 1932), p. 52, n. 6Google Scholar. The text is defective before the point at which the twelve stones might be treated.

²⁰ Jerome (quoted below, p. 78) makes them collectively symbolize virtues. Most of the other commentators follow him.

²¹ Victorinus, stating that ‘iaspis aquae color est et sardi ignis’ (‘the colour of jasper is that of water, and of sard that of fire’) (CSEL 49, 46), makes them symbolize flood and fire of past and future judgement. Most later writers follow this. Apringius adds that green jasper and red sard symbolize the divine and human natures of Christ, an idea adopted by several later writers (see now Meier, Gemma Spiritalis 1, 121–2), but not by Bede, who in commentary on Revelation iv.3 (PL 93, col. 143A) merely abbreviates Victorinus. Pseudo-Alcuin has the original idea that jasper denotes the green pastures of Paradise (PL 100, col. 116c; noted Dronke, P., ‘Tradition and Innovation in Medieval Western Colour Imagery’, Eranos Jahrbuch 41 (1972), 51–106, at 84)Google Scholar. This is taken up by Haymo of Auxerre (PL 117, cols. 937–1220, at 1204–8; McNally (Bible, p. 117) dates c. 865), who otherwise relies on Ambrosius and Bede. Ideas about sard and jasper current more generally (as expressed mainly in visual art, as well as in textual exegesis) are discussed Schade, H., ‘Der Stein unter dem Throne Gottes: Hinweise zur Symbolik frühmittelalterlicher Kunst’, Geist und Leben 36 (1963), 115–27.Google Scholar

²² Ambrosii Autperti Expositio in Apocalypsin, ed. Weber, R., CCSL Continuatio Medievalis (hereafter CM) 27–27A (Turnhout, 1975), at 27A, 818–23, and 27, 207–10.Google Scholar

²³ Expositio super Septem Visiones Libri Apocalypsis, PL 17, cols. 843–1058, at 1204–8; dated McNally, Bible, p. 117.

²⁴ Except for a snippet of Gregory the Great on sardonyx (quoted below, p. 80), which Ambrosius applies to sard, CCSL, CM 27A, 821 (twice quoting it), and 27, 208.

²⁵ He does not depend on any of the Latin writings I have encountered, including Pliny (unless very indirectly) and Isidore. The general homogeneity in style and shortness of his descriptions suggests a single source as against more than one.

²⁶ Since he describes topaz as golden. (On reputed colours of topaz see below, end of n. 162.) The Greek source was not, however, Epiphanius, with whom (quoted ‘Lapidary Traditions, part I’, p. 42) he disagrees in saying, for example, ‘Sardius fulvi coloris est, qui pro eo quod in rubicundum colorem aliquantulum declinare videtur, parumper obscuratur’ (‘Sard is of a tawny colour, which is slightly obscured, inasmuch as it is seen to shade off somewhat into a reddish colour’) (PL 17, col. 1042B).

²⁷ For instance, ‘Chalcedonius rubeum habet colorem, sed tamen obscurum, nisi cavando attenuetur: ipsa vero concavitas pulcherrimum colorem et gratissimum ei confert’ (‘Chalcedony has a red colour but, however, an obscure one, unless refined by hollowing out: then the very concavity confers upon it a most beautiful and gratifying colour’) (PL 17, col. 1042B). It would be most interesting to discover where Berengaud got his information. Meier, Gemma Spiritalis 1, 56, seems to imply that her vol. 11 will have something significant to say about Berengaud; the section ‘Bearbeitbarkeit / Bearbeitung / Gravuren’ (1, 281–99) makes scant mention of him. Friess, G., Edelsteine im Mittelalter (Hildesheim, 1980)Google Scholar, which has (unsystematized) leanings towards the tangible, is likewise unhelpful.

²⁸ Stated at length, but not in so many words, In Esaiam xii–xviii, ed. M. Adriaen, CCSL 73A (Turnhout, 1963), 612.

²⁹ See below, pp. 85–6.

³⁰ The only significant overlap seems to be that Bede's red sard symbolizes martyrdom, while Berengaud's tawny sard tending towards red denotes a life of tribulation ending in martyrdom.

³¹ See Haupt, G., Die Farbensymbolik in der sakralen Kunst des abendländischen Mittelalters (Dresden, 1941), pp. 43–9Google Scholar and passim; Dronke, ‘Tradition and Innovation’, pp. 67–9 and works cited pp. 51–2. The long chapter on ‘Farben’, Meier, , Gemma Spiritalis 1, 142–236Google Scholar, is less satisfactory.

³² See principally André, J., Étude sur les termes de couleur dans la langue latine, Etudes et commentaires 7 (Paris, 1949), 260–3Google Scholar and passim; and Heijn, N. V., Kleurnamen en kleurbegrippen bij de Romeinen (Utrecht, 1951)Google Scholar, passim; Dronke, ‘Tradition and Innovation’, pp. 61–6.

³³ Bonner (Saint Bede, p. 10) lists some.

³⁴ In Esaiam xv, CCSL 73 A, 608–14. On this and on biblical jewels generally see ‘Lapidary Traditions, part I’, pp. 20–1.

³⁵ PL 93, cols. 197B–C: ‘And the foundations of the wall of the city were garnished with all manner of precious stones (Rev. xxi.19). By the names of the several stones are indicated variously the forms, or the order, or the diversity of the virtues with which the heavenly Jerusalem is all built. For it is difficult for persons in isolation to flower with all the virtues. Furthermore, Isaiah, when describing the beauty of the said city, saying, Behold, I will lay thy stones with fair colours, and lay thy foundations with sapphires, etc. (Isa. liv. 11), immediately added as if by way of exposition, All thy children shall be taught of the Lord (Isa. liv. 13).’

³⁶ Above, n. 20. Bede himself uses it elsewhere – e.g., De Tabernaculo 111, ed. Hurst, D., Bedae Venerabilis Opera Exegetica, CCSL 119A (Turnhout, 1969), 103Google Scholar.

³⁷ E.g. a quotation from the Clouds of Aristophanes in the section on sapphire, CCSL 73A, 612.

³⁸ PL 93, col. 197c: ‘The first foundation was jasper (Rev. xxi.19). There are many varieties of jasper. For one is of green colour, and appears stained as if with flowers. Another has the likeness of emerald, but of a thick colour: by it, they say, all apparitions can be put to flight. Another gleaming somewhat redly as if with a colour mixed with snow and the foam of ocean waves.’ This last sentence makes so much less sense than Jerome's that one wonders if Migne's text of Bede is corrupt at this point.

³⁹ CCSL 73A, 613. Jerome's whole sentence translates ‘There is also another jasper resembling snow and the foam of ocean waves, of a substance gleaming somewhat redly as if mingled with blood.’

⁴⁰ ‘By jasper therefore is signified the greenness of faith unwithering …’

⁴¹ CCSL 73A, 613 (Song of Songs v.10), 609 (Isaiah). Bede alters Jerome's Old Latin quotation to agree with the Vulgate wording.

⁴² PL 93, cols. 197D–8A.

⁴³ CCSL 73A, 612.

⁴⁴ In Danielem, CCSL 75A, 891–2.

⁴⁵ PL 93, col. 200A: ‘Whence, also, it is called tharsis among the Hebrews from its likeness to sea colour.’

⁴⁶ CCSL 73A, 610.

⁴⁷ PL 93, col. 202D; quoted fully below, pp. 99–100.

⁴⁸ ibid. col. 199c: ‘It possesses the likeness of red earth.’

⁴⁹ PL 76, col. 82c. (I owe the identification of this source to Meier, Gemma Spiritalis 1, 121.) Bede probably also used Augustine directly: see below, n. 172.

⁵⁰ On whom see ‘Lapidary Traditions, part I’, pp. 20–1.

⁵¹ PL 93, cols. 197D–8A, followed by material from Jerome as noted above: ‘The second, sapphire (Rev. xxi. 19). The colour of this stone together with the mystery of it Moses expounded when, describing the appearance of God, he said, Under his feet as it were a paved work of a sapphire stone, and as it were the heaven when it is clear (Exod. xxiv. 10).’

⁵² CSEL 55, 71 / Migne, Patrologia Græco-Latina (hereafter PGL) 43, col. 334: ‘Moreover, it is written in the Law that the vision of Moses was revealed on the mountain as if upon a sapphire stone.’ The mention of Moses is not taken up by Jerome (CCSL 73 A, 612 is the potential context), nor could Bede have derived it from Gregory the Great (whose references to sapphire are adequately quoted Meier, Gemma Spiritalis 1, 133–4, nn. 466–7).

⁵³ Quoted below, p. 100.

⁵⁴ The influence here is from Epiphanius' second section and will be discussed below, p. 93.

⁵⁵ CSEL 35, 755 / PGL 43, col. 340: ‘Like the sea's tint … moreover, should one wish to hold it up against the sun …’

⁵⁶ See below, p. 82. Damigeron (Cava §59) has the similar idea of holding beryl up against the sun, but compares it to oil, not to any kind of water. Garrett, R. M. (Precious Stones in Old English Literature (Leipzig, 1909), pp. 10–11)Google Scholar asserts that the beryl was known to Bede by sight and suggests that for his description of it he may have relied on his own observation. The text which Garrett took as evidence (PL 95, cols. 313–15 (quoting col. 314)) is, however, rather obviously not by Bede. It is the account by Thomas of Elmham, in the fifteenth century, of the earliest books in St Augustine's, Canterbury, discussed James, M. R., The Ancient Libraries of Canterbury and Dover (Cambridge, 1903), pp. lxiii–lxviiGoogle Scholar, and re-ed. ibid. pp. 500–2.

⁵⁷ CSEL 35,755 / PGL 43, col. 340: ‘The stone chrysolite … gives forth a sheen of gold.’

⁵⁸ PL 93, col. 199D: ‘The stone chrysolite shines like gold.’

⁵⁹ ‘Pinguedo rosea’ (PL 93, col. 200c). The ‘rosy’ colour would be Bede's variation on Epiphanius §10 rubrum (‘ruddy’). The word pinguis is used in his main source for topaz (quoted below, p. 96). Meaning literally ‘fat’, it is sometimes used of colour by classical writers (e.g. Pliny, Nat. Hist. XXXVII. 115, on jasper), rather as we might say ‘thickish’ as well as ‘oily’. The related noun which bears that transferred meaning is however not pinguedo but pinguitudo (e.g. Pliny, ibid. 105, on sard). Pinguedo is used normally of animal lard or resinous substances of a similar consistency. (The most useful dictionary for these words is Forcellini's Lexicon Totius Latinitatis.) Bede's use here of pinguedo rather than pinguitudo is to be rationally explained as conditioned by Epiphanius §11 (CSEL 35, 746 / PGL 43, 325–6), where topaz ground in a mortar is said to exude a fluid substance with medicinal uses. (MLWL cites Bede as using pinguedo in an additional sense glossed ‘abundance’, too abstractly: it relates to fatty substances and/or good things to eat (the ‘fat of the land’ familiar in the AV). PL 93, col. 198D (‘internae pinguedine charitatis’), expounding both a literal oil and a figurative wine, is an example.)

⁶⁰ See below, n. 121.

⁶¹ The Greek notes in his Expositio Actuum Apostolorum are of bare textual variants and the meanings of single words. They imply no working knowledge of the language. Later in life, when he had acquired such a working knowledge, he reworked the same text in his Retractatio, which frequently uses Greek inflectional forms and grammatical connections to clarify the equivalent Latin text. (Bolton, W. F., ‘An Aspect of Bede's Later Knowledge of Greek’, Classical Review 77 (1963), 17–18CrossRefGoogle Scholar, misguidedly ignores this.) The Expositio was composed after Explanatio Apocalypsis (Laistner, Hand-List, pp. 20–5). It precludes the use of a Greek text as source for the latter. Bede's commentaries on Acts are ed. Laistner, M. L. W., Bedae Venerabilis Expositio Actuum Apostolorum et Retractatio (Cambridge, Mass., 1939)Google Scholar. On the Greek in them see further ibid., esp. p. xv; Jenkins, C., ‘Bede as Exegete and Theologian’, Bede, his Life, Times and Writings, ed. Thompson, A. H. (Oxford, 1935), pp. 152–200, esp. 157–65Google Scholar; cf. Meyvaert, P., ‘Bede the Scholar’, Famulus Christi: Essays in Commemoration of the Thirteenth Centenary of the Birth of the Venerable Bede, ed. Bonner, G. (London, 1976), pp. 40–69, at 48–51.Google Scholar

⁶² In fact Bonner (Saint Bede, p. 10) denies it.

⁶³ See ‘Lapidary Traditions, part I’, pp. 20–1.

⁶⁴ Blake, R. P. and de Vis, H., Epiphanius on the Twelve Stones (London, 1934), p. cxxii.Google Scholar

⁶⁵ Blake, , Epiphanius, pp. 122–72.Google Scholar

⁶⁶ ibid. p. cxxiii.

⁶⁷ Outlined above, p. 77; cf. below, p. 88.

⁶⁸ PL 93, col. 200B: ‘Beryl is as if you were to look upon water struck by the brilliance of the sun, giving forth in return a pleasant reddish colour. But it does not shine unless cut into a hexagonal shape in polishing. For its brightness is enhanced by the reflection of the angles. [Migne prints repercussus, which is ungrammatical and certainly wrong. Pliny has repercussu.] Now it signifies those men who are wise of intellect, but shine forth even more with the light of grace from above. For that water stands for depth of perception, Solomon is witness, who said, The words of a man's mouth are as deep waters (Prov. xviii.4).’

⁶⁹ See, e.g., Smith, G. F. H., Gemstones, 13th ed., rev. Phillips, F. C. (London, 1962), pp. 303–7Google Scholar; cf. Meier, , Gemma Spiritalis 1, 167–70, 172 and 177–9.Google Scholar

⁷⁰ Nat. Hist. XXXVII.76–9.

⁷¹ Neither can Meier (Gemma Spiritalis 1, 191, n. 241) find any.

⁷² In the Latin translation: ‘Beryllium lapis glauci, id est caesii, coloris est, marinae tincturae similis et aeris, amethysti et paederotis habens speciem et aquatioris, id est albidioris, hyacinthi’ (CSEL 35, 755 / PGL 43, col. 340). In the shortened Greek version: ‘Λίθος βηρύλλιον γλαυκίξων μέν ⋯στι, θαλασσοβαøής, ἔχων εἷδος καì τ⋯ς ύδαρεστέρας ύακίνθου’ (ed. de Mély, F., ‘Les Lapidaires grecs: Textes’, Les Lapidaires de l'antiquité et du moyen âge 11 (Paris, 1898), 197.Google Scholar)

⁷³ Blake, Epiphanius, P. 155.

⁷⁴ ibid. pp. 163–4, correcting the obvious error reported by Blake in the extant Georgian, ‘the bloodstained garment of Jacob’.

⁷⁵ See Haupt, Farbensymbolik, esp. pp. 84–99 and 137. The principal exegetic uses of red were as the colour of earth (arising from interpretation of the name Adam as ‘red earth’), or the blood of martyrs, or the fire of love or of the Holy Spirit. The first three of these are applied to gems by Bede. (A reverse example of the second appears in the Old English vernacular: ‘Drihten … sealde his þone readan gim, þæt wæs his þæt halige blod’ (‘The Lord … gave us his red jewel, that was his holy blood’) (The Blickling Homilies, ed. Morris, R., EETS o.s. 58, 63 and 73 (1874–1880, repr. London, 1967), 9–11)Google Scholar.)

⁷⁶ E.g. De Tabernaculo in, CCSL 119A, 100–1. That gold (the substance) could in Bede's vernacular be called read is thus irrelevant to his operation of Latin colour distinctions.

⁷⁷ On the standard nature of the equation see Haupt, Farbensymbolik, esp. pp. 67–70, with some quotations pp. 134–5. Meier's references are scattered, Gemma Spiritalis I, 197–8, 200, 210–11 and 227; as at pp. 143–5 and 530, her presentation is such as to belittle Haupt's contribution to learning.

⁷⁸ I have not found an exact example of it in earlier writers. Jerome (In Hiezechielem, CCSL 75, 394) comes close, referring gold in treasuries to ‘sense and mind’ as gathering up riches in heaven. I have not attempted a complete investigation, and suspect that exact examples are to be found.

⁷⁹ As Ambrosius clearly assumed, for he says: ‘Quia vero auri vocabulum sapientiam figuret, Scriptura testatur quae dicit, Accipite sapientiam sicut aurum’ (CCSL, CM 27A, 819; the quotation again, ibid. p. 822): ‘That the word gold indeed signifies wisdom, Scripture bears witness, which says, Receive wisdom like gold (Prov. xvi.16).’ The Vulgate reads ‘Posside sapientiam, quia auro melior est’ (= AV ‘Better is it to get wisdom than gold’).

⁸⁰ This emerges from Meier's work, perhaps surprisingly, and contradicting her apparent intention. Throughout her chapter on ‘Farben’ I cannot find a single instance of another colour in a gem signifying wisdom. (Wisdom is not, of course, the only thing symbolized by a golden colour.)

⁸¹ See esp. below, p. 93.

⁸² Even Haymo, who draws largely on Bede (see above, n. 21), only takes over in this regard the gryphons and Arimaspians (see below, p. 86).

⁸³ PL 93, cols. 198A, c, 201A, c and 198C–9B.

⁸⁴ ibid. col. 201D.

⁸⁵ Allegorizing: Blake, Epiphanius, pp. 123, 131, 132, 137, 138(?), 141, 146, 152 and 154; non-allegorizing: pp. 124 and 164.

⁸⁶ According to Exodus xxviii.21 the twelve jewels of the high priest's breastplate were each to be engraved with the name of one of the twelve tribes. Epiphanius goes beyond the bible in stating which tribe's name was engraved upon each gem, their order being matched to that of the birth of Jacob's sons (Genesis xxix.31–xxx.24 and xxxv. 16–18). Bede goes so far as to mention the engraving and give it a figural significance in his final summing-up (quoted below, p. 100). Even this was unusual among Latin commentators (on the Apocalypse and, it would seem, more generally). Jerome's compendious passage on biblical jewels quotes the Exodus verse but does not elaborate on it (In Esaiam, CCSL 73 A, 611). I see no mention of the tribes of Israel in Meier, Gemma Spiritalis 1.

⁸⁷ Conspicuously, Blake, Epiphanius, pp. 143, 149, 150, 153 and (167–)170; cf. pp. 147 and 158.

⁸⁸ ibid. p. 132 (with punctuation made more coherent).

⁸⁹ Set forth, e.g., by Pliny, Nat. Hist. II. 182–90, esp. 189. The idea would of course have been familiar to an educated man like Bede. It may have been current in popular belief as well, at any rate later in the Anglo-Saxon period, for it appears in vernacular literature in Riddle vi of the Exeter Book. The most vivid image of the cosmological process is Bede's own in De Temporum Rations xxxiv (Bedae Opera de Temporibus, ed. Jones, C. W. (Cambridge, Mass., 1943), esp. p. 246)Google Scholar.

⁹⁰ PL 93, col. 201D: ‘Now it signifies souls continually given to heavenly devotion, and in a manner, as far as is permissible to mortals, approaching the conversation of the angels.’

⁹¹ ‘Quantum mortalibus fas est’ in this context may well suggest to a modern reader the legend of Daedalus and Icarus. In view of Bede's general unfamiliarity with the pagan classics (demonstrated Blair, P. Hunter, ‘From Bede to Alcuin’, Famulus Christi, ed. Bonner, , pp. 239–60)Google Scholar, he is unlikely to have known it. He would have had in mind, though (as he was to in De Temporum Ratione xxxiv), the human difficulty of approaching too close to a large fire.

⁹² Ambrosius: ‘In quo scilicet et caelestis vitae conversatio, et in his quae agenda sunt vera discretio, atque in Dei ac proximi amore pia caritatis coniunctio declaratur’ (‘In which is shown forth both the conversation of the heavenly life, and true discretion in necessary actions, and a pious conjunction of charity in the love of God and one's neighbour’) (CCSL, CM 27A, 823). Berengaud: ‘Significat autem sanctos viros ab omnibus sordibus peccatorum alienos, terram corporibus, coelum mentibus semper inhabitantes; de qualibus fuit Apostolus, qui dicebat, Nostra autem conversatio in coelis est’ (‘Now it signifies holy men estranged from all defilements of sins, dwelling on earth in their bodies but always in heaven in their minds. One such was the Apostle, who said, For our conversation is in heaven (Phil. in.20)’) (PL 17, col. 1044A–B). The image was also absent from Bede's use of this verse from Philippians for sapphire (see above, p. 79).

⁹³ That of topaz in the Thebaid (see below, p. 95) signifies primarily that monastic communities most abound in Egypt, but also that ‘quicunque Soli iustitiae vicinus habitaverit, aetherei nimirum luminis fulgore coloretur’ (‘Whoever shall have lived in proximity to the Sun of Justice shall surely be coloured by an ethereal brilliance of light’) (PL 93, col. 201 A). This adaptation is necessary to fit the colour of topaz. (Purists may care to note that Bede does not unambiguously say those tinged by it will take on that colour.) That of chrysoprase ‘in India, that is near the sun's rising’ signifies martyrs ‘because they shine like the sun in the Father's kingdom’ (‘in India, id est prope solis ortum … quia velut sol in regno Patris fulgere’) (ibid. col. 201C).

⁹⁴ CSEL 35, 753–4/PG L 43, cols. 338–9.

⁹⁵ Quoted below, p. 90.

⁹⁶ Compare PL 93, col. 199A-B, ‘Contra huiusmodi alites semen divini verbi nobis praeripere anhelantes, quique sanctorum invigilant simplici per coeleste desiderium intentione, quasi uno oculo admirandi, ut gemmam fidei caeterarumque virtutum investigare et effodere queant’ (‘Against winged beings of this sort panting with desire to snatch away from us the seed of the divine word, all the saints keep watch through yearning for heaven with undivided attention, as if gazing forth with a single eye that they may discover and dig up the gem of faith and the other virtues’), with Blake, Epiphanius, p. 141: ‘They converted men from unbelief to the true faith, as does the hyacinth gem’ (Epiphanius has identified ligure with jacinth), ‘which is in the depth of the abysses, and none hath the power to take it out save by the blood of the lamb and the rushing pinions of the eagle. Such also is the holy daring of Elijah, say I, and of the three youths, who were in Babylon, and of the most blessed Thekla, who was brave in the time of her trial: not that it was their power which brought about salvation, but the coming of the Lamb, who removed the sins of the world, and like to the eagle in the powers of His Godhead, His ascension into heaven and His going up.’

⁹⁷ It seems probable, from the spread of material presented by Laufer, The Diamond, pp. 6–21 (see ‘Lapidary Traditions, part I’, p. 24, n. 4; cf. pp. 49–50), that the two stories are ultimately of common origin. If Bede guessed that, plainly he preferred the encyclopaedists as sources of information to the erratic Epiphanius.

⁹⁸ See ‘Lapidary Traditions, part I’, pp. 20–1.

⁹⁹ Siegmund, P. A., Die Überlieferung der griechischen christlichen Literatur (Munich, 1949), p. 143.Google Scholar

¹⁰⁰ The end is in fact defective, and the surviving text has not been perfectly transmitted: e.g. Blake (Epiphanius, pp. cxiv–cxv) plausibly postulates that in at least one place there is a gap in all the extant versions. The loss of one exact section from the middle of a work still requires explanation. The likely answer is that at an early stage in its compilation the ‘Collectio’ took the form of several separate booklets or gatherings small enough for the second and third sections (at least) of Epiphanius to need whole booklets to themselves, and that that containing the second section was mislaid before the collection came to be copied as a unit. Mayr-Harting, H., The Coming of Christianity to Anglo-Saxon England (London, 1972), pp. 172–3Google Scholar, points to definite evidence from about the time and place for which this would have to be postulated that ecclesiastical writings of similarly small unit length were kept in just this manner.

¹⁰¹ See ‘Lapidary Traditions, part I’, p. 24, and cf. pp. 41–2.

¹⁰² PL 93, col. 199D: ‘Rightly is it placed in sixth position, since our Lord was both incarnate in the sixth age of the world and crucified on the sixth day of the week for the salvation of all the world.’

¹⁰³ Nat. Hist. XXXVII. 120: ‘apud Medos’ (‘among the Medes’). ‘Red Sea’ in antiquity and in medieval times normally comprised both our Red Sea and the Persian Gulf. It is, for example, defined so by Pliny (Nat. Hist. vi. 107–8) and by Bede himself (De Natura Rerum§42 (PL 90, cols. 261–2 / Btdae Venerabilis Opera 1: Opera Didascalica, ed. Jones, C. W., CCSL 123A (Turnhout, 1975), 226)Google Scholar) and so shown by red colouring in the world map on fol. 56v of the eleventh-century English manuscript British Library, Cotton Tiberius B. v (not untypically of medieval world maps). On Bede's need for this confirmation see below, p. 91.

¹⁰⁴ Nat. Hist. xxxvii. 103–4 and 95 (see below, n. 125 and p. 92).

¹⁰⁵ ibid, xxxvii.65 and VII. 10.

¹⁰⁶ ibid, xxxvii.86–7.

¹⁰⁷ ibid, xxxvii.76.

¹⁰⁸ ibid, xxxvii. 121 (see below, n. 129).

¹⁰⁹ Variously formulated Jones, Bedae Opera de Temporibus, p. 129, and idem, ‘Bede's Place in Medieval Schools’, Famulus Christi, ed. Bonner, pp. 261–85, at 281; based principally on an unsound argument (Bedae Opera de Temporibus, p. 359) purporting to show that Bede did not know bk xviii. Whether or not he knew the whole Natural History is unproven, but he has been shown (see, e.g., Laistner, , Expositio et Retractatio, pp. 133 and 140Google Scholar) to have used other parts often denied him, and has not been shown to have lacked any part of it. I hope to return to these questions in detail elsewhere.

¹¹⁰ Etym. xvi.ix.3.

¹¹¹ ibid, xvi.ix.1.

¹¹² ibid. xvi.vii.1(–2) and xii.ii.17.

¹¹³ ibid, xvi.viii.4.

¹¹⁴ ibid, xvi.vii.8 and xvi.xv.1; see above, p. 79.

¹¹⁵ Etym. xvi.vii.7.

¹¹⁶ Ed. Mommsen, pp. 86–8.

¹¹⁷ ibid. p. 195.

¹¹⁸ ibid. p. 136, saying that the best type ‘sweetly bears the flowers’ of both translucence ‘and purple tint’ of two extreme varieties.

¹¹⁹ PL 93, col. 201D: ‘And, purified, sweetly bears the flower.’ This suits Bede's context so much better than does the original phrase that we may safely trust Migne for the reading. (Neither of the derivative texts - Haymo, Expositio in Apocalypsin (see above, n. 21) and Hrabanus Maurus, De Universo (see ‘Lapidary Traditions, part I’, p. 22) – bears on it.) On corruption of Solinus cf. ibid. p. 54.

¹²⁰ PL 93, cols. 198C-9A: ‘The emerald is of such intense green that it surpasses all green plants, leaves and jewels, making the air around it vibrant with greenness.† It also, appropriately, improves oil in greenness†, however much it be saturated by nature. There are many varieties, but the most valuable are the Scythian. Second place is held by the Bactrian kind, third place by the Egyptian. It signifies those souls which are ever green in faith … The same stone's country of origin, too, most beautifully fits them [the souls]. The land is rich, but uninhabitable. For though it is overflowing with gold and gems, all is under the sway of gryphons, the most ferocious avians, or rather flying beasts. For they are quadrupeds with the body of a lion but the head and wings of an eagle. The Arimaspians – notable for having one eye, in the middle of their foreheads – contend with them to seize these stones, the creatures raging and the Arimaspians hoarding them, both with remarkable cupidity.’

¹²¹ Migne's text of Bede reads merito (‘rightly, appropriately’). Meier (Gemma Spiritalis 1, 293) holds the true reading to be mero, as in Solinus and Isidore. Mero in Hrabanus Maurus (PL in, col. 466D) supports her, but pending a critical edition of Explanatio Apocalypsis it may be wiser to suspend judgement, since this may not be simply a question of textual transmission. Reading mero, one might translate the obelized clause: ‘It also improves (or is improved by) unmixed oil in greenness (or unmixed green oil)’ or ‘It is improved in green by wine and oil’ or ‘It is improved by wine and green oil’. Any or all of these translations can be justified from Bede's Latin and the authorities at his disposal. I hope to publish elsewhere a detailed discussion of the questions of wording and geology involved.

¹²² What Pliny wrote was ‘mira cupiditate et feris custodientibus et Arimaspis rapientibus’ (‘the creatures guarding them and the Arimaspians plundering them, both with remarkable cupidity’). The reversal of the two verbs in Explanatio Apocalypsis probably results from scribal accident in a text of Pliny. The gryphons whose land produced the gems had no need to raid the Arimaspians for them: understood in their classical sense, the reversed verbs are inconsistent with the story. But the voicing of intervocalic unvoiced stops in late imperial and early post-imperial spoken Latin led scribes of that period not infrequently to write þ for b. The confusion of spellings which resulted lasted for centuries in some words, including the verbs rapere and rabere and their derivatives. Thus in the travellers’ tales, similar in spirit, of the Epistola Alexandri ad Aristotekm (ed. Boer, W. Walther (The Hague, 1953), p. 6, line 7, etc.Google Scholar), the catch-phrase rabida ferarum genera ‘savage animal species’ often reads rapida in the manuscripts (earliest ninth-century; but see Sisam, K., Studies in the History of Old English Literature (Oxford, 1953), p. 83, n. 3Google Scholar, on their antecedents). In the context of gryphons, Bede and his contemporaries probably understood rapientibus as rabientibus and took custodientibus as conveying that the Arimaspians took good care of the gems once they had got hold of them. Bede may have had his doubts about the true reading: his figural interpretation (quoted above, n. 96) seems calculated to accommodate both.

¹²³ Set out ‘Lapidary Traditions, part I’, pp. 35–7, with general discussion pp. 39–41.

¹²⁴ PL 93, col. 199D: ‘Sard, which is altogether of bloody colour.’ Cf. discussion ‘Lapidary Traditions, part I’, pp. 41–2.

¹²⁵ PL 93, col. 198B: ‘Chalcedony shines forth like the fire of a lantern of pale aspect.’ Cf. discussion ‘Lapidary Traditions, part I’, p. 43. The words pallenti specie and the prefix re- of renitet seem to have been occasioned by Pliny's report (Nat. Hist. xxxvii.104), repeated by Isidore (Etym. xvi.xiv.5), that the stones are most often found when reflecting the full moon.

¹²⁶ PL 93, col. 199D: ‘The stone chrysolite shines like gold, having burning sparks.’

¹²⁷ See above, pp. 80–1.

¹²⁸ Alphabetic recension, §17: ‘Chrysolitus lapis lucidissimus est et auro similis, et scintillat velut ignis.’ Pitra's recension, §47, has for this just ‘Est spissus, lucidus, similis auro’; but the longer original, from which both versions derive, may safely be assumed to have contained a phrase at least similar to ‘scintillat velut ignis’. On that original's availability in England by Bede's time see ‘Lapidary Traditions, part I’, pp. 18–19.

¹²⁹ Operating sometimes at the level of single words – e.g. PL 93, col. 202B, ‘Purpureus ergo decor, coelestis regni habitum, roseus vero atque violaceus’, where Bede works in two words from Pliny xxxvii.121 not used in his slightly longer quotation from it immediately preceding.

¹³⁰ See ‘Lapidary Traditions, part I’, p. 44. Cf., generally, Damigeron's first sentence on melas, which in Pitra's §40 has Stellas, in Alphabetic §37 scintillulas. The difference has probably arisen by scribal corruption, but it is not possible to say which reading, if either, preserves the original.

¹³¹ PL 93, col. 200C: ‘it is said to burn the holder's hand’.

¹³² ibid. col. 201C: ‘Chrysoprase is green with an admixture of gold, carrying with it a certain purple sheen.’

¹³³ ibid. col. 198B: ‘has brilliance in the open but not indoors’.

¹³⁴ ibid. col. 198A: ‘Which, struck by the rays of the sun, emits from itself a burning brilliance’; ‘that it is said to be found in the Red Sea’.

¹³⁵ See above, pp. 82–3.

¹³⁶ See below, p. 106.

¹³⁷ See above, n. 103.

¹³⁸ Nor by Epiphanius: Meier, Gemma Spiritalis 1, 212, n. 343, is misleading.

¹³⁹ See Heijn, Kleurnamen, pp. 26–35 and 117; André, Termes de couleur, pp. 90–102.

¹⁴⁰ Nat. Hist. XXXVII.95.

¹⁴¹ See ‘Lapidary Traditions, part I’, p. 43, n. 6.

¹⁴² Quotation above, p. 80. Mayr-Harting (Coming of Christianity, p. 213 (cf. pp. 313–14)) points to a restatement of this same deduction from the same verse of Exodus in Bede's commentary on the Song of Songs (PL 91, col. 1167D). Bede was the first Latin exegete to comment on the colour of sapphire in that context.

¹⁴³ See André, Termes de couleur, pp. 135–6. The existence of these connotations would have been apparent in at least some of Bede's Latin reading (e.g., Pliny, Nat. Hist. xxxvii. 93–6refers repeatedly to the fulgor of carbuncles; Isidore, Etym. xiii.ix.2 distinguishes fulgor from fulgus and fulmen ‘quia incendit et urit’ ‘because it ignites and blazes’), and his use of fulgor elsewhere in the jewel passage suggests he was aware of it: once in connection with the fire of a lantern (see above, pp. 90–1), several times of the sun (see above, p. 91 and n. 93) and figurally associated qualities. (Fulgor of the sun is a common classical usage.)

¹⁴⁴ This is obvious from its place in the order and may be confirmed by, for example, comparing the geographical details (Blake, Epiphanius, p. 131) with CSEL 35, 750/PGL 43, cols. 332–3.

¹⁴⁵ Epiphanius would have used the Septuagint bible, so the Latin Epiphanius which Bede read must have had something resembling the Latinized quotation of the same Septuagint verse by Jerome in the passage Bede had already used (CCSL 73A, 608 and 609): ‘Ecce ego praeparabo tibi carbunculum lapidem tuum, et fundamenta tua sapphirum’ (‘Behold, I shall prepare for thee thy stone the carbuncle, and thy foundations sapphire’).

¹⁴⁶ Blake, Epiphanius, p. 129.

¹⁴⁷ As tenth main stone, ibid. p. 153.

¹⁴⁸ Cf. ‘Lapidary Traditions, part I’, p. 22.

¹⁴⁹ CSEL 35, 750 / PGL 45, col. 333. To the ancients ‘sapphire’ apparently meant lapis lazuli and ‘jacinth’ what we call sapphire: see Eichholz's notes on Pliny (Loeb ed., x, 262 and 266). To Pliny (Nat. Hist. xxxvii.120) sapphire was normally blue but could be purple. Solinus does not mention it. Isidore (Etym. xvi.ix.2) regards it as mixed and classifies it with purple gems. The Western exegetic tradition, from Jerome, found blue more congenial as a symbolic colour than purple (see e.g. Haupt, Farbensymbolik, pp. 99–117; Dronke, ‘Tradition and Innovation’, pp. 67–8); hence, perhaps, the agreement on blue jacinth, although to the ancients jacinth was purple (Pliny xxxvii. 125; Isidore xvi.ix.3; Solinus, ed. Mommsen, pp. 135–6). Solinus and Isidore mix the colours somewhat, but Isidore still classifies jacinth among purple gems. (It is the minor element of this mixture of colours in whose allegorizing Ambrosius Autpertus (CCSL, CM 27A, 819) takes advantage of the fact that purpura can refer both to blue-purple and to the colour of blood (see below, n. 201). Ambrosius’ description of the gem does not offer any parallel to Bede's rubicundum: Meier's translation ‘Purpurrot’ (Gemma Spiritalis 1, 150–1) is unfortunate.)

¹⁵⁰ ‘For our Lord Jesus Christ arose from the dead. He fulfilled and incarnated this, and this is the vision of the stone of marvel. It is luminous and gleaming like a coal of the colour of blood, and it glows in its brilliancy like lighted lamps (or torches), which brilliancy none can take away or hide from the lamps. Of a verity in this wise are illuminated and made glorious all the saints who have become worthy of receiving the Holy Spirit from Christ, and of all those illuminated ones who believe in God through truth are all the souls now made glorious by him, and in their meekness is clearly cast upon them the guise of an angel-like coming and by their thoughts also, just as the countenance of Moses the meek became divinely fair, just as from the eye of the sun the dazzling rays gleam forth, wherefore because of the glory of the illumination none could gaze on the countenance of Moses’ (Blake, , Epiphanius, p. 131Google Scholar; my italics).

¹⁵¹ ‘Qui … fulgorem. Quia coelestibus semper intentus sanctorum animus, divini luminis quotidie radiis innovatus, compunctior quodammodo atque ardentior aeterna perquirit, aliisque inquirenda suadet’ (‘That is, the soul of the saints, ever intent on heavenly things, renewed daily with the rays of the divine light, in a manner the more urgent and ardent pursues eternal things and persuades others to ask after them’) (PL 93, col. 198A).

¹⁵² See above, p. 80.

¹⁵³ It is possible that Bede used Epiphanius' figural section for one further description: if his variety of chrysolite ‘caerulei viridisque colons’ (PL 93, col. 200C) comes in detail from any previous writer, it is probably Epiphanius, as represented in the garbled Georgian (Blake, , Epiphanius, p. 153)Google Scholar. But this ‘blue and green’ is more likely to be Bede's extrapolation from the ‘marine’ colour of Jerome and Isidore (see above, p. 89).

¹⁵⁴ See above, p. 81.

¹⁵⁵ Gemma Spiritalis 1, 180 (n. 207), 207–8, 287–8, 353 and 359.

¹⁵⁶ Sancti Ambrosii Opera, Pars Quinta: Expositio Psalmi CXVIII, ed. Petschenig, M., CSEL 62 (Vienna, 1913), 372–5.Google Scholar The psalm in question is cxix in the English bible.

¹⁵⁷ I.e. very probably as identified by Wellmann and Wirbelauer (cit. ‘Lapidary Traditions, part I’, p. 11, n. 6), Xenocrates of Aphrodisias, a physician of the mid- to late first century A.D. who wrote extensively on more or less magical medical matters, not as held by more recent German scholars (e.g., Ullmann, M., ‘Xenokrates (7)’, PW, Supplement band 14 (Munich, 1974), 974–7)Google Scholar, the rather earlier Xenocrates of Ephesus. Space does not permit full rehearsal of the arguments here.

¹⁵⁸ For this word, which is variously garbled in all the manuscripts and editions, I adopt the undoubtedly correct emendation of Jaeger, W., ‘Studia Pliniana et Ambrosiana ad Xenocratem Ephesium [sic] emendandum’, Mélanges de philologie, de littérature et d'histoire anciennes offerts à J. Marouzeau (Paris, 1948), pp. 297–302Google Scholar. Ambrose took his account of topaz, including the acknowledgement to Xenocrates, from a now lost work of Origen. Large fragments of the Greek corresponding to this passage survive (ed. Pitra, J. B., Analecta sacra Spicilegio Solesmensi parata 11 (Paris, 1884), 341–2Google Scholar; quoted Wellmann, ‘Steinbücher’, pp. 89–90; Jaeger offers some textual improvements). The surviving Greek lacks so many details present in Ambrose that it must be regarded (as Jaeger emphasizes) as a scholium rather than as a direct survival of the Origen; but its frequent verbal correspondences to Ambrose, like the Greek words embedded in Ambrose's text, undoubtedly reflect the original. Its word πυρίχρουν (neuter, agreeing with χρ⋯μα) makes sense for the first time of the Ambrose context. The manuscript readings might be interpreted as ‘equally golden’ or ‘all-golden’ (or ‘every-coloured’ as Isidore (Etym. xvi.vii.9) seems to have read and understood it); we may note that whatever Bede had before him at this point it cannot have given rise to his colour rosea.

¹⁵⁹ The adjective of ‘resplendenti similis’ needs a referent. The presence of lampas in Bede's corresponding phrase seems cogent reason to believe that Ambrose wrote ‘resplendenti lampadi similis’ and that the oddly declined noun dropped out of the text at some such early stage in its transmission as the odd words pyrichrus and glyptali were corrupted. (The earliest manuscript of the Ambrose passage is ninth-century.) For the sense cf. Berengaud's ‘Topazius aerei aureique coloris esse dicitur, iubar suum circumquaque spargens’ (‘Topaz is said to be of the colour of air and of gold, scattering its radiance around it on all sides’) (PL 17, col 1043A). Hrabanus Maurus, De Universo, in a sentence (PL III, col. 468D) derived from this of Bede's (cf. above, n. 119), reads lampat for Bede's lampas. The three examples of the unclassical verb lampare in Blatt's, FranzNovum Glossarium Mediae Latinitatis fasc. L (Copenhagen, 1957)Google Scholar are all continental and none earlier than the ninth century, so Hrabanus' word is unlikely to constitute evidence relevant to the text of Bede.

¹⁶⁰ Glyptali is my emendation of the manuscripts' genitali, which has passed without comment by modern editors but does not make Latin sense in context. Genitalis pertains to the coming into existence of things, not to what modern English calls ‘creative’ artistry on things that already exist. Thesaurus Linguae Latinae vi.ii (Leipzig, 1934), 1813Google Scholar, includes this phrase with genitalis under the heading ‘de actionibus quae genitarum rebus dant’, but all the other examples there are unexceptional. Ambrose's context demands a sense ‘pertaining to sculpting or engraving’; his quodam prepares the ground for some unusual word. A coinage glyptalis would be of a piece with his introduction of other Greek words into the passage. No other word glypt- occurs in Latin, according to the Thesaurus: unlike pyrichrus, it would not even present known elements to the scribes, so its early corruption into a very similar Latin word is easily understood. Comparison with the Origen scholium rather suggests that the whole phrase is Ambrose's gloss on ‘it is εὔγλυøος’’, so his introducing the Greek word by a Latinized cognate is also understandable.

¹⁶¹ Pl 93, col. 200C–D: ‘The stone topaz, inasmuch as it is not widely obtainable, is accordingly expensively priced and valuable. It is said to have two colours, one of the purest gold, and the other shining with an ethereal translucence. A rosy fatness, and a reticent purity, next neighbour to the stone chrysoprase in size and colour; because most greatly radiant when it is struck by the sun's brightness, surpassing the most highly esteemed translucencies of all jewels, uniquely provoking the most covetous pleasure of the eyes in its appearance. If you choose to polish it you dull it, but if you leave it in its natural state it will shine out. It is said to be marvellous to kings themselves, so that they are not aware of possessing anything comparable to it among their riches.’

¹⁶² ‘And she, though she had an abundance of regal ornaments, was yet amazed beyond measure at the colour of it, and was at pains to ensure that the appearance of so precious a stone should remain unknown no longer; and so, sought out as a result of her attention, the stone came into wider circulation.

‘We have said how the stone topaz came to be known; now let us speak more explicitly of its nature. There is in this stone a crasis, that is a certain temperate blending, of two colours, leek-green and gold-hued, resembling chrysoprase, expressed as it were in certain shapes of either colour, which skilled jewellers are said to be able to bring out more fully. There is a variety, moreover, which is flame-coloured and very clear and gold-coloured and oily, similar to a resplendent <lamp>, especially when it is struck by direct sunlight. It is indeed most beautiful and marvellous, beyond all chrysoprases in size and, as I said, more oily in appearance. It has this property, that if you try to smooth and polish it it is made rougher; and it is worn away by wear. It has moreover, for certain ‘glyptic’ work, the property of being euglyphus, that is easily engraved; and it is marvellous and, that it might be held worthy of the highest attention, it is found with difficulty and seldom discovered for human use, just like that which the rich queen marvelled at.’

The ancients included in ‘topaz’ both modern topaz (= Ambrose's yellow variety) and peridot (predominantly green). It is the latter to which belong the tactile properties and which is found on the island (see Eichholz's notes to Pliny xxxvii. 107–9 (Loeb ed., x, 250–2). Pliny calls it the largest of all gemstones. The normal colour of topaz was yellow to Greek writers generally, to Latin writers green (cf. ‘Lapidary Traditions, part I’, p. 42). Ambrose confused the properties much less than most subsequent ecclesiastical writers.

¹⁶³ ‘Lapidary Traditions, part I’, p. 42.

¹⁶⁴ See ibid. pp. 41–4; on the date ibid. pp. 7, 28 and 54.

¹⁶⁵ Nat. Hist. XXXVII.108–9. Isidore, Etym. xvi.vii.9, does not unambiguously name the island.

¹⁶⁶ See above, nn. 158–60.

¹⁶⁷ See below, n. 209.

¹⁶⁸ See e.g. Evans, Joan, Magical Jewels of the Middle Ages and the Renaissance, particularly in England (Oxford, 1922), pp. 29 and 95–109Google Scholar. Church disapproval was the main reason why later medieval magical lapidaries prescribe not directly the engraving of efficacious signs on gems; but ‘If you should find …’ a jewel so engraved.

¹⁶⁹ See ‘Lapidary Traditions, part I’, p. 51; Henig, Corpus (see above, n. 1), pp. 159–62.

¹⁷⁰ See above, end of n. 162.

¹⁷¹ The word mercium here was probably prompted by Augustine's mercedem (Enarrationes in Psalmos, PL 137, col. 1569), as suggested ‘Lapidary Traditions, part I’, p. 42. The mercenary dichotomy has a curious subsequent history. Marbod derived from Bede's description his first five lines on topaz (PL 171, col. 1748A): ‘Nominis eiusdem topazion insula gignit, / Qui quanto rarius, tanto magis est pretiosus. / Hic species tantum binas perhibetur habere: / Alterius puro color est vicinior auro, / Clarior alterius tenuisque magis reperitur.’ His poem's secular tone implies deliberate contrast to Explanatio Apocalypsis and with Cives celestis patrie, which he likewise knew and only drew on once (see ‘Lapidary Traditions, part I’, p. 59, n. 2). Bede's jewel passage and close derivatives were copied as balancing pendants to Marbod's poem in some manuscripts. If Bibiothèque Nationale, lat. 14470 (see ‘Lapidary Traditions, part I’, n. 11), whose following item (ptd PL 171, cols. 1771–4 as ‘Mystica seu Moralis Applicatio … nusquam edita’) is a slightly reduced version of Bede's passage, was rightly used as basis for the Migne edition, that may even have been Marbod's intention. (Riddle, Marbode, is of course wrong to attribute this and other following items – cf. below, n. 230 – to Marbod in any serious literary sense.) Doubtless in any case he felt a wry satisfaction in using only that portion of Bede's passage which was least devout in tone.

¹⁷² See below, pp. 104–5.

¹⁷³ PL 93, cols. 202D–3A: ‘Whose designer and founder God, who is the foundation of foundations, and deigned even to become a priest for us, that by the sacrifice of his own blood he might at the same time wash clean and dedicate the walls of that city, possesses as his own all things whatsoever that are the Father's. Wherefore the same stones were commanded to be set upon the breast of the high priest and inscribed with the names of the patriarchs, that by a most beautiful mystery it might be signified how all the spiritual unctions which every one of the saints individually and partially received were in the mediator of God and man, the man Jesus Christ, gathered perfectly together.’

¹⁷⁴ PL 94, cols. 539–60. DDL is at 551D–2A.

¹⁷⁵ McNamara, M. (The Apocrypha in the Irish Church (Dublin, 1975), pp. 22, 54–6, 76–8, 86 and 128–35)Google Scholar discusses some of them.

¹⁷⁶ E.g. that beginning ‘Duodecim abusiva in hoc saeculo sunt’ (col. 545c–D), which draws on a homily De duodecim abusivis saeculi generally accepted to have been written in Ireland c. 630 × 650 (ed. S. Hellmann, Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, ed. A. Harnack and C. Schmidt, 3rd. ser. 4 (Leipzig, 1909), 1–62; discussed Laistner, M. L. W., Thought and Letters in Western Europe A.D. 500–900 (London, 1957), pp. 144–6.Google Scholar

¹⁷⁷ E.g., the names and descriptions of the three Magi (col. 541D), quoted McNamara, Apocrypha, p. 56. If my dating of the Collectanea is correct, this is the earliest known example of what was to become the traditional picture of the Magi. It may actually be a Hiberno-Latin composition; it is unlikely to have been translated from Greek as some writers have suggested. I discuss it further below, n. 184.

¹⁷⁸ On whom see Herren, M., ‘Some New Light on the Life of Virgilius Maro Grammaticus’, Proc. of the R. Irish Acad. 79c (1979), 27–71Google Scholar, esp. 33, 35, 46, 61–9; and cf. the clear-headed remarks of Macalister, R. A. S., The Secret Languages of Ireland (Cambridge, 1937), pp. 83–8.Google Scholar On Virgilius Maro Grammaticus’ circle see Herren, M., ‘The Pseudonymous Tradition in Hiberno-Latin: an Introduction’, Latin Script and Letters A.D. 400–900: Festschrift Presented to Ludwig Bieler, ed. O'Meara, J. J. and Naumann, B. (Leiden, 1976), pp. 121–31, at 126–30Google Scholar. There is a new edition: Virgilio Maro Grammatico: Epitomi ed Epistole, ed. Polara, G. (Naples, 1979).Google Scholar

¹⁷⁹ E.g. Dekkers, E., Clavis Patrum Latinorum, 2nd ed. (Steenbrugge, 1961), no. 1129Google Scholar; McNally, R. E., ‘Isidorian Pseudepigrapha in the Early Middle Ages’, Isidoriana, ed. Diaz, M. C. Diaz y (Leon, 1961), pp. 305–16, at 313.Google Scholar

¹⁸⁰ PL 94, col. 543c on ‘Cloud’; col. 548A–B on ‘Wind’, ‘Nature’ and ‘Diamond’ (on which last cf. ‘Lapidary Traditions, part I’, pp. 24–5); riddles 3, 2, 4 and 9 as ed. Ehwald (see below, n. 181).

¹⁸¹ The Epistola ad Acircium, as part of which Aldhelm published the riddles, dates from the few years immediately following 685: see Lapidge, M. and Herren, M., Aldhelm: the Prose Works (Ipswich and Totowa, N.J., 1979), p. 12.Google ScholarEhwald, R. (Aldhelmi Opera, MGH, Auct. antiq. 15 (Berlin, 1919), 41)Google Scholar thought it could be deduced from the text of the Epistola that the riddles did not previously circulate separately. It seems better to suppose with Lapidge (Aldhelm, p. 13) that they may have done. How much earlier than the date of the Epistola these four particular riddles were composed we cannot know. The general probability remains, however, that they would have reached the Collectanea's compiler(s) as part of the Epistola, hence after 685.

¹⁸² See below, pp. 104–5.

¹⁸³ See above, pp. 91–4.

¹⁸⁴ The arguments are too lengthy for inclusion in this article and do not lend themselves to intelligible summarizing. I hope to publish an article on the Collectanea shortly.

¹⁸⁵ ‘Lapidary Traditions, part I’, p. 23 accordingly needs correction.

¹⁸⁶ ‘On (the?) Twelve Stones. Chrysolith, which stone light and day hide(s), so that at night it is fiery but during the day pallid. Sapphire, a similar gem, but it is not hexagonal; it is found in the Red Sea. Struck by the sun's rays it emits a burning brilliance. Sard, a stone purple in colour, which serpents fear for its brightness, so they say. Asbestos is tinged with a bloody colour from the sun itself. When it is kindled they say it cannot be extinguished. Jacinth, of which they say it can sense the winds and calm(s) of weathers. Carbuncle, of red colour, which the eyes love, breathes its splendour from afar and is not seen close to. Cerdamios, which is brilliant orange in the open air, but indoors is of an azure colour. Chrysoprase, of gold and purple as if carrying with it a mingled light; which eagles love. Beryl is covered by a golden cloud, and it is hexagonal. It is said to burn in the hand anyone who holds it. Emerald, which, they say, of a purple colour, resembles jacinth in power, and from it they think the motions of the moon can be stirred up. Sardonyx, a base stone, is of a black colour; it is formed in the deep waters. Agate, a stone shining with drops of gold, efficacious against scorpions, and which taken in the mouth assuages thirst.’

Notes

a gemmis: the sense required is that of nom. sg. gemma, whether the apparent dat. pl. gemmis results from authorial mis-declension or scribal homoeoteleuton.

b fulgorem: see above, n. 143. Here a colour contrast is obviously intended.

c colore purpureo: the following comma might be omitted, giving ‘Emerald, which they say resembles purple-coloured jacinth in power …’ Either punctuation begs questions discussed below.

¹⁸⁷ Ed. Mommsen, p. 136: ‘Where jacinth, there also chrysoprase occurs; which stone light hides and darkness manifests. For there is in it this diversity, that by night it is fiery, by day pallid.’

¹⁸⁸ Nat. Hist. xxxvii. 26.

¹⁸⁹ ibid, xxxvii. 76.

¹⁹⁰ Ed. Mommsen, p. 152: ‘The Arab produces also malachite, of a thicker green than emerald, mightily efficacious against pre-natal dangers to children.’ (Solinus' ingenita misinterprets and misapplies innato of Pliny, Nat. Hist. xxxvii.114.) ‘In the Red Sea he produces also ‘iris’, hexagonal like crystal, which, struck by the sun's rays, casts from itself a likeness of the heavenly bow in a reddish reverberation of the air.’

¹⁹¹ Nat. Hist, xxxvii.136–7.

¹⁹² Meier does not regard it as important enough to mention it in her section on apotropaic effects (Gemma Spiritalis, 1, 414–26); her examples of efficacy against beasts and venom (pp. 423–4) and against demons and subjoined animals (pp. 417–18) do not specify serpents; I know no Latin commentary except the Epiphanius which has the motif. On secular lapidaries see below, nn. 193–6.

¹⁹³ Pliny, Nat. Hist. xxxvi.142: Isidore, Etym. xvi.iv.3 (cf. ‘Lapidary Traditions, part I,’ p. 17, n. 7, and p. 22, n. 4); Damigeron, Pitra §20, Alphabetic §27. The early medical texts do not mention snakes as such in this context but the ‘Lombard’ Dioscorides (Romanische Forschungen 13 (1902), 238)Google Scholar and the Old English Leechdoms (ed. Cockayne, 11, 296; cf. ‘Lapidary Traditions, part I’, p. 26, n. 4) agree that jet is good against their venom.

¹⁹⁴ Nat. Hist. xxxvii. 139–40 and 142: effective against spider-bites and scorpions and renders scorpions' poison inert; Solinus, ed. Mommsen, pp. 53–4: good against scorpion-stings.

¹⁹⁵ Isidore, Etym. xvi.xi.1, partly confuses their identity, giving achates the black colour which in Pliny belongs only to gagates. More thoroughgoing confusion may have been one reason why there is not a separate entry for achates either in the ‘Lombard’ Dioscorides or in Ad Paternianum (where the medical powers of gagates are more moderate).

¹⁹⁶ De Mély, Lapidaires 11, 197: ‘Among the varieties of these stones there is also an agate having the colour of a lion. Ground (and made into a pasted with water and smeared upon animal-bites it drives out the poison of scorpion and vipers and the like.’ The Latin (CSEL 35, 754 / PGL 43, col. 340) is almost identical with the Greek epitome. Cf. the very similar information given by Damigeron (Pitra §17, Alphabetic §4).

¹⁹⁷ CSEL 35, 746 / PGL 43, col. 324: ‘And another variety, sardachates, around which stone is to be seen a white band as on an agate. Those who concoct fabulous stories assert that it can put evils to flight and call it the Peacemaker.’ The sentence is missing from the extant Greek epitome.

¹⁹⁸ E.g. Pliny's numerous varieties (Nat. Hist. xxxvii.139) with names compounded in -achates do not include a sardachates, nor does that name appear in Solinus or Isidore.

¹⁹⁹ See ‘Lapidary Traditions, part I’, p. 41.

²⁰⁰ Nor in any quoted by Meier, Gemma Spiritalis 1, 121 and 147–52.

²⁰¹ Of course πνρωπός is not synonymous with purpureus, but taken, as the context indicates, as a colour-word, its semantic field certainly overlaps (see refs. above, n. 149). The choice of purpureus would also be influenced by Epiphanius’ initial statement that sard is the colour of blood. Blood is often called purpureus in both classical and patristic Latin.

²⁰² These words could easily derive from the Greek but could not derive from the Latin Epiphanius (cf. quotations ‘Lapidary Traditions, part I’, p. 41). DDL's only other entry which begins so, ‘Achates lapis …’, preserves successive words of its source text.

²⁰³ Cf. ‘Lapidary Traditions, part I’, pp. 18 and 24.

²⁰⁴ Ed. Mommsen, p. 57. Isidore, Etym. xvi.iv.4 (or xiv.iv.15) would also do, but as the colour is taken from Solinus the inextinguishableness probably is also.

²⁰⁵ Agreeing with Pliny, Nat. Hist. xxxvii.146, and Isidore.

²⁰⁶ Ed. Mommsen, p. 136: ‘From the same soil [i.e. country (as jacinth and chrysoprase)] we obtain haematite, with a bloody redness and on account of that called haematites [< αἶμα ‘blood’].’

²⁰⁷ Ibid.: ‘This it is that senses the breezes and does as the sky does; nor does it gleam equally when it is cloudy as when it is a clear day.’

²⁰⁸ Cf., mutatis mutandis, ‘Lapidary Traditions, part I’, pp. 30–1 and 22. For ideas current in Ireland specifically see my article on ‘The Ever-New Tongue’ (see above, n. 3).

²⁰⁹ Pliny, Nat. Hist. xxxvii.62–3, at length, esp. ‘non alia gratior oculorum refectio est’; Solinus, ed. Mommsen, p. 87: ‘Nihil his iucundius, nihil utilius vident oculi’; Isidore, Etym. xvi.vii.1, after Pliny. All three contrast emerald with all other gems.

²¹⁰ Ed. Mommsen, p. 98: ‘Moreover, there are various kinds of thunder-stone. The German is white, yet has an azure sheen, and if you have it in the open air it catches the glitter of the stars.’

²¹¹ ibid. p. 195: ‘The chrysoprases also, which carry with them a mixed light of gold and leek-green, they have likewise adjudged to belong to the beryl group.’ Corruption of porraceo to purpuraceo, and further to purpura ceu, is found in extant manuscripts of Solinus.

²¹² Pliny, Nat. Hist. x.12 (aetites), xxxvi. 149–51 (aetita); Solinus, ed. Mommsen, p. 159 (aetites); Isidore, Etym. xvi.xiv.2 (aetites) and xvi.xv. 19 (aetitis); cf. Laufer, ‘The Diamond’, p. 9.

²¹³ The recension I call Pitra's: see ‘Lapidary Traditions, part I’, pp. 13–15.

²¹⁴ Ed. Mommsen, p. 195: ‘The Indians work down beryls into hexagonal shapes … The beryl group divides into multifarious varieties … there are chrysoberyls, which, gleaming less strongly, are enveloped by a golden cloud.’

²¹⁵ De Civitate Dei xxi.5, quoted ‘Lapidary Traditions, part I’, p. 47. The other possible source, Isidore, Etym. xvi.iv. 5, has indicative adurit, not infinitive adurere.

²¹⁶ See above, n. 149.

²¹⁷ Ed. Mommsen, p. 160: ‘Magi … ex ea lunares motus excitari putant’. Mommsen's text of Solinus reads ex ea, Isidore, Etym. xvi.xv. 17, ex eo; but, since Isidore took the phrase from Solinus, at least some early Solinus manuscripts must have read eo. DDL's author is more likely to have used Solinus for this as for the other gems.

²¹⁸ Pitra's recension §6; Alphabetic §46 ‘valentissimus …’: ‘Most powerful for all divination by water.’

²¹⁹ Cf. unambiguous similar statements: §§Pitra 2, 5, 14, 18, 29; Alphabetic §§15, 40 and 48; and Pliny, Nat. Hist. xxxvii.192.

²²⁰ Etym. xvi.viii.2–4: ‘Sard is so called because it was found first at Sardis; it has a red colour, outstanding among marbles, but basest of gems … Sardonyx is called by a combination of the two names, for it has the lustre of onyx and sard. It is composed, furthermore, of three colours, black underneath, white in the middle, and red on top. It is found in the Indies and Arabia, discovered in torrents.’

²²¹ Ed. Mommsen, pp. 53–4 (the beginning quoted ‘Lapidary Traditions, part I’, p. 50): ‘The stone agate … Crete yields what they call curralliacbates, resembling coral but dotted with drops of shining gold, and efficacious against the stings of scorpions. India yields a variety exhibiting the appearance now of woods and now of animals, to have looked on which benefits the eyes and which taken in the mouth assuages thirst.’

²²² Cf. the speculation above, p. 99.

²²³ In Tenga Bithnua (see above, n. 3) implies that material from Damigeron reached Ireland in the seventh century, but this is not necessarily so.

²²⁴ Cf. the use by Ambrosius Autpertus of Gregory's phrase on sardonyx (quoted above, p. 80) in his comments on sard in Revelation (CCSL, CM 27, 208, and twice 27A, 821).

²²⁵ See ‘Lapidary Traditions, part I’, p. 38.

²²⁶ ibid. pp. 41–5.

²²⁷ ibid. pp. 28–9.

²²⁸ ibid. p. 29.

²²⁹ Chevalier, U., Repertorium Hymnologicum 1 (Louvain, 1904), nos. 3271 and 3275Google Scholar; Walther, H., Initia Carminum (Göttingen, 1959), no. 2812.Google Scholar

²³⁰ Whether Marbod (as a pendant to his poem), St Anselm (as below, n. 234), or Amato of Monte Cassino. The first two attributions are at least as early as manuscripts V and J respectively (Riddle, Marbode, maintains the misattribution to Marbod, as he does for the material of Bedan and Arabic provenance (PL 171, cols. 1771–6)); the third was proposed recently by Lentini, A., ‘Il ritmo “Cives caelestis patriae” e il “De duodecim lapidibus” di Amato’, Benedictina 12 (1958), 15–26Google Scholar (unfortunately followed by Meier, Gemma Spiritalis 1, 142). Other reputed authors include Bede (manuscripts R and S), Herrad of Landsperg (by editorial inference from manuscript T), and Frithegod (see below, p. 122).

²³¹ This copy was discovered by Dr Michael Lapidge. The manuscript's other contents are discussed Eden, P. T., Theobaldi Physiologus, Mittellateinische Studien und Texte 6 (Leiden, 1972), 9Google Scholar. Dr Lapidge emphasizes that Eden's dating ‘s. xi/xii’ is palaeographically too late.

²³² Text published Dronke, ‘Tradition and Innovation’, pp. 78–9; earlier, inaccurately, Giles, J. A., Anecdota Bedae, Lanfranci, et Aliorum: Inedited Tracts, Letters, Poems, etc., Publ. of the Caxton Soc. 7 (London, 1851), 66–7Google Scholar. The manuscript as a whole is discussed Rigg, A. G. and Wieland, G. R., ‘A Canterbury Classbook of the Mid-Eleventh Century (the ‘Cambridge Songs’ Manuscript)’, ASE 4 (1975), 113–30.Google Scholar

²³³ Analecta Hymnica, ed. Dreves, G. M. 11 (Leipzig, 1887), 94–5Google Scholar, with music p. 117. The manuscript is no. 1672 in Gamber, K., Codices Liturgici Latini Antiquiores, Spicilegii Freiburgensis Subsidia 1 (Fribourg, 1963)Google Scholar. Dreves states (p. 8) that the main text is tenth-century, but (p. 10) that Cives celestis patrie is one of a series of additions belonging mostly to the eleventh century. Salmon, P. (Les Manuscrits liturgiques latins de la Bibliothèque Vaticane 1, Studi e testi 251 (Rome, 1968), 52)Google Scholar gives a single date for the manuscript s. x^cx. The music is ed. Stäblein, B., Monumenta Monodica Medii Ævi 1: Hymnen (Kassel, 1956) (hereafter MMMÆ), pp. 51–67 and 522–8Google Scholar. Stäblein holds (pp. 522–3) that the manuscript's literary and liturgical content argues its compilation between, at the outside, c. 990 and 1030. His arguments tend (as does the manuscript tradition of Cives celestis patrie) to favour something nearer the latter date.

²³⁴ As part of Anselm's Enarrationes in Apocalypsin (with which the manuscript begins), and so ed. PL 162, cols. 1580–2; manuscript discussed Hauréau, B., Notices et extraits de quelques manuscrits latins de la Bibliothèque Nationale 1 (Paris, 1890), 74–88Google Scholar; and Bibliothèque Nationale: Catalogue général des manuscrits latins, ed. Lauer, P. H. 1 (Paris, 1939), 248Google Scholar. J lacks the last three stanzas.

²³⁵ Text as reported Riddle, Marbode, pp. 119–22, an edition of the poem (with poor translation) based on V, collated with Q and K. (I assume that Riddle's ‘14, 4 multiplicatas’, given without comment for all three manuscripts, is a misprint like his ‘11, 3 intetinctus’.) The manuscript is described Jørgensen, E., Catalogus Codicum Latinorum Medii Aevi Bibliothecae Regiae Hafniensis (Copenhagen, 1926), p. 48.Google Scholar

²³⁶ Ed. Lentini, ‘Il ritmo “Cives caelestis patriae”’, pp. 23–6. The manuscript's main text is s. xi^cx, but the hymn is added in a hand which Franceschini, E. (‘Un nuovo codice del “De duodecim lapidibus preciosis”’, Aevum 26 (1952), 183)Google Scholar dates to the fourteenth century. Lentini thinks it could be thirteenth.

²³⁷ Lateinische Hymnen des Mittelalters, ed. Mone, F. J., 111 (Freiburg im Breisgau, 1855), 28–30.Google Scholar

²³⁸ Text not previously edited; manuscript discussed Warner, G. F. and Gilson, J. P., Catalogue of Western Manuscripts in the Old Royal and King's Collections in the British Museum 1 (London, 1921), 284–6Google Scholar. N lacks the last two stanzas.

²³⁹ Facsimile ed. Derolez, A., Lamberti S. Audomari Canonici Liber Floridus (Ghent, 1968)Google Scholar; discussed Schmidt, P. G., ‘Die mittellateinische Dichtung im Liber Floridus’, Liber Floridus Colloquium, ed. Derolez, A. (Ghent, 1973), pp. 51–8, at 53–4)Google Scholar. Schmidt points out that the poem is preceded by a picture of the heavenly city (65r) in which the names of twelve jewels are matched one by one to those of twelve apostles. The link between poem and picture is not close, however, as the twelve stones of the picture are those of the Exodus list. (The names of apostles make a composite set which does not fit any point in the narrative of Gospels or Acts.)

²⁴⁰ Aurelii Prudentii Clementis Carmina, ed. Dressel, A. (Leipzig, 1860), p. 209Google Scholar, quoting the old press-mark D. 5. 4, which according to H. Narducci (Catalogus Codicum Manuscriptorum praeter Graecos et Orientales in Biblioteca Angelica olim Coenobii Sancti Augustini de Urbe 1 (Rome, 1893), 205a), belonged to a 1639 manuscript of Aristotle's Physics. Dressel's press-mark is without doubt a misprint for D. 5. 3 (the modern MS 435), described by Narducci, ibid., as ‘sec., ut videtur, xi; principio et fine carens, notis musicis refertus. Liber coralis … Fuit alicuius veteris Monasterii, in quo utebantur peculiari liturgia, non eadem ac Romana.’ The close agreement of the readings with D, but their greater corruptness, makes Narducci's dating much more credible than Dressel's (he assigned the manuscript to the tenth century) and suggests that the ‘peculiar liturgy’ was something Gallican. Dressel does not report any reading for the first stanza, but in his context (commentary on Prudentius, Psychomachia, lines 851–65) it is quite unclear whether that should be taken as meaning that the manuscript does not contain it.

²⁴¹ Text as reported Riddle, Marbode, pp. 119–22 (see above, n. 235); manuscript described Wilmart, A., Codices Reginenses Latini 1 (Rome, 1937), 106–7.Google Scholar

²⁴² Text ed. (a little inaccurately) Garrett, , Precious Stones, pp. 27–31Google Scholar; manuscript discussed Warner and Gilson, Catalogue 1, 276.

²⁴³ Manuscript described Coxe, H. O., Catalogus Codicum MSS qui in Collegiis Aulisque Oxoniensibus Hodie Adservantur 1.iii: Collegii Mertonensis (Oxford, 1852), pp. 39–40Google Scholar. S is a direct copy of R (or else a very close derivative) with some attempt at correcting gross errors.

²⁴⁴ Text ed. Engelhardt, C. M., Herrad von Landsperg (Stuttgart and Tübingen, 1818), pp. 148–52Google Scholar. Engelhardt's earlier remarks on the history of the manuscript are more accessibly summarized E., and Rott, J.-G., Hortus Deliciarum: le ‘Jardin des Délices’ de Herrade de Landsberg (Strasbourg, 1945), pp. 4–5Google Scholar. On its place in the burnt Strasbourg library see the valedictory notice by Reuss, R., Revue critique d'histoire et de littérature (Sept. 1871), pp. 160–80, at 166 and 168.Google Scholar

²⁴⁵ Text ed. PL 171, cols. 1771–2, and Riddle, Marbode (as above, n. 235); where their reports differ I have followed Migne. On the manuscript see ‘Lapidary Traditions, part I’, p. 12, n. 11.

²⁴⁶ Manuscript described Catalogue of the Manuscripts Preserved in the Library of the University of Cambridge, ed. Hardwick, C. 111 (Cambridge, 1858), 670–3Google Scholar, in whose foliation the poem is at 112r–v. The same catalogue (1 (1856), 76–7) adds that stanza 5 of Cives celestis patrie is inserted ‘in a later hand’ in the manuscript Dd. 3. 16, for which no date is given. The insertion is at 12v, in a version of the lapidary treatise of Albertus Magnus (thirteenth-century) which runs from 7v to 15v.

²⁴⁷ Notices et extraits 1, 76–7. Hauréau does not record which particular manuscripts support any of his readings. It would seem from his ‘13, 4 scintillasque’ and ‘15, 2 fundamenta’ that he had one of the same class as T. He lists as also containing the poem Troyes 2269, St Omer 115, Rouen A 528, Berne 48, and a Vienna manuscript described Denis, M., Codices Theologici Vindobonenses 1 (Vienna, 1793)Google Scholar, col. 1013 (none of which he would date earlier than the thirteenth century), but seems to imply that he did not use them. Chevalier, Repertorium, identifies what presumably is Hauréau's last citation as Vienna, pal. 4760, 36V. Walther, H., Initia Carminum: Ergänzungen und Berichtigungen (Göttingen, 1969)Google Scholar, gives Hauréau's Rouen A. 528 as being now numbered A. 362, and adds Rouen U. 43, fol. 122 (manuscript described Bibliothèque de l' École des Chartes 47 (1886), 629–32)Google Scholar. Walther, Initia Carminum, lists the further manuscripts Berlin, oct. lat. 199 (s. xii), 5v; Berlin, lat. fol. 307 (s. xii), 24v; Phillipps 1728 (s. xii), fol. 78; Brussels (old) 2834; Cambrai 830 (s. xii), 130V; Erfurt, Ampl. qu. 295 (s. xiii), 37r-v (see Thorndike, L., ‘De Lapidibus’, Ambix 8 (1960), 6–23, at 10)CrossRefGoogle Scholar; Göttweich 107 (s. xii), 25v; Zwettl 41 (s. xii/xiii), fol. 74. Riddle (Marbode, pp. 136 and 138) adds Berlin 956 (s. xii²), 24V–, Brussels 1684 (s. xv), 42V–3 (these presumably identical with two of the above); Vatican, lat. 3267, 59r–v; Amiens 65 (s. xii); and dates Rouen, A. 528, fol. 187–, to the eleventh century (which if right would preclude Marbod's authorship, as MSS B, C and D definitely do).

²⁴⁸ I count as divergent every word whose form varies from the reconstructed original by more than do medieval from classical spelling conventions, or which is misplaced in order. So counted, C gets nearly 96% of words right, (H) and Q about 93%, K 92% (if Riddle reports all variants), T about 91%, D 90%, and M 89%; a cluster of manuscripts around 86% (joined by J for the part it contains) ends with B under 85%, followed by X 79%, L 73%, and the incomplete J and N around 67%.

²⁴⁹ Here inter alia belongs T's transposition of extat and constat between 1, 6 and 6, 1.

²⁵⁰ Especially V's transpositions of adjacent words, implausibly conspicuous to derive solely from random error, even of a common type. The most striking instance is 15, 3 felix Deo et proxima in VQK, RS and B, whose transposition of the words et Deo, which does violence to both sense and rhythm if the text be read not sung, can be accounted for by a preference for o rather than e as a high melismatic vowel; plausibly, both because o is the more sonorous and because it is markedly the easier to enunciate near the top of one's vocal register. The latter would be a practical consideration here for monks not endowed with tenor voices.

²⁵¹ This is the likeliest reason, e.g., for M1, 33 opifex, a reading otherwise found only in class I manuscripts.

²⁵² A good translation (from V) into the language of nineteenth-century hymnody, in the metre of the original, is to be found King, C. W., The Natural History of Precious Stones and of the Precious Metals (London, 1867), pp. 337–9.Google Scholar

²⁵³ MMMÆ 1, 65–6. Stäblein gives two transcriptions as being of equal validity (the other a tone higher; either equally with or without flattening of B). I have followed the version I find most coherent. (It is also the closest to Dreves's version.) Most other (edited) surviving manuscripts are literary not liturgical, so had no cause to include the music. Mone mentions that M has neumes, Dreves that so does Göttweich 107, but neither seems ever to have been transcribed. C has neumes for the first line (Rigg and Wieland, ‘Canterbury Classbook’, p. 123). The practice of musical cross-reference by means of short cues was widespread in liturgical manuscripts (see Planchart, A. E., The Repertory of Tropes at Winchester, 2 vols. (Princeton, N.J., 1977)Google Scholar, passim, on the few surviving Anglo-Saxon examples). Planchart shows (pp. 47–50) that the largest of these, London, British Library, Cotton Caligula A. xiv, is probably, like our C, a Canterbury manuscript.

²⁵⁴ E.g. 1, 3 artifex: most words in 4, 1–2 and 10, 1–2; 13, 2 decore violaceus (see above, pp. 79, 90, 90, 95 and 91 respectively). At these and other points Bede confirms the conclusions of regular internal textual criticism.

²⁵⁵ Especially 12, 2 virore medioximus. Medioximus is not normally used with a dependent noun; viror, unlike its cognate verb virere, is normally confined to meaning ‘greenness’ as in 2, 2 (which would make no sense in a description of jacinth), not burgeoning ‘strength’ generally. The concept ‘median’ applies because Bede says jacinths are liable to suffer two opposing faults, tenuousness of colour and opacity, the best specimens being free of either.

²⁵⁶ See ‘Lapidary Traditions, part I’, p. 22 (corrected as above, n. 1). Even Hrabanus Maurus, De Universo, which reproduces most of the jewel passage verbatim, has only one of the closing paragraphs.

²⁵⁷ See above, pp. 79 and 86.

²⁵⁸ Quoted above, p. 75.

²⁵⁹ Quoted in part above, pp. 99–100.

²⁶⁰ See above, p. 78.

²⁶¹ ‘The Lord that keepeth Israel shall neither slumber nor sleep.’ These psalms are cxxi and cxxii in the English bible. In the latter the Vulgate has ‘in turribus tuis’ where the AV has ‘within thy palaces’.

²⁶² Bede's choice of Solomon as witness, and emphasis that wisdom must be consummated in action (see above, pp. 82–4), contrast markedly with his exegesis of topaz, the next gem, explicitly in terms of the contemplative monastic life (see above, p. 98 and n. 93).

²⁶³ Gg. 5. 35 is of course as a whole a very cosmopolitan collection; but Dr Lapidge points out that the originally separate section which includes Cives celestis patrie (the ‘Part II’ of Rigg and Wieland, ‘Canterbury Classbook’) contains mostly Insular pieces, and that the one immediately following (ibid. p. 123, ‘Alphabetical hymn on All Saints’) is by the same Anglo-Latin author (probably from Winchester) as wrote hymns to saints Æthelwold, Birinus and Swithun extant in MSS Rouen 1385, British Library, Royal 15. C. vii and Cotton Nero E. i. (The overwhelmingly ‘Anglo-Norman’ distribution of the manuscripts in my stemma is (in view of n. 247 above) probably factitious, a product of the differential interest of editors, and cannot safely be called in evidence here.)

²⁶⁴ Ker, N. R., Catalogue of Manuscripts Containing Anglo-Saxon (Oxford, 1957), no. 2.Google Scholar

²⁶⁵ See ‘Lapidary Traditions, part I’, p. 29.

²⁶⁶ Ambrosius Autpertus comes close to it with ‘auri colore tesplendet’ (CCSL, CM 27A, 822), but the form of the compound goldbleoh, a ἅπαξ λεγόμενον, is likely to reflect a compound adjective, probably nominative, in the text from which the lemma was drawn. The entry belongs to a small glossary (Ker's art. c) not containing any other item which would clinch Cives celestis patrie as a source. (It is pure coincidence that the word clibanus occurs in a lemma on the same page of the manuscript: it belongs to a somewhat later glossary (Ker's art. d) and is probably derived from Isidore (Etym. xv.vi.5), as are the majority of items in that glossary (cf. H. D. Meritt, Fact and Lore about Old English Words (Stanford, Cal., 1954), p. 2). The manuscript page in question is British Library, Add. 32246, 9V; the items on it from Ker's art. c are printed Anglo-Saxon and Old English Vocabularies by Thomas Wright, ed. Wülcker, R. P. (London, 1884), col. 140, 2–6 and 10–24Google Scholar; those from art. d, ibid. col. 138, 2–30, col. 140, 7–9, and col. 140, 25–col. 141, 36.) See also below, n. 271.

²⁶⁷ Bale, J., Scriptorum Illustrium Maioris Brytannie Catalogus 1 (Basle, 1557), 137Google Scholar; followed by Pits, J., Relationes Historicae de Rebus Anglicis (Paris, 1619), p. 175.Google Scholar These references were found by Dr Lapidge, who considers Bale, unlike Pits, reliable enough for the possibility to be raised ‘that he had seen a manuscript in which the metrical vitae of Wilfrid and Audoenus (St Ouen) plus Cives celestis patrie were attributed nominatim to Frithegod’. On Frithegod see Lapidge, , ‘The Hermeneutic Style in Tenth-Century Anglo-Latin Literature’, ASE 4 (1975), 67–111, esp. 78–81.Google Scholar

²⁶⁸ E.g. 10, 3–4 where both balance of construction and the sense of rubere require an adjectival phrase in line 3 parallel to that in line 4, and punctuation no stronger than a comma at the end of line 4 (C's scribe had the right instincts); and, most notably, 6, 1, where the author violates what elsewhere is clearly his norm, to move in two-line units as the music does, and where the scribes responsible for the majority of manuscripts have seen fit to substitute a participle accordingly.

²⁶⁹ (Agius (16.1) is not rare.) The significance of these and most of the following words was brought to my notice by Dr Lapidge.

²⁷⁰ ‘Raving’, in which sense it is not infrequent in secular lapidaries, e.g. the Latin Damigeron on aëtites (Pitra §1, Alphabetic §18).

²⁷¹ Even auricolor, a natural compound to modern English perceptions, is in Latin redolent of poetic artifice (see André, Termes de couleur, pp. 231–2). It was used by the fourth-century Christian poet Juvencus; Mittellateinisches Wörterbuch (Munich, 1959– )Google Scholar gives only four subsequent citations, including Cives celestis patrie.

²⁷² The combination of the relatively rare six-line stanza with relatively smooth rhyme might possibly be made to lead somewhere (even as late as the tenth century); but octosyllabic lines (stressed mainly proparoxytone, as here) were the staple of hymn-writers from Ambrose and Prudentius on.

²⁷³ One cannot say certainly a priori whether the author of Cives celestis patrie would have used a pre-existing chant or composed (or had composed for him) a new one. The considerations rehearsed Planchart, Repertory of Tropes 1, 14–15 (cf. pp. 135, 140 and 391–2), suggest, however, that the tune is very likely to have been newly composed, and that since the member of a monastic community normally responsible for musical composition (where required), the cantor, ‘had in addition the task of providing the monastery with new hymns’, author and composer are quite likely to have been the same man. Either way, the shape of the melody fits the first stanza rather well, and we may think it rather a noble one (‘Eine formal und rhythmisch kunstvoll durchkonstruierte Weise’ as Stäblein puts it, MMMÆ 1, 527). Either way, it exemplifies music current in the author's milieu. But the repertory of church music was cosmopolitan, and one can seldom convincingly assign peculiarities in a given item as between two (or more) possible origins for at least one of which proper comparative material does not exist. (Stäblein, ibid. p. 524, esteems the Moissac hymnal a rich repository of early Gallican musical peculiarities; but he characterizes such pieces inter alia by a certain angularity and lack of finish, and clearly would not include Cives celestis patrie among them.) In the one genre for which appreciable contemporary evidence of Anglo-Saxon usage does exist, French and English music tend to be very similar (Planchart, Repertory of Tropes, passim, esp. 1, 298–9) and arguments for provenance rely almost entirely on the words (ibid., esp. 1, 144–72). The most striking characteristic of this tune, its wide range of an octave and a third, would be exceptional for plainsong anywhere.