¹ The manuscript in its entirety consists of New Haven, Beinecke Rare Book and Manuscript Library, 401 (26 leaves) + 401 A (2 leaves) + Oxford, Bodleian Library, lat. theol. d. 24 (2 leaves) + Bodleian Library, Don. f. 458 (2 leaves) + Cambridge, University Library, Add. 3330 (2 leaves) + Oslo and London, Schøyen Collection, 197 (2 leaves) + London, British Library, Add. 50483K (1 leaf) + Philadelphia, Free Library, J.F. Lewis Collection, European Text Leaf 121 (1 leaf). See Ker, N. R., Catalogue of Manuscripts Containing Anglo-Saxon (Oxford, 1957), pp. 10–11 (no. 12)Google Scholar, and Schailor, B., Catalogue of Medieval and Renaissance Manuscripts in the Beinecke Rare Book and Manuscript Library, Yale University, II: MSS 250–500 (Binghamton, NY, 1987), pp. 280–4.Google Scholar

² Napier, A. S., Old English Glosses, Chiefly Unpublished (Oxford, 1900), no. 11Google Scholar. The ink glosses in Cambridge, UL, Add. 3330 were published by Napier, ibid. no. 12; those in the Schøyen leaves (formerly Malibu, California, J. Paul Getty Museum, Ludwig XI.5) were published by Meritt, H. D., ‘Old English Aldhelm Glosses’, MLN 68 (1952), 553–4Google Scholar; and those in Philadelphia were published by Collins, R., Anglo-Saxon Vernacular Manuscripts in America (New York, 1976), pp. 32–3.Google Scholar

³ Napier, , Old English Glosses, p. xxxiiiGoogle Scholar, and Meritt, H. D., ‘Old English Glosses, Mostly Dry-point’, JEGP 60 (1961), 441–50 (no. 1).Google Scholar

⁴ All but five of the dry-point glosses appear in Beinecke 401; the other five are in Cambridge, UL, Add. 3330. I have not examined the leaves in the Schøyen collection, but their position in the manuscript leads me to believe that they contain no dry-point glosses.

⁵ See Lapidge, M., ‘The Study of Latin Texts in Late Anglo-Saxon England: the Evidence of Latin Glosses’, Latin and the Vernacular Languages in Early Medieval Britain, ed. Brooks, N. (Leicester, 1982), pp. 99–140Google Scholar, and also Wieland, G., ‘The Glossed Manuscript: Classbook or Library Book?’, ASE 14 (1985), 153–73.Google Scholar

⁶ On medieval recipes for ink, see Wattenbach, W., Das Schriftwesen im Mittelalter, 3rd. ed. (Leipzig, 1896), pp. 233–44Google Scholar and Bat-Yehouda, M. Zerdoun, Les Encres noires au moyen âge (jusqu' à 1600) (Paris, 1983).Google Scholar

⁷ See Wattenbach, , Schriftwesen, p. 227Google Scholar, and Bischoff, B., Latin Palaeography, trans. Cróinin, D. Ó and Ganz, D. (Cambridge, 1991), p. 18.Google Scholar

⁸ On the use of the wax tablet and stylus in the Middle Ages, see Wattenbach, , Schrifwesen, pp. 51–89 and 219–22Google Scholar; R. H. and Rouse, M. A., ‘The Vocabulary of Wax Tablets’, Harvard Lib. Bull. ns 1.3 (1990), 12–19Google Scholar; and Bischoff, , Latin Palaeography, pp. 13–14.Google Scholar

⁹ See Bischoff, B., ‘Über Einritzungen in Handschriften des frühen Mittelalters’, Mittelalterliche Studien, 3 vols. (Stuttgart, 1966–1981) I, 88–92.Google Scholar

¹⁰ I have designated the glossing Hands A through F, based on their order of appearance in the manuscript.

¹¹ Hand B's characteristic letter forms include Insular f, g, h, r and s, square a, round d, straight-limbed y, high e in ligature and st ligatures.

¹² See Ker, Catalogue, no. 12 and Napier, , Old English Glosses, p. xviiGoogle Scholar. I do not think that Hand B is the same as the second ink glossator because of instances where the ink glossator wrote differing glosses over ones by Hand B.

¹³ These are mentioned in the notes to the text of the glosses, below (pp. 207–13).

¹⁴ Both ink glossators are dated to the second half of the tenth century; see Ker, Catalogue, no. 12.

¹⁵ Nos. 46, 81 and 91. These instances demonstrate some of the complexities of reading scratched glosses. When viewed with a strong, oblique light the glosses of Hand B are clear, but because the instrument used by Hand C scrapes the parchment rather than makes an impression, his glosses are invisible. Upon shifting the page so that the light is dim, Hand C's glosses become visible while Hand B's disappear. Had the glosses by Hand B been clearly visible, Hand C would probably not have written other glosses over them, providing evidence that dry-point glosses were as difficult to see in the tenth century as they are today.

¹⁶ Lowe, E. A., Codices Latini Antiquiores, 11 vols. and supp. (Oxford, 1934–1971; 2nd ed. of vol. II (1972)) II, ixGoogle Scholar, lists the manuscript among those examined and rejected for a pre-800 date. See also Morrish, J., ‘Dated and Datable Manuscripts Copied in England during the Ninth Century: a Preliminary List’, MS 50 (1988), 512–38, at 527.Google Scholar

¹⁷ See Lowe, E. A., ‘Membra Disiecta’, RB 39 (1927), 191–2Google Scholar: ‘The Anglo-Saxon script has some resemblance to Mercian charters, and Nicholson's guess of a Worcester origin may not be wide of the mark, though some features remind one of Canterbury.’

¹⁸ On the language of the ink glosses, see Napier, , Old English Glosses, p. xxxii.Google Scholar The following Kenticisms are found in the dry-point glosses: c for WS æ. (nos. 7, 22 and 28): e for WS æ: (51); e for WS ie (30 and 137); e for WS y (58). The confusions in spelling also show Kentish influence: æ for WS e (39, 68, 70 and 107); æ for WS ea (106); and perhaps æ for WS i (101). The Kentish letter e appears several times in inflectional endings, most likely showing a confusion in spelling with e rather thn a true dialectal variant (14, 48 and 98). See Zupitza, J., ‘Kentische Glossen des neunten Jahrhunderts’, ZDA 21 (1987), 1–58Google Scholar, and Williams, I., ‘A Grammatical Investigation of the Old Kentish Glosses’, Banner Beiträge zur Anglistik 19 (1905), 92–166.Google Scholar

¹⁹ See Bishop, T. A. M., English Caroline Minuscule (Oxford, 1971), esp. pp. xxi–xxiiiGoogle Scholar and Dumville, D. N., English Caroline Script and Monastic History: Studies in Benedictinism, A.D. 950–1030 (Woodbridge, 1993), esp. pp. 86–90.Google Scholar

²⁰ See Lapidge, M., ‘The Hermeneutic Style in Tenth-Century Anglo-Latin Literature’, ASE 4 (1975), 67–111, at 77–85.Google Scholar

²¹ Cambridge, Corpus Christi College 326 (Christ Church, s. x²), London, British Library, Royal 5. E. XI (Christ Church, s. x/xi) and Royal 6. A. VI (Christ Church, s. xi¹), London, Lambeth Palace 200 (St Augustine's, s. x²), Oxford, Bodleian Library, Bodley 97 (Christ Church, s. xiⁱⁿ) and Salisbury, Cathedral Library, 38 (Christ Church, s. x^ex). Also from Canterbury is London, BL Cotton Cleopatra A. iii (St Augustine's, s. x^med), a glossary which contains four batches from De laudibus virginitatis.

²² Lambeth 200 has no Old English glosses. The dry-point glosses in CCCC 326, Royal 5. E. XI and Royal 6. A. VI are noted by Napier, , Old English Glosses, p. xxxiii.Google Scholar Those in Salisbury 38 he prints in ‘Collation der altenglischen Aldhelmglossen des Codex 38 der Kathedralbibliothek zu Salisbury’, Anglia 15 (1893), 204–9.Google ScholarMeritt, H. D. prints dry-point glosses from CCCC 326 and Royal 5. E. XI in Old English Glosses, a Collection (New York, 1945), nos. 1 and 2.Google Scholar A careful examination of Bodley 97 might well reveal dry-point glosses which have not yet been noticed.

²³ The chief manuscripts of this tradition are Oxford, Bodleian Library, Digby 146 and Brussels, Royal Library, 1650, each of which contains over 5000 Old English glosses. On the language of these glosses, see Goossens, L., The Old English Glosses of MS Brussels, Royal Library, 1650 (Brussels, 1974), pp. 54–139.Google Scholar

²⁴ I have confined my discussion of the dry-point glosses to those written by Hand B, the main glossator. The other glossators wrote too few glosses for an adequate analysis.

²⁵ Only two of the glosses are in Latin, nos. 2 (universa; omnes) and 43 (ingenti: inmensi).

²⁶ There are, however, no syntactical marks among the glosses. R.I. Page notes the presence of dry-point syntactical glosses in a Boethius manuscript (Cambridge, UL, Kk. 3. 21) in ‘New Work on Old English Scratched Glosses’, in Studies in Early English Language and Early Literature in Honour of Paul Christophersen, ed. Tilling, P. M. (Ulster, 1981), pp. 105–15, at 110Google Scholar, and in a Sedulius manuscript (CCCC 173), in ‘More Old English Scratched Glosses’, Anglia 79 (1979), 27–45, at 44.Google Scholar On syntactical glosses in general, see Robinson, F. C., ‘Syntactical Glosses in Latin Manuscripts of Anglo-Saxon Provenance’, Speculum 48 (1973), 443–75CrossRefGoogle Scholar, and Korhammer, M., ‘Mittelalterliche Konstruktionshilfen und altenglische Wortstellung’, Scriptorium 34 (1980), 18–58.CrossRefGoogle Scholar

²⁷ No. 27.

²⁸ No. 9.

²⁹ No. 74.

³⁰ On grammatical glosses, see Wieland, G., The Latin Glosses on Arator and Prudentius in CUL Gg.5.35 (Toronto, 1983), pp. 47–97.Google Scholar

³¹ No. 47.

³² No. 8. This gloss is by Hand A.

³³ Wieland distinguishes five types of glossing (lexical, grammatical, syntactical, prosodic and commentary) in The Latin Glosses on Arator and Prudentius.

³⁴ Because there are no letter forms, it is impossible to prove that these marks are by the main glossator, but they show the same type of impression in the parchment, and they only appear on those leaves which he glossed.

³⁵ No. 75. The gloss reads oferteoh. The passage reads ‘ut humanis faucibus et mortalium palato omne, quicquid iocundum ac delectabile illatum sentitur, mellitae dulcedinis gustum ammodum incomparabiliter praecellit’ (ed. Ehwald, R., Aldhelmi Opera, MGH, Auct. Antiq. 15 (Berlin, 1919), 234)Google Scholar: ‘just as the taste of honeyed sweetness quite incomparably excels everything that is experienced as pleasing and delectable when brought to human mouths and the palate of mortals’, trans. Lapidge, M. and Herren, M., Aldhelm: the Prose Works (Cambridge, 1979), p. 63.Google Scholar

³⁶ No. 45. ‘Quid enim, quaeso, in rerum visibilium videri valet natura, quod tam ingenti studio auctoris sui praecepto pareat’ (ed. Ehwald, , Aldhelmi Opera, p. 233)Google Scholar: ‘What, I ask, in the nature of visible things can be seen, that obeys the command of its begetter’, trans. Lapidge, and Herren, , Aldhelm: the Prose Works, p. 63.Google Scholar

³⁷ No. 105. ‘Non enim splendida meri argenti species turpiter deformatur’ (ed. Ehwald, , Aldhelmi Opera, p. 236)Google Scholar: ‘For the radiant beauty of pure silver is not shamefully debased’, trans. Lapidge, and Herren, , Aldhelm: the Prose Works, p. 65.Google Scholar

³⁸ Korhammer had discovered that, for didactic purposes, syntactical glosses in English manuscripts frequently construe the verb as the first word in the sentence, regardless of standard Old English word order: see ‘Mittelalterliche Konstruktionshilfen’, pp. 41–6; see also Wiesenekker, E., Word be Worde, Andgit of Andgite: Translation Performance in the Old English Interlinear Glosses of the Vespasian, Regius, and Lambeth Psalters (Huizen, 1991), pp. 54–8.Google Scholar

³⁹ ‘Quamdiu enim antiquas inhabitare sedes et exigua fovere tuguria … ille, qui inter ceteras magistratus officio fungitur, decreverit’ (ed. Ehwald, , Aldbelmi Opera, p. 233)Google Scholar: ‘For as long as that bee who among others discharges the office of magistrate, shall decree that they should inhabit their ancient dwellings and care for their little cottages’, trans. Lapidge, and Herren, , Aldhelm: the Prose Works, p. 62.Google Scholar

⁴⁰ No. 83. Ælfrics Grammatik and Glossar, ed. Zupitza, J. (Berlin, 1880), p. 52.Google Scholar

⁴¹ No. 24.

⁴² ‘Old English Glosses’, p. 441, n. 4.

⁴³ ‘Illud etiam commemorandum de apum concordi sodalitate et theatrali quodam spectaculo stupendum autumno’ (ed. Ehwald, , Aldhelmi Opera, p. 233)Google Scholar: ‘This also is to be remembered, I suggest, concerning the harmonious fellowship of the bees, and to be admired as some theatrical spectacle’, trans. Lapidge, and Herren, , Aldhelm: the Prose Works, p. 62.Google Scholar

⁴⁴ The phonological similarity between these words is greater considering that the glosses in this manuscript, and other Kentish glosses, frequently show a loss of d in this position. See no. 12 and I. Williams, ‘A Grammatical Investigation’, §90.4.b.

⁴⁵ No. 75. For the context, see above, n. 38. See also Meritt, Old English Glosses, A Collection, no. 2.52.

⁴⁶ We know that at least one teacher translated Latin texts into Old Englïsh for his students, namely Æthelwold: ‘Dulce namque erat ei adolescentes et iuuenes semper docere, et Latinos libros Anglice eis soluere’: Wulfstan of Winchester's Life of St Ætheltvold, ed. Lapidge, M. and Winterbottom, M. (Oxford, 1991), pp. 46–8 (ch. 31).Google Scholar See also Bullough, D. A., ‘The Educational Tradition in England from Alfred to Ælfric: Teaching utriusque linguae’, SettSpol 19 (1972), 453–94.Google Scholar

⁴⁷ In his studies of the Latin glosses of Arator and Prudentius, Wieland concluded that ‘Arator's work was studied more for grammar and lexicon, while Prudentius' was examined more for allegorical and anagogical meanings, and that Arator occurred earlier in the medieval curriculum than Prudentius.’ See ‘Latin Lemma – Latin Gloss: the Stepchild of Glossologists’, Mittellateinisches jahrbuch 19 (1984), 91–9, at 98.Google Scholar

⁴⁸ For example, both Meritt and Page have examined and printed dry-point glosses from Cambridge, UL, Kk. 3. 21 (Boethius: Abingdon, s. x/xi), yet I have seen in the manuscript a large amount still unpublished, many of which are clearly visible.

⁴⁹ A grant from the English Speaking Union of the United States and from the John F. Enders Foundation of Yale University made it possible for me to travel to England to examine the leaves there. I would like to thank Professor Fred C. Robinson and Dr Robert Babcock for reading and commenting on earlier drafts of this paper.