^{page 77 note 1} Tatuini Opera Omnia: Ars Tatuini, ed. M. de, Marco, Corpus Christianorum Series Latina 133 (Turnhout, 1968), 1–93.Google Scholar

^{page 77 note 2} By Cobbs, S. P., ‘Prolegomena to the Art Grammalica Talvini’ (unpubi. Ph.D. thesis, Chicago, 1937), pp. 69–71Google Scholar; Löfstedt, B., Der biberno-lateinische Grammatiker Malsachanus (Uppsala, 1965) p. 17, n. 4Google Scholar; de Marco, pp. vii–xi.

^{page 77 note 3} None of these manuscripts was listed in Ker's, N. R.Catalogue of Manuscripts containing Anglo-Saxon (Oxford, 1957)Google Scholar, but they have been included in the supplement to the Catalogue, ASE 5(1976), 121–31 (p. 131).Google Scholar

^{page 77 note 4} Codices Latini Antiquiores, ed. Lowe, E. A., 12 vols. (Oxford 1934–1975), Supplement, no. 3775.Google Scholar

^{page 77 note 5} CLA v, no. 672.

^{page 77 note 6} On this attribution, see an article which I have in preparation.

^{page 78 note 1} CLA VIII, no. 1127.

^{page 78 note 2} Other surviving fragments of this manuscript are Carlsruhe Fragm. Aug. 116–19, 122, 126, 128–31, 234 and 136 and Aug. cxvi (binding); St Paul in Carinthia, Stiftsbibl., 979/O; Vienna, Nationalbibl., Lat. 482; and Zürich, Staatsarchiv, A.G. 19, no. XIII. On this manuscript, see the article ref007 to above, p. 77, n. 6.

^{page 78 note 3} The two corrupt forms dignin and digin indicate the otherwise unattested form *dignin (= West Saxon þinen).

^{page 78 note 4} The form croþt either represents an archaic spelling of croft, or is a scribal error for croft.

^{page 78 note 5} Nraf and baraes are corrupt renderings of nraef (fearn spelt backwards).

^{page 79 note 1} The horizontal stroke over the gloss in P is the only example in the Tatwine glosses of the practice of indicating a vernacular gloss by means of such a line, as found in e.g. the Leiden glossary (ed. Hessels, J. H. (Cambridge, 1906), § 23d).Google Scholar

^{page 80 note 1} Brunner, K., Altenglische Grammatik nach der angelsächsischen Grammatik von Eduard Sievers neubearbeitet (Halle, 1942; revised repr. 1951), §44, n. 6.Google Scholar

^{page 80 note 2} Ibid. 199, n. 1.

^{page 80 note 3} Three Northumbrian Poems, ed. Smith, A. H. (London, 1968), p. 38.Google Scholar

^{page 80 note 4} The form of the correction of vilps into the unmetathesized vlisp in both L and P may imply a marginal alteration in the archetype.

^{page 80 note 5} Campbell, A., Old English Grammar (Oxford, 1959, repr. with Corrections 1974), §60.Google Scholar

^{page 80 note 6} Ibid.

^{page 80 note 7} Ibid.

^{page 80 note 8} Ibid. 26, n. I.

^{page 80 note 9} Ibid. 243.

^{page 80 note 10} Ibid.§261 and 299.

^{page 80 note 11} Ibid. 51, n. 2.

^{page 81 note 1} The sources for the life of Tatwine are Bede, Hisloria Ecclesiastica v.23 and 24; Continuatio Baedae sub annis 733 and 734 (both works ed. Plummer, C., Venerabilis Baedae Opera Historica (Oxford, 1896, repr. 1975))Google Scholar; and the epitaph printed in Hearne's, T. edition of Leland's Collectanea (Oxford, 1715) 111, 116Google Scholar, and recently reprinted by Lapidge, M., EHR 90 (1975), 822–12.Google Scholar

^{page 81 note 2} Alnus is glossed as scip in what is clearly indicated as a list of trees.

^{page 81 note 3} There is thirteenth-century evidence that the manuscript of the Corpus glossary (see below), Cambridge, Corpus Christi College 144, was in the library of St Augustine's by then.

^{page 82 note 1} Bosworth, J. and Toller, T. N., An Anglo-Saxon Dictionary (Oxford, 1882–1898)Google Scholar, with a supplement by Toller, T. N. (1908–1921), s.v. sandbyll.Google Scholar

^{page 82 note 2} Pheifer, J. D., Old English Glosses in the Épinal-Erfurt Glossary (Oxford, 1974), p. 63.Google Scholar

^{page 83 note 1} Steinmeyer–Sievers 111, 510, 5.

^{page 83 note 2} Ibid., 684, 52.

^{page 83 note 3} BT, s.v. scill.

^{page 83 note 4} This line is quoted twice in Sergius's explanationes in Donatum ed. Keil, H., Grammatici Latini IV (Leipzig, 1865), 486–565Google Scholar, a grammatical treatise known to the English at an early date.

^{page 83 note 5} Old English Glosses (A Collection), ed. Meritt, H. D. (New York and London, 1945), p. 28, 384.Google Scholar

^{page 83 note 6} Mr P. Sims-Williams has pointed out that *fau would be the British form derived from Latin fagus: cf. Breton faou Welsh ffawydd < *ffaw + gwŷdd.

^{page 84 note 1} Josef, Hofmann, ‘Altenglische und althochdeutsche Glossen aus Würzburg und dem weiteren angelsächsischen Missionsgebiet’, Beiträge zur Geschichte der deutschen Sprache und Literatur 85 (1963), 56.Google Scholar

^{page 84 note 2} The related glossary in Berne 258 has grist (Steinmeyer–Sievers 1, 497, 1–2).

^{page 84 note 3} Wright–Wülcker, 304–37.

^{page 84 note 4} 414, 25.

^{page 84 note 5} Leehdoms, Wortcunning and Stareraft of the Anglo-Saxon, ed. Cockayne, O., Rolls Ser. (1864–1866), 11, 38a.Google Scholar

^{page 84 note 6} CGL 11, 563–97.

^{page 85 note 1} Virgilii Maronis Grammatici Opera, ed. Huemer, J. (Leipzig, 1886), p. 76.Google Scholar

^{page 85 note 2} E.g. carmen de virginilate (ed. Ehwald, R., Aldhelmi Opera Omnia, Monumenta Germaniae Historica Auct. Ant. 15)Google Scholar, praefatio, line 38.

^{page 85 note 3} Cobbs, S. P. (‘Prolegomena’, p. 71)Google Scholar understands gif it as gifeðe, an identification I find difficult on morphological grounds, although semantically convincing.

^{page 85 note 4} Gif it cannot be defended semantically by comparison with fortisan: wen is Harley 3376, for the two are not exactly parallel: wen is is an adverbial phrase which may be used together with gif.

^{page 85 note 5} The Old English Prudentius Glosses at Boulogne-sur-Mer, ed. Meritt, H. (Stanford, 1969), p. 15, 143.Google Scholar

^{page 85 note 6} Cartilago is not included in this summary, since it and its Old English equivalent gristle are of infrequent, glossary, occurrence and thus shed no light on the affiliations of the Tatwine glosses.

^{page 85 note 7} Cp often has two glosses for the same lemma, the one gloss using an interprelamentum also found in Ep–Erf and the other gloss being independent.

^{page 87 note 1} Ed. Keil, H., Grammatici Latini v (Leipzig, 1868), 410–39.Google Scholar

^{page 87 note 2} Pheifer, , Old English Glosses, p. xlvi.Google Scholar

^{page 87 note 3} The ð in the Phocas items ðrotae and ðaca, and the d in the borrowed dignin might ref052 the different origins.

^{page 88 note 1} Elym. XII.vi. The glossator has lost the point of the etymology by substituting nimis for multum and reversing the word order.

^{page 88 note 2} Ed. Keil, , Grammatici Latini III (Leipzig, 1839), 441–56.Google Scholar

^{page 88 note 3} Charisii Artis Grammaticae Libri V, ed. Barwick, C. (Leipzig, 1964), p. 43, 10Google Scholar: ‘lien σπλήν lienis’.

^{page 88 note 4} Institutiones gyammaticae, ed. Hertz, M., Grammatici Latini (II and III, 1–77; Leipzig 1850 and 1859) II, 149, 7fGoogle Scholar. Splen here takes on the status of a Latin word in its own right: ‘“lien”, “rien” vel “rēn” et “splēn splenis”’. Both renis and splenis occur in a list of nouns in -is in the ars Tatuini (37, 1117), renis alone in the ars Bernensis (ed. Hagen, H., Anecdota Helvetica (Leipzig, 1870), pp. 62–142)Google Scholar (117, 15), splenis alone in the unprinted grammar in Amiens, Bibliothèque Municipale, 426 (54v), all of which are insular in origin.

^{page 89 note 1} I should like to acknowledge the kindness of Professor P. A. M. Clemoes, Mr Patrick Sims-Williams and especially Dr Michael Lapidge in reading this article in typescript and making many helpful suggestions.