¹ Beowulf, 2nd ed., Manchester, 1913, p. xiv.

² Mod. Lang. Review, xi, 335.

³ Gospel of St. John, p. x.

⁴ This implies, of course, no interruption in the scribe's work; but simply, as Skeat points out, that his orthography was in a transitional state.

⁵ Lautlehre d. as. Version d. “Ep. Alex,” p. 5.

⁶ Die Wunder des Ostens, p. 41.

⁷ Beowulf: Quel. u. Forsch., lxii, 240. Cf. P. G. Thomas, Mod. Lang. Review, i, 207.

⁸ Ib., p. 241.

⁹ Now in preparation for the E. E. T. S.

¹⁰ Mod. Lang. Review, i, 203: “… up to the point where the first hand ceases to appear … there are but 11 examples of as against 786 of . From this point to the end … there are 117 examples of as against 482 of . The total number of forms in Beowulf is thus 1268, of 128.” Prof. Klaeber tells me his count tollies with that in the above table.

¹¹ Line references, throughout this discussion, are given as in Cook's Concordance to Beowulf.

¹² Beowulf, Untersuchungen, p. 238, note 1.

¹³ Sievers, likewise, (Zschr. f. Deutsche Phil., xxi, 358) refers gewiofu,—“ bei dem ersten Schreiber, der io nur als u-umlaut von i, nicht auch von e kennt ”,—to gewif. Here, however, as shown by the presence of geogoþ and freoþu, the question is one, not of phonology, but simply of scribal usage.

¹⁴ In Mod. Lcmg. Notes, v, 44, Davidson tries to establish the inaccuracy of the scribe in both Beovmlf B and Judith. His argument in opposition to ten Brink, based, as it is, on the scribe's use of þ and Ð, is by no means so cogent as the analysis of the eo- and io-forms, but is given here as of supplementary value. The second scribe, he claims, is inaccurate not only in his insertion of io-forms into Beowulf B, but also in his reduction of nearly every þ to Ð. The word siþþan, for instance, variously spelt in A, occurs 18 times in B, where it is invariably written syÐÐan. Furthermore, says Davidson, “An examination of Cook's edition of ‘Judith,‘ … fails to reveal a single þ and but one medial—oþre 109—in the 350 lines.” “That this evident levelling of þ and Ð to Ð in B is the work of the scribe and not due to original differences in the text is … reasonably certain…. Indeed, the use of medial and final þ in B is so infrequent and of such a nature that it impresses one as an oversight in the copying of a manuscript in which þ was present.”

In Quellen und Forschungen, lxxi, 103, this usage of Ð is noted by T. G. Foster, who unhesitatingly (accepts ten Brink's theory; but the accuracy of the second scribe is rendered questionable by the “remarkable” “mixture of forms” which Foster himself cites from Judith.

Before ten Brink had advanced his theory, Hornburg had said of the second scribe (Archiv f. n. Sprachen, lxxii, 384): “Eigentümlich ist demselben die ziemlich konsequente Setzung des io für eo und ió für eó; eigentümlich ist ihm ferner die Form telge für talige v. 2068, madelade [MS. maþelade] v. 2426, morna v. 2451, siex v. 2905, u. s. w. Wir sehen aus diesen Bemerkungen dass sich der zweite Abschreiber mit einer gewissen Selbständigkeit bewegt. Daraus folgt, dass er Verfasser etlicher Abschnitte im Gedichte sein kann.”

¹⁵ Consistent use of io-spellings is indicative of accuracy only when he scribe's original is assumed, as by ten Brink, to be one in which io-forms predominate.

¹⁶ Cf. ll. 1946, 1987, 2018, 2127, 2174, 2219, 2336, 2367, 2408, 2425, 2486, 2559, 2599, 2663, 2681, 2693, 2710, 2842, 2883, 2892, 2913, 2931, 2951, 2961, 2972, 2999, 3142, 3150, 3169.

¹⁷ Englische Studien, xiii, 314. Cf. T. G. Foster, Quellen imd Forschungen, lxxi, 51 ff. Foster's dismissal of Moller's conclusion on the ground of “unsettled orthography” is an unwarrantable evasion of the argument.

¹⁸ This statement, of course, is not intended to rule out an Anglian copy.