^{page 1 note 1} For the first one, see In Quest of Virgil's Birthplace, Harvard University Press, 1930Google Scholar.

^{page 1 note 2} ‘Where was Vergil's Farm?’ Chap. II. of Harvard Lectures on the Vergilian Age, Harvard University Press, 1928Google Scholar.

^{page 1 note 3} ‘Further Considerations on the Site of Vergil's Farm,’ Classical Quarterly, XXV. (1931) 65–76Google Scholar. Unless otherwise stated, my references are to the recent article. To the former I will refer as Harv, Lect.

^{page 1 note 4} P. 70.

^{page 2 note 1} Macrobius, (Saturn. V. 2, 1)Google Scholar is only an apparent exception in calling Virgil a Venetian (Veneto rusticis parentibus nato). Macrobius is speaking of the tenth Augustan region of which Mantua is a part. See Nardi, , La Giovinezza di Virgilio, translated by MrsRand, (Harvard Press, 1930), p. 15Google Scholar. I cite the translation in this article rather than the original work, since the former includes various corrections and additions to the latter.

^{page 2 note 2} See Brummer, J., Vitae Vergilianae, p. 1Google Scholar.

^{page 2 note 3} Brummer, , op. cit., p. 73Google Scholar: uico Andico, qui abest a Mantua milia passuum XXX.

^{page 2 note 4} Geschichte der rümischen Literatur, §§ 248, 4; 479, 4. Schanz speaks not only for himself but for Norden, Wessner and others. He gives the literature on the subject.

^{page 2 note 5} Thilo, and Hagen, , Appendix Seruiana, 1902, p. 327, 24–33Google Scholar.

^{page 2 note 6} P. 71.

^{page 3 note 1} P. 71.

^{page 3 note 2} P. 71.

^{page 3 note 3} See In Quest of Virgil's Birthplace, p. 169, n. 99.

^{page 3 note 4} P. 20.

^{page 3 note 5} P. 30.

^{page 3 note 6} P. 31.

^{page 3 note 7} P. 71.

^{page 4 note 1} p. 73.

^{page 4 note 2} Brummer, , op. cit., p. 73, 6Google Scholar. See In Quest of Virgil's Birthplace, p. 136.

^{page 4 note 3} p. 75.

^{page 4 note 4} Ibid, n. 1.

^{page 4 note 5} In l. 16 (cf. my Plate, Figure 114) a similar q, with a waving stroke above, stands for qu(a)e. In l. 20 we have que again, this time with an oblique abbreviation-stroke above the letter. I venture to guess that Crinitus varied his symbols of abbreviation to suit his taste. Thus in l. 19 we have quae again, but this time with a stroke not like that in l. 15 but like that for que in l. 20. A study of the eccentricities of the script of this scholar would be interesting and informing.

^{page 5 note 1} See Thompson, E. M., An Introduction to Greek and Latin Palaeography, Oxford, 1912, p. 471, Facs. 200Google Scholar.

^{page 5 note 2} Steffens, , Lateinische Paläographie, Taf. 115bGoogle Scholar.

^{page 5 note 3} Vicus Andicus, Pubbl. della Reale Accad. Virgiliana di Mantova, VIII., 1930Google Scholar. The author's learning is applied to a fantastic purpose, which leads him to make out Andes a river and to shift Virgil's birthplace from Pietole to Cerese, but his study of the manuscripts of Probus and the edition of Egnatius is sober and sound.

^{page 5 note 4} In Bucolicis quod ad Probi commentariolum attinet secuti sumus netustatem illam quemadmodum ex uetustissimo codice manu scripto Bobii quondam a Georgio Merula inuento adnotauimus. See In Quest of Virgil's Birthplace, p. 164, n. 82.

^{page 5 note 5} Ibid.

^{page 5 note 6} ‘Giulio Pomponio Leto,’ in La Via Letteraria, II. (1919), pp. 30, 48Google Scholar. This publication is not accessible to me.

^{page 5 note 7} Le Scoperte dei Codici Latini e Greci nei Secoli XIV e XV, Florence, 1914, p. 246Google Scholar. (He mentions the commentary of Probus as known to Pomponius Laetus.)

^{page 5 note 8} Op. cit., p. 5.

^{page 6 note 1} P. 74. It was not correct to state that I confined myself to a study of the variations in the Life itself, when I had pointed out (p. 168, n. 91) ‘a number of other certain instances’ in the Commentary.

^{page 6 note 2} P. 138.

^{page 6 note 3} I refrained from an attempt to classify the manuscripts of the Z class, though pointing out that PM seem to stand together now and then against V, which seemed to bear traces of emendation (op. cit., pp. 168 sq., nn. 92, 95). According to Dal Zotto, who nevertheless admits the possibility of my (tentative) view, V is the parent of PM (op. cit., p. 7). For my part, I will not be rash enough to deny this possibility, or the possibility that V is the manuscript used for the edition of 1471. Judgment must be suspended till all the evidence is in. One interesting thing about V is the presence of variants added from some codex antiquus, presumably that of Bobbio (see e.g. Thilo, and Hagen, , Appendix Seruiana, pp. 325, 20Google Scholar; 326, 17). Do these variants attest a fresh consultation of the ancient book after the copy on which VPM depend was made? Or, as Dal Zotto thinks, have PM merely neglected certain features of that copy? Such questions must wait for their answer.

^{page 6 note 4} P. 73. His further statement that I try to show ‘that the evidence of the four is no better than the evidence of the one, because [italics mine] all five (so he suggested) go back to one original’ involves a complete misapprehension of what I said and of what a textual critic should say. If all five go back to one original by independent paths, then the evidence of four against one is of course decisive. The point is that four of the five are not independent, but are dependent on a copy made from the original.

^{page 7 note 1} op. cit., p. 10.

^{page 7 note 2} I provided for the latter possibility (op. cit., p. 138), and it may be necessary to assume it. Conway (p. 73) finds it strange that, despite my ‘desire to raise the credit of the Bobbio codex,’ I should assume ‘that Egnatius’ own statement is untrue, and that some one manuscript intervened between his edition and the ancient codex of Bobbio.’ But where does Egnatius say that no copy intervened between his text and the manuscript? At least he himself made a copy for the printer, and in copying could have committed scribal errors. Nor is his statement (see above, p. 5, note 4) a lie if someone else made the copy, whether the manuscript was at Bobbio or elsewhere. Dal Zotto thinks it possible that the old codex may have been acquired by Egnatius at that time, and he is reminded of the manuscripts, both Greek and Latin, owned by the Aldine Press (op. cit., p, 11). That possibility, too, I would not deny, though we might expect a more positive statement from Egnatius, such as Aldus made on the subject of the ancient manuscript of Pliny's Letters brought down to him from Paris, which forms the basis of his famous edition of 1508. This is another matter on which judgment must await further investigation.

^{page 7 note 3} op. cit., p. 137.

^{page 7 note 4} Conway remarks (p. 75) that ‘the omission of annis liberali does not suggest that Egnatius‘ codex was a better or older source than that of those from which our other authorities are derived.’ He misses my point, here and elsewhere. I did not assert that E is everywhere better than VMPR; rather I sought to show from the presence of an error of precisely this kind that his is an honest text unadulterated with the traditional humanistic emendations. If he knew the edition of 1471 or a manuscript of the type of VPM, he did not use it.

^{page 7 note 5} Thilo, and Hagen, , Appendix Scruiana, pp. 344–346Google Scholar. See Zotto, Dal (op. cit., p. 12)Google Scholar, who states that Hagen, to whom only Keil's material was accessible, calmly cites E for seventeen variants in this passage, which E did not contain. One wonders what edition of Egnatius was accessible to Keil. A new collation is obviously needed. Incidentally, Dal Zotto (p. 8) calls the Basel edition of 1534 ‘una seconda edizione.’ The real date of the second edition is 1520. It appeared in Venice (in aedibus Georgii de Rusconibus) and Harvard College Library has a copy of it.

^{page 7 note 6} Even to the point of doing what he can for the reading of VMPR (non tibi sed for non sinis et) in the quatrain on the burning of the Aeneid (pp. 74 sq.). He does not mention the all-important point, made in my statement, that non sinis et is the reading in Donatus's Life of the poet (op. cit., p. 168, n. 92). Conway remarks that I follow Brummer's punctuation:

‘Tucca uetat Variusque simul; tu, maxime Caesar,

non sinis et Latiae consulis historiae.’

And in order to keep non tibi sed in the last line, he would punctuate:

Tucca uetat Variusque simul tu maxime Caesar;

Non tibi sed Latiae consulis historiae.

The construction surely becomes less easy on this interpretation, and a glance at my facsimile will show that, for the main point at issue, I am following not only Brummer's punctuation (not quite accurately given by Conway), but that of PM and R (V is not reproduced, but I venture to guess that it agrees with the rest). In all of these tu has a capital initial and there is no pause after ‘Caesar.’

^{page 8 note 1} P. 75.

^{page 8 note 2} Thilo, and Hagen, , Appendix Seruiana, p. 333, 19Google Scholar.

^{page 8 note 3} Cf. e.g. ibid., p. 325, 11, 15.

^{page 8 note 4} P. 73.

^{page 8 note 5} op. cit., p. 138.

^{page 8 note 6} See my articles, especially the first one, on ‘A New Approach to the Text of Pliny's Letters,’ in Harvard Studies in Classical Philology, XXXIV., 79–191, and In Quest of Virgil's Birthplace, p. 165 n. 83Google Scholar.

^{page 9 note 1} P. 73.

^{page 9 note 2} Fol. 173 in the Harvard copy.

^{page 9 note 3} They are not mentioned in the colophon: P. V. Eucolica & Georgica & Aeneidos Libri XII. Venetijs excusi M. D. VII. Die ultimo Iunij. Lauretani Principis Anno Sexto. Bernardinus Stagninus Impensam fecit, Ioannes Baptista Egnatius Venetus Emendauit. Dij Caeptis Faueant.

^{page 9 note 4} The remarks that follow may be noted in my facsimile (p. 134), to which Conway refers.

^{page 9 note 5} Illud quod in Maroniani carminis emendatione secuti sumus, diligentem lectorem admonitum praecipue uelim nos in eo emendando multa ad uetustae consuetudinis normam, cuius Vergilium obseruantissimum fuisse constat, direxisse, in plerisque antiquioribus potius auctoribus quam recentibus grammaticis adstipulatos.

^{page 10 note 1} Fol. 174.

Neque enim ita a uetusta consuetudine steti, ut semper id secutus sim; gustum tamen quendam praebere uolui ut, si probares, haberes et tu quid in aliis eiusdem notae sequerere.

^{page 10 note 3} It would be interesting to go through his text and see what they are.

^{page 10 note 4} Quorum censura iam sibi tantum uindicabat licentiae, ut pro libidine sua poetarum albo quoddam adscriberent, quoddam uero penitus eraderent.

^{page 10 note 5} Quae passim sub eruditissimorum hominum titulis emendata circumferuntur, addita quoque eorum epistula, quam uiri illi eruditi in codicibus a se primum emendatis apposuissent.

^{page 10 note 6} Adque ad hanc normam alia quaedam parui omnino momenti facile corriges in graecis dictionibus, in quibus quandoque accentus deerit, saepe superfluet, nonnunquam litera una subsultabit; quae facile legendo percipias.

^{page 10 note 7} Nos in nostris Racemationibus aestimatione quadam censuimus legendum ‘omnipatentis’; neque enim uideo quid ibi sibi uelit epitheton Iouis. Tu uide ne ex emendatione nostra ‘omnipatentis’ quoque legendum censeas.

^{page 11 note 1} In Seruio illud de Oaxe ex scripto manu Seruio reposuimus, quod miror a Politiano aut non uisum aut omnino dissimulatum, ubi duos quoque uersus Attacini Varronis, qui corruptissime legebantur, ex Apollonii lectione restituimus. The comment of Servius is that on Ecl. I. 65. The verses as given by Egnatius (without verse distinction) are: Quos magno anchiale partus adducta dolore: & geminis capiens Tellurem oaxida palmis scindere dicta. The passage in Apollonius is Argon. I. 1129 sq. Thilo, in the Thilo and Hagen edition of Servius, III. 1 (1887), p. 15, notes that Salmasius made use of Apollonius, but makes no mention of the contribution of Egnatius over a century before.

^{page 11 note 2} P. 10, n. 2.

^{page 11 note 3} See op. cit., p. 137, n. 92.

^{page 11 note 4} op. cit., pp. II sq.

^{page 12 note 1} Ibid., p. 15.

^{page 12 note 2} See the facsimile in the De Vries series, fol. 4^r, l. 16 (XLI. 13, 7), where trecenta milia septem written in figures, the VII looking altogether like un.

^{page 12 note 3} P. 72.

^{page 12 note 4} op. cit., pp. 140, 141, n. 97, with corroboration not only from Nardi (a Mantuan) but from Funaioli (a Milanese), whose experience manuscripts Conway will admit is certainly not inferior to our own.

^{page 13 note 1} P. 72.

^{page 13 note 2} See In Quest of Virgil's Birthplace, p. 137.

^{page 13 note 3} Ibid., p. 61.