^{page 87 note 2} Butcher, W. G. D., ‘The Caesura in Vergil,’ Classical Quarterly, viii p. 127Google Scholar.

^{page 87 note 3} This was written before the appearance of Prof. A. Gercke's Die Entstehung der Aeneis. I am glad to see that he takes maior maius to allude to the Eclogues and Georgics, quoting in addition Ecl. iv. 5, magnus ab integro saeclorum nasciturordo (pp. 74, 75).

^{page 87 note 4} Vol. vi. p. 239.

^{page 88 note 1} Prof. Gercke (I. c.chap, iv.) draws similar inferences from the lines of Propertius, but thinks (following Rothstein) that they also refer to the prooemium to book i.:

Arma uirumque cano, Troiae qui primus ab oris litora, multum ille et terris iactatus et al to ui superum saeuae memorem Iunonis ob iram, multa quoque et bello passus, dum conderet urbem inferretque deos Latio, genus unde Latinum the Albanique patres atque altae moenia Romae. (i. 1–7).

He would conclude therefore that ‘im Jahre 26 v. Chr. Vergils Aeneis mit den Versen i. 1–7+ vii. 37 et seq. begann, und dass der Dichter die ersten drei oder vier Jahre seiner Arbeit an dem neuen Epos lediglich auf die Dichtung der römischen Ilias verwendet bat ’ (p. 76). This seems very plausible, though it does not appear necessary to refer

iactaque Lauinis moenia litoribus

to the foundation of Lavinium, which does not occur in the existing epic, and therefore to suppose a later curtailment of Vergil's plan. The line quoted above (Aen. vii. 157) is enough to justify the reference in Propertius.

I might say here that I am unable to follow Prof. Gercke in his elaborate attempt to dissect the Aeneid and date its various portions, but I am in substantial agreement with chap. iv., so far as it deals with the priority of book vii. In chap. v. Prof. Gercke proceeds to analyze book vii. and the rest of the last half of the Aeneid, and to label the fragments ‘alt,’ ‘jung,’ etc. His judgments would in many cases run entirely counter to the literary considerations I have tried to bring forward in this paper.

^{page 89 note 1} Hermes xxiv (1889) 1–34Google Scholar

^{page 89 note 2} Literary History of Rome, p. 487.

^{page 89 note 3} Class. Rev. xx. 113et seq.

^{page 90 note 1} Metam. xi. 119–124.

^{page 90 note 2} Nouvells Promenades Archéologiques, p. 272, n.

^{page 90 note 3} Virgils Epische Technik, pp. 85–93.

^{page 90 note 4} In the first book Aeneas appears, as well as the sovereign and the ambassador, so the episode is longer and more elaborate. There are also parallels to the meeting with Euander in book viii.

^{page 91 note 1} Class. Rev. xx. 304 sq.

^{page 91 note 2} See note above, Cereale…Cererem, vii. 111, 113.

^{page 91 note 3} There are various parallels with the Euander episode; note especially viii. 171: auxilio laetos dimittam opibusque iuuabo compared with i 571: auxilio tutos dimittam opibusque iuuabo.

^{page 92 note 1} It is perhaps worth while to note that each passage begins with a reference to Sicily: vii. 298–9,

etlaitum Aeneam classemque ex aethere longe Dardaniam Siculo prospexit ab usque Pachyno. i. 34,35,

uix e conspectu Siculae telluris in altum uela dabant laeti.

The chiastic arrangement (Laeidm…siculopro Spexit con Spectu Sicular…laeti) seems to indicate that the correspondence is not accidental.

^{page 93 note 1} I have note d a number of passages where an author's bold construction seems a case of ‘going one better’ than himself in a previous passage, or the predecessor he is imitating. E.g. would Horace have written curis…exptditis, Odes I. xxii. 11, if it had not been for solutis…curis, Catullus xxxi. 7. 2 ? Again, perhaps ‘mihi cumque salue rite uocanti’ (Odes I. xxxii. 15, 16) can best be explained as a further experimentation of Horace in the use of the participles, led up to by quippe…reuisens in the preceding ode (11. 13, 14), which is of course modelled on áre in Greek with the participle.

^{page 93 note 2} Nettleship (Verg. p. 131) has already pointed out that Amata in the last six books plays a part some what similar to that of Dido in i. and iv., but the idea of Amata as understudy to Dido will bear more emphasis. Vergil's treatment of Amata is distinctly unsympathetic, in great contrast to his portrait of Dido. But compare Mr. Garrod's remarks on the way Dido outgrew the original intention of her creator (English Literature and the Classics, pp. 150 sqq.

^{page 94 note 1} Thiasus is not used elsewhere in Vergil, except in Bucolics v. 30, 31: thiasos inducere Bacchi et foliis lentas intexere mollibus hastas.

^{page 95 note 1} Compare Georgics iv. 521, ‘nocturnique orgia Bacchi.’