So familiar are we with the story of the Aeneid that we are apt to forget what violence it does to the tradition generally current in Virgil's time. That tradition is represented by the third Aeneid; there Aeneas is brought as far as Sicily, after a course of wandering corresponding fairly with that described by Livy and Dionysius. But in order to bring in the new element of the story, Aeneas must be carried to Carthage from Sicily before he can be allowed to go on to Latium. The fifth book, as it now stands, implies a second visit to Sicily after the tragedy of Carthage. It is difficult to suppose that so awkward a combination as this can have entered into the original plan of the Aeneid. As things now stand it might occur to the reader that the fifth Aeneid would naturally have followed the third, as the sixth might naturally have followed the fourth. Virgil had not, probably, at the time of his death, harmonized the Sicilian and Carthaginian episodes in a manner satisfactory to himself.

The way in which Virgil, for the purposes of his epic, has altered the story of Dido, is as striking and characteristic as anything in the whole range of his poetry. In the universally accepted tradition Dido's tragic end was due to her resolution not to become the wife of Iarbas; and what in Virgil is represented as coming upon her as a curse for the breach of her vow is, in the genuine story, the honourable result of her constancy. No doubt Virgil felt that Varro's version of the story, according to which not Dido, but her sister, was sacrificed for love of Aeneas, would have been tame and pointless in his epic poem; he therefore ventured on a bolder flight, and carried the day. No part of the Aeneid, if we may trust Ovid, was more eagerly read than the fourth book; and all readers were forced to acknowledge the skill with which he made their tears flow in a fictitious cause.

In comparing Virgil's account of the early fortunes of Dido with that of Pompeius Trogus (Justin 18. 4-6) the reader is struck with some minute coincidences of language which may show that both writers drew upon the same source, but that Virgil for the sake of brevity mutilated the narrative. Take the two accounts of Dido's flight from Tyre. Sychaeus, it will be remembered, is in Trogus' narrative called Acerbas.

Justin 18. 4. 8, qua (fama) incensus Pygmalion oblitus iuris humani avunculum suum eundemque generum sine respectu pietatis occidit. Dido then is Pygmalion's daughter, and great-niece of her husband. In Virgil Pygmalion is only the germanus of Dido, Aen. 1. 346, sed regna Tyri germanus habebat Pygmalion, scelere ante alios immanior Ille Sychaeum Impius ante aras atque auri caecus amore Clam ferro incautum superat, securus amorum Germanae. Justin l. c.

omnes.

'See Ausonius, Epigr. 118.

Elissa fugam molitur adsumptis quibusdam principibus in societatem, quibus par odium in regem esse eandemque fugae cupiditatem arbitrabatur. . . . Sed Elissa ministros migrationis a rege missos navibus cum omnibus opibus suis prima vespera imponit, provectaque in altum compellit eos onera harenae pro pecunia involucris involuta in mare deicere. Tunc deflens ipsa lugubrique voce Acerbam ciet . . . tunc ipsos ministros adgreditur; sibi quidem ait optatam olim mortem, sed illis acerbos cruciatus et dira supplicia imminere, qui Acerbae opes, quarum spe parricidium fecerat, avaritiae tyranni subtraxerint. Hoc metu omnibus iniecto comites fugae accepit.

This is a clear and intelligible narrative. Dido associates with herself some of the nobles who, as she thinks, hate Pygmalion as much as she does, and she further devises a means to work upon their fears. But Virgil abbreviates the narrative till it becomes difficult to understand: Conveniunt, quibus aut odium crudele tyranni, Aut metus Servius explains this passage, which evidently appeared to him difficult, by reference to a narrative perhaps not unlike that of Trogus; metuebant laedendi, hoc est, qui timebant ne laederentur; unde est illud in quarto (545) et quos Sidonia vix urbe revelli; quia non voluntate sed aut odio aut timore convenerat.

acer erat.

Then again Virgil's naves quae forte paratae is very vague. Servius explains it by reference to a narrative quite different to that of Trogus: moris enim erat ut de pecunia publica Phoenices misso a rege auro de peregrinis frumenta conveherent. Dido autem a Pygmalione ad hunc usum paratas naves abstulerat; quam cum fugientem a fratre missi sequerentur, aurum illa praecipitavit in mare, qua re visa sequentes reversi sunt. Licet et alio ordine historia ista narratur.

The fragment of Timaeus (23 Müller) in which these events are narrated gives an account which compared with that of Justin is an abridgment. τοῦ γὰρ ἀνδρὸς αὐτῆς ὑπὸ Πυγμαλίωνος ἀναιρεθέντος, ἐνθεμένη τὰ χρήματα εἰς σκάφος, μετὰ τίνων πολιτῶν ἔφευγε, καὶ πολλὰ κακοπαθήσασα τῇ Λιβύῃ προσηνέχθη, καὶ διὰ τὴν πολλὴν αὐτῆς πλάνην Δειδὼ προσηγορεύθη ἐπιχωρίοις.

The fourth Aeneid, however much it may differ from the received tradition, contains a few touches for which Virgil may perhaps be indebted to it. Justin 18. 6 gives the following account of Elissa's death. Diu Acerbae viri nomine cum multis lacrimis et lamentatione flebili vocato ad postremum ituram se quo suae urbis fata vocarent respondit. In hoc trium mensium sumpto spatio, pyra in ultima parte urbis instructa, velut placatura viri manes inferiasque ante nuptias missura, multas hostias caedit et sumpto gladio pyram conscendit, atque ita ad populum respiciens ituram se ad virum, sicut praeceperint, dixit, vitamque gladio finivit. Timaeus 1. c. τοῦ τῶν Λιβύων βασιλέως

θέλοντος αὐτὴν γῆμαι, αὐτὴ μὲν ἀντέλεγεν, ὑπὸ δὲ τῶν πολιτῶν συναναγ καζομένη, σκηψαμένη τελετὴν πρὸς ἀνάλυσιν ὅρκων ἐπιτελέσειν, πυρὰν μεγίστην ἐγγὺς τοῦ οἴκου κατασκευάσασα καὶ ἅψασα, ἀπὸ τοῦ δώματος αὑτὴν eis tǹv πupáv čppuyev. The vow of constancy, the pyre and the sword, the excuse for raising the pyre, are adopted by Virgil. It may again be observed that Timaeus' account is the shorter, and also that it differs from that of Trogus as to the manner of Elissa's death.

As for the fortunes of Aeneas after his landing in Latium, there were two main traditions, one of which represented Aeneas as obtaining the hand of Lavinia only after war with her father Latinus, the other that there was no fighting with Latinus at all, but that war arose after his death in consequence of the claims of Turnus to the hand of Lavinia.

The first, which is the basis of the version adopted and modified by Virgil, is alluded to by Servius on Aen. 4. 620, who quotes Cato as his authority. A quarrel breaks out between Aeneas and Latinus in consequence of plundering on the part of Aeneas' companions; in the battle which ensues Latinus is slain. So Servius on Aen. 9. 745, si veritatem historiae requiris, primo proelio interemptus est Latinus. See also Serv. Aen. 1. 259. Livy 1. 1 gives a slightly different account alii proelio victum Latinum pacem cum Aenea, deinde adfinitatem iunxisse tradunt. The other tradition is given by Livy in the following terms: alii (tradunt) cum instructae acies constitissent, priusquam signa canerent, processisse Latinum inter primores, ducemque advenarum evocasse ad conloquium: percontatum deinde qui mortales essent, unde aut quo casu profecti domo, quidve quaerentes in agrum Laurentem exissent, postquam audierit multitudinem Troianos esse, ducem Aeneam, filium Anchisae et Veneris, cremata patria et domo profugos sedem condendaeque urbi locum quaerere, nobilitatem admiratum gentis virique, et animum vel bello vel paci paratum, dextera data fidem futurae amicitiae sanxisse. There is a general resemblance between this description and Virgil's words in the seventh Aeneid (229): Dis sedem exiguam patriis litusque rogamus Innocuum, et cunctis undamque auramque patentem. ... Fata per Aeneae iuro dextramque potentem, Sive fide, seu quis bello est expertus et armis. It may be that Virgil, though varying the tradition for his own purposes, is working upon the same materials as Livy.

The account given by Dionysius represents Latinus as at war with the Rutuli when Aeneas landed. Latinus is forbidden by oracles to fight with the stranger, and advised rather to ally himself with the "Enves. Aeneas advances his claims, and receives from Latinus an assurance which recalls Dido's Non ignara mali miseris succurrere disco. The Trojan hero marries Lavinia; the Aborigines and Rutuli receive the name of Latini, but afterwards the Rutuli, under the

leadership of Turnus, who is branded as an avтóμolos, desert the alliance. Turnus fights for his lost love, and both he and Latinus die in the battle.

The account given by Dionysius tallies on the whole with that attributed by Servius, Aen. 6. 760, to Cato. The Etruscan element in the story, represented by Mezentius, is treated by Virgil quite in a way of his own. For, however they may differ in details, the tradition as given both by Cato and by the authorities whom Dionysius follows represents Mezentius as falling in a war which arose some time after the death of Turnus. Mezentius is indeed an ally of Turnus, but is not killed until after the final settlement of Aeneas in his kingdom; according to Cato, it was by Ascanius, according to Dionysius' authorities, by Aeneas himself, three years after the battle in which Turnus and Latinus were slain. As in the case of Dido, Virgil does violence to the accepted order of events. Turnus must be slain before Aeneas can finally obtain the hand of Lavinia; thus the last half of the Aeneid is provided with its element of romance; and Mezentius falls before Turnus in a war in which both are simultaneously engaged.

It is evident that Virgil had a tradition or traditions to work upon, many of the details of which are now lost, but which are most fully preserved by Dionysius. Fragments of them are preserved by Servius on Aen. 7. 51, Amata. . . duos filios voluntate patris Aeneae spondentes sororem factione interemit . . . Hos alii caecatos a matre tradunt, postquam amisso Turno Lavinia Aeneae iuncta est. Does this imply that there was, independently of the Aeneid, a story according to which Turnus died before the marriage of Aeneas with Lavinia? In any case it implies that Amata survived Turnus, and this is different from the account in the Aeneid. Another instance is mentioned by Asper, quoted in the Verona scholia on Aen. 7. 484, Tyrrhum aiunt fuisse pastorem aput quem Lavinia delituit tum cum Ascanium timens fugit in silvas. Hic Latini vilicus dicitur fuisse. Comp. Serv. ad 1.

The considerations on which I have been dwelling will be found, I think, to throw some light on the difficulties with which Virgil had to contend. The traditions on which alone he could work had neither form nor life. Aeneas had never, so far as we can see, not even in the Homeric poems, been a hero in the sense in which the word can be used of Achilles, Ulysses, Ajax, or Diomed. Even in Homer the protection of Aphrodite and Apollo hangs heavily around him. In the place where he is worshipped he is a mere name; a shadowy demi-god associated with the worship of Aphrodite. As a founder of cities he has no characteristic to distinguish him from the many fabulous oikioraί of Greek and Italian towns. The Homeric

heroes do not found cities, but destroy them; the civilizing and beneficent hero, on whose features Dionysius dwells with pleasure, is the creature, if not of philosophy, at least of a late and reflective stage of mythology. To make out of so shadowy a being as the Aeneas of legend a hero of war and peace, fit to be the founder of an imperial city, was no easy task, especially for a poet who considered it his first duty to construct his epic in words, manner, and arrangement, on the model of the Iliad and Odyssey.-[H. N.]