musque Potitius" above v. 281. "Asper inmani corpore" Lucr. 5. 33, of the serpent in the garden of the Hesperides. Serv. collects different notices of this Thybris, the one most germane to Virg.'s description representing him as a robberchief, and connecting his name with ὕβρις, a word, as others have remarked, associated with violent floods by Aesch. Prom. 717, Hdt. 1. 189. Livy 1.3 makes Tiberinus a king of Alba.

331.] A quo cognomine' may = "a cuius cognomine:" comp. 2. 171 &c.: or 'cognomine' may be adj., as in 6. 383. A third way would be to separate 'quo' from 'cognomine,' taking the latter with 'diximus,' by way of surname, as in 12.845 &c. Evander, as Serv. remarks, identifies himself with the Italians, mentioning the name incidentally, a proof of the vagueness of Virg.'s historical notices. The general story seems to have been that Thybris or Tiberinus was drowned in the Albula: the version however which made him a robber-chief speaks of him simply as having lived on its banks.

333.] The cause of Evander's exile was variously given, some ascribing it to parricide or matricide: Lewis, p. 284. Virg. perhaps means to negative these stories, as Ovid does: see the next note. Pelagi extrema sequentem,' as Heyne remarks, is said in the character of an ancient Greek speaking of the unknown west. Donatus made 'pelagi' locative, taking 'extrema sequentem of encountering dangers, not unlike "ferro extrema secutam" 6. 457. 'Sequi' is similarly used 10. 193., 12. 893, seeking a distant though unmoving object being regarded as tantamount to pursuing a flying one. Comp. 4. 361 note.

334.] Serv. notes that Fortune and Fate are not philosophically consistent. The inconsistency is kept up by the epithets, though they are apparently similar, 'omnipotens' referring Evander's landing to the all disposing power of chance, 'ineluctabile' to the destiny of his birth, which he

VOL. III.

335

could not escape. "Ineluctabile tempus" 2. 324, "inexorabile fatum" G. 2. 491. Ov. F. 1. 481 seems to refer to this passage when he makes Evander's mother say "Sic erat in fatis: nec te tua culpa fugavit, Sed deus," though he is speaking of the cause of Evander's leaving home, not of the cause of his reaching Italy. But Virg. may include both: see the next line.

335.] Egere' probably refers to the entire voyage, 'drove me to leave my home and settle here.' Comp. "acti fatis" 1.33. The 'tremenda monita' of Carmentis are like the "iussa ingentia" of Apollo 7. 241, which is generally parallel. Rom. has 'tremendae.'

336.] For Carmentis or Carmenta, and for the other accounts of Evander's parentage, see Lewis 1. c. 'Auctor Apollo' 12. 405, in a different sense: with 'auctor' here comp. "fatis auctoribus" 10. 67.

337-368.] Evander shows Aeneas the various places which afterwards became famous as parts of Rome, the Carmental gate, the Asylum, the Capitol, and the Forum. He welcomes him to his homely palace, and puts him to rest for the night.'

337.] Vix ea dicta" "sunt." 'Dehinc' dissyll., G. 3. 167. The altar of Carmentis was shown in the time of Dionys. Hal. (1.32) close to the Carmental gate. Med. (originally) and Rom. have 'arma.'

338.] Wagn. restored 'Romani' from Med., Pal., Gud. &c. Rom. and two of Ribbeck's cursives have 'Romano,' the old reading, which may have arisen, as Wagn. thinks, from the two first letters of 'nomine.' The old editions used to point after 'portam,' constructing 'quam memorant' with 'honorem.' With the present pointing either Romani' or 'Romano' gives good sense. If the latter seems the more poetical, we may urge that the name was not strictly Roman, the very object of the context being to show that it came from Carmentis. The Carmental

I

340

gate was otherwise called the "porta scele rata," being that through which the Fabii passed.

339.] In a Greek author 'honorem' would be interpreted as cogn. acc. after 'memorant:' in Latin it is simpler to take it in apposition to 'portam:' see however on G. 3. 41. In either case it is the poetical equivalent of the prosaic expression "in honorem."

340.] Prima' is explained by Serv. with reference to the later prophecy of the Sibyl. Comp. Livy 1.7 (of Carmenta) "quam fatiloquam ante Sibyllae in Italiam adventum miratae eae gentes fuerant." It is a strange instance of Virg.'s habit of introducing things incidentally, if indeed we are not rather to call it a proof that he had not thoroughly digested the materials of his story, as we should have expected that more stress would be laid on a prediction like this. 'Futuros' not to be taken with 'magnos' and 'nobile.' The two things which Carmentis predicted as in the future were the mighty family of Aeneas and the glorious Pallanteum.

341.] Aeneadae' includes the Romans, Lucr. 1. 1; indeed they must have been the chief burden of the prophecy, as the connexion of Troy with Pallanteum really began with the foundation of Rome. 'Nobile Pallanteum' probably refers not only to the glories of the place under Evander and his successors, but to those of the Palatine in more historical times. Rom. has 'nomine,' and 'nobine' is the reading of Pal. and (originally) Gud.

342.] Quem Asylum rettulit' has not been satisfactorily explained. The general sense is doubtless that given by Donatus, "qui postea a Romulo Asyli nomen accepit," but it does not appear how this is to be got out of the words. Serv. thinks the reference is to the Athenian Asylum, apparently taking 'rettulit' to mean produced by imitation. Heyne and Wagn. explain it "appellavit." Gossrau interprets it "restituit," comp. 5. 598. The choice seems to lie between the two last

345

views, or some modification of them. No authority is quoted for "referre" in the sense of "appellare:" but we may perhaps comp. "renuntiare aliquem consulem." If we take Gossrau's view, we may suppose the meaning to be not so much revived it as an asylum as changed it into an asylum, comparing the use of "reddere." This is perhaps supported by Florus 1. 1, "Erat in proxumo lucus: hune asylum facit." For the site of the asylum comp. Livy 1. 8, "locum, qui nunc saeptus descendentibus inter duos lucos est, asylum aperit," and see Lewis, p. 419, and the authors there referred to.

343.] The Lupercal was a cavern in the Palatine, connected by some of the ancients with the wolf that suckled Romulus and Remus, by others, as by Virg. here, with Evander and the Arcadian worship of Pan. See Lewis, pp. 238, 384. For the Lupercalia see Dict. A. s. v. 'Sub rupe:' comp. v. 295 above, E. 10. 14, 15, the latter of which describes an Arcadian mountain scene.

344.] Called after the Parrhasian (Arcadian) custom the place of Lycaean Pan,' i. e. dedicated to Pan, the god of Arcadia, and called by his Lycaean name, Lupercal' being supposed to be connected with "lupus" as 'Lycaeus' with λύκος. 'Panos' is the possessive gen., and 'dictus' seems to include the two notions of naming and dedicating, for which see on 6. 138. Schrader conj. 'monte,' which is actually found in two inferior MSS., and supported by Ov. F. 2. 421, "Quid vetat Arcadio dictos de monte Lupercos? Faunus in Arcadia templa Lycaeus habet." Panos,' the Greek gen., seems to be found in all the MSS. Parrhasius' is applied to Evander 11. 31, the name of the town Parrhasia being put for the whole of Arcadia.

345.] The precise site of Argiletum is disputed. Cic. Att. 12. 32 mentions it as a place where he owned some shops, and there are similar allusions to it in Mart. 1. 3.1 &c. The name as usual was accounted for by various contradictory legends, some

350

355

making Argus the son of a haruspex, killed by his father for disclosing the meaning of the human head found at the Capitol, others talking of a Roman Argillus, who was put to death in the time of the first or second Punic war, while another etymology derived the word from "argilla." Even those who made Argus the guest of Evander represented his death differently, though they agreed in the fact that he was killed for conspiring against his host. See Serv., and comp. Varro L. L. 5. § 157. 346.] Testaturque locum' calls the spot to witness what happened there, perhaps including, as Serv. thinks, a protestation of his own innocence as a host. Elsewhere the spot where a thing happened is said 'testari' what happened there, as in Prop. 4. 7. 21, "Sunt Agamemnonias testantia litora curas." But it may mean, as Mr. Long suggests, bears witness to the spot. 'Docet' explains, 6. 891.

347.] Tarpeius mons' was the old name of the Capitoline mount, Varro L. L. 5. § 41. 'Sedem' apparently refers to the temple of Jupiter, not as belonging to Evander's time, but as built afterwards.

348.] 'Aurea: Pliny 33. 3 says even the bronze tiles of the Capitol were gilded at the restoration of the edifice by Catulus. 349.] "Religio fani," "sacrarii," "signi" are found in Cic.: see Freund.

350.] Rom. has 'silvas.' Med. first reading has 'tenebant,' Gud. originally 'tremebat.'

351.] Gud. has a variant 'hoc...saxum.' 353.] Cum saepe' 1. 148, note. 'Nigrantem' from its connexion with the storm, ἐρεμνὴν αἰγίδα 11. 4. 167.

with clouds, and thunder and lightning follow. Thus aἰγís elsewhere is simply a name for the whirlwind. Comp. 2. 616 note. In Hom. it seems to be a goatskin used as a shield belt, and sometimes the shield itself (Dict. A. s. v.): Virg. apparently follows later writers in regarding it as a breastplate: comp. v. 437. Heyne, supposing it to be a shield, was puzzled to understand how it could be held in the right hand, and so punctuated before 'dextra' (which is also the pointing of Serv.), thus introducing a collocation of 'que' unknown to Virg. except under peculiar circumstances. Wagn. justly observes that whether shield or not, it is clearly not used for purposes of defence, so that there can be no reason why it should not be held in the right hand. 'Nimbos cieret' is from the Homeric νεφεληγερέτα.

355.] The account given by Macrob. Sat. 1. 7 is that Janus was established as king of Italy in a city called Janiculum, when Saturn came to the country, after which they reigned jointly, Saturn building a town which was called Saturnia. Varro L. L. 5. § 42 speaks of Saturnia and its supposed remains, Ovid F. 1. 241 foll. of Janiculum. "Disiectas moles" 2.608.

356.] "Veterum monumenta virorum " 3. 102. Here 'veterum virorum' goes with 'reliquias' as well as with 'monumenta.' Serv. rather gratuitously remarks on 'virorum "hoc sermone ostendit etiam Saturnum virum fuisse."

357.] Janus has already been associated with Saturn 7. 180. Pater, the Latin title of a God (see on G. 2. 4), is constantly connected with Janus, Hor. 2 S. 6. 20, 1 Ep. 16.59. 'Arcem' Pal., Med. (first reading), 'urbem' Rom., Med. (second reading). The words are constantly confounded, and

354.] Zeus is represented as shaking his aegis, Il. 4. 167., 17. 593. In the latter passage the effect is that Ida is covered

360

the former is more appropriate to a mountain settlement.

358.] Huic,' 'illi' are rather carelessly introduced after 'hanc, hanc.' Forb. rightly remarks that 'huic' is applied to Janiculum as being in thought nearer the speaker and consequently first named in the preceding verse. See Madv. § 485 a. Fuerat' again comes in somewhat loosely after 'condidit,' referring to the same time. See Madv. § 338. obs. 6.

359.] Dictis' may be a participle, but on a comparison of 7. 249, 284, it is perhaps better to take it as a substantive, the abl. being one of circumstance. Serv. mentions the doubt. Ad tecta subibant' approached the house; without 'ad' it would have been entered: comp. vv. 362, 3. And so Donatus.

360.] Passim' dispersedly. "Laeta boum passim campis armenta videmus" 3. 220. Mugire videbant:' see on 4. 490. 361.] For the site of the 'Carinae,' which is more or less disputed, see Dict. G. vol. 2, pp. 822, 823. Pompey had a house there, which afterwards became M. Antony's. Rom. has 'Cavernis.' For 'lautis' Med. has 'latis.' Dryden renders the line Once oxen lowed where now the lawyers bawl.'

362.] For 'victor' one of Ribbeck's cursives reads 'nobis,' with 'victor' as a variant. 363.] Peerlkamp may be right in his interpretation of 'subiit' stooped to enter, comparing Ov. M. 5. 282 “subiere minores Saepe casas superi" (add Id. F. 4. 516., 5. 505); see however Id. M. 1. 121. The lengthening of the last syllable is sufficiently accounted for by the caesura, especially before the aspirate, without supposing with Lachm. (see Excursus on G. 2. 81, second edition) that it is really long in Virg.

365

Rom. and Med. (first reading) have 'subit.' 'Cepit' need merely be i. q. "accepit:" but there is force in Serv.'s remark "mire dictum ut alibi, (9.644) ‘nec te Troia capit."

364.] 'Aude' of making a moral effort, like "sapere aude" Hor. 1 Ep. 2. 40. So Aesch. Prom. 999, τόλμησόν ποτε Πρὸς τὰς παρούσας πημονὰς ὀρθῶς φρονεῖν.

365.] Finge' like 'aude' seems to express effort: comp. 6.80, G. 2. 407. "Nec, si miserum Fortuna Sinonem Finxit, vanum etiam mendacemqueinproba finget" 2.80. Rebus egenis' seems to be constructed with both 'asper' and 'veni.' 'Rebus egenis' 6. 81., 10. 367. "Rebus in dubiis, egenis," Pl. Capt. 2. 3. 46. Dryden says of this and the foregoing line (Dedication to Aeneid) "For my part, I am lost in the admiration of it: I contemn the world when I think of it, and myself when I translate it."

366.] Fastigia tecti,' the sloping roof: see on 2. 302.

367.] "Ingentem Aenean" 6. 413, where there is a similar contrast.

368.] Schrader ingeniously conj. 'spoliis,' which is the reading of one MS., the third Gothan. Forb. remarks that the couch was of leaves, with a bearskin over it. "Pelle Libystidis ursae" 5.37 note. Virg. seems to have imitated Od. 14. 48 foll., as Heyne remarks.

369-406.] That night Venus entreats Vulcan to make a suit of armour for Aeneas, reminding him that she had asked no favour while the Trojan war lasted. He chides her for her hesitation, and readily consents.'

369.] "Nox ruit" 6. 539. The conception of night as winged is found Eur. Or. 177, Aristoph. Birds 695. We have already had a hint of this image 2. 360., 6. 866.