Quin ipsae stupuere domus atque intima Leti 481.] 'Ipsae' not only the patients, but the agents, the prisons and torturers themselves. 'Intima Tartara' is rightly made by Wagn. epexegetic of 'domus,' like "urbem et promissa Lavini Moenia,' A. 1. 258, both being constructed with 'Leti.' Letum:' personified as in A. 6. 277, where it appears as one of the figures at the gate of Orcus; here it seems to be the presiding genius of the whole place. 'Intima,' the depths. 482.] Caeruleos' of the dark livid colour of the serpent, not unlike 'ater.' It recurs A. 7. 346, "caeruleis unum de crinibus anguem," which gives some slight support to caeruleis' in this passage, the first reading of Med. For 'inplexae' Med. a m. s. and a few others have 'inpexae,' which occurs in a parallel place Tibull. 1. 3. 69, Rom. and fr. Vat. innexae' (comp. A. 6. 281), while others of less authority give amplexae.' Wagn. cites Hor. Epod. 5. 15, "Canidia brevibus inplicata viperis Crines et incomptum caput," where however incomptum' might be used to confirm inpexae.' Capillus horrore inplexus atque impeditus" is quoted by Forc. from Appul. Apol. The sense here seems to be that the Furies had snakes twisted among their hair, i. e. growing from their heads and matted or entwining themselves with the natural hair. 483.] Inhians,' a-gape on Orpheus. 'Tenuit ora' may include both abstinence from barking and fixedness of countenance. See on A. 2. 1. 484.]Rota orbis' is difficult, as we should rather have expected' orbis rotae.' We may either make orbis' a genitive of quality, as we might say in prose 'a wheel of circular form,' or taking orbis' for the wheel, suppose after Heyne that 'rota' is put for the rotation-a sense of course not inherent in the word, which would then be used improperly, and so not needing to be supported by explicit instances, such as those which Voss adduces, and Forb. controverts. Comp. E. 9. 58 (note), "ventosi ... murmuris aurae," where the difficulty 485 is somewhat similar. 485-503.] He was returning, followed by his wife, and just on the point of emerging from the shades, when in a moment of forgetfulness he broke the condition imposed, and looked back upon her. She fled, complaining loudly of his madness and her fate, and he was not allowed to return to seek her.' 485.] Virgil simply indicates the giving of the consent by the epithet 'reddita,' and only mentions the condition parenthetically as an afterthought. This mode of telling the story was doubtless adopted on grounds of art, such as those which Horace (A. P. 43, 44, 136 foll.) applies to the larger question of the conduct of the plot of an epic; and it is so far successful that it keeps the mind fixed on Orpheus as the central figure, while it does not perplex those who already know the legend in its details. When he came to the composition of the Aeneid, he seems to have seen the necessity of being more explicit, though even there his narrative is sufficiently different from the naive garrulity of Homer. Ovid, whose mode of narration is more rapid, tells the whole story from first to last (M. 10. 1 foll.). 487.] Legem,' condition, A. 11. 322. So 'leges' and 'foedera' are coupled G. 1. 60. Again we are left to collect from the context that Orpheus was specially ordered not to look back. The injunction, as Cerda remarks, seems to be one of the same kind as that mentioned E. 8. 102 (note). 66 490 Ignoscenda quidem, scirent si ignoscere Manes : 490.] From a gloss in Dresd. Serv. Wagn. infers (rightly, as appears from Serv. on v. 498) that a punctuation was once current, connecting 'iam' with suam,' quae paene sibi iam erat reddita ;" and this he would approve but for the injury to the rhythm. But all that could be gained from it may be extracted from the passage as it stands, where' suam' is meant to be emphatic, he looked back on his recovered Eurydice, just as daylight was actually upon them.' 491.] Victus animi,' like 'animi dubius,' 3. 289, a construction common in Virgil with a participle or adjective, while other writers employ it with a verb, as Lucr. 1. 922, "nec me animi fallit." The genitive seems to mean 'with respect to,' though it is possible that it may originally have been local. See Madv. § 296 b. obs. 3, where the usage seems needlessly restricted to expressions which denote doubt and anxiety.' 'Victus' apparently means not master of himself.' 492.] Effusus labor' is like "incassum fusos... labores " A. 7. 421. In both passages, or at any rate in the latter, Virgil may have been thinking of Lucr. 2. 1165, "in cassum magnum cecidisse labores," where any attempt to alter 'magnum' or separate it from cassum' only robs the passage of its force, destroying the image of toil falling into a vast bottomless void. Not unlike is 'effudit curas,' Juv. 10. 78, though that is said of voluntary abandonment of exertion. 'Tyrannus' occurs several times in Virgil, in some passages (e. g. A. 7. 266) evidently without any invidious connotation, while there is perhaps none where such a meaning is absolutely required. As however the invidious sense was current when Virgil wrote (see the passages from Cic. referred to by Forc.), it seems natural to introduce it wherever, as here, the passage would be improved by it. 'Inmitis' seems to imply that the condition was a cruel one, and that Pluto will not relent even thus far a second time. 495 493.] Foedera :' see note on v. 487. The best commentary on 'terque fragor stagnis auditus Avernis' is Martyn's citation of Milton, Par. Lost. 9. 782: “Earth felt the wound, and Nature from her seat Sighing through all her works gave signs of woe, That all was lost," and, again, ib. 1000, "Earth trembled from her entrails, as again In pangs, and Nature gave a second groan : Sky lowered, and muttering thunder, some sad drops Wept at completing of the mortal sin Original." Serv. has a curious notion that the sound was one of joy among the shades, and quotes a passage from Lucan's lost Orpheus, gaudent a luce relictam [Heyne conjectures 'reductam' or 6 revectam,' but 'a luce relictam ' may = ' luce carentem '] Eurydicen, iterum sperantes Orphea, Manes." Voss's opinion that the sound is occasioned by the force exerted to bring Eurydice back would surely spoil the poetry of the passage. 'Avernis,' adj., A. 6. 118. Fragm. Vat. and other copies have 'Averni.' 66 494.] Here as well as in the next line, 'quis' goes with 'tantus furor.' With the expression comp. A. 4. 682, "Exstinxti te meque, soror." "Ovid, M. 10. 60, denies that Eurydice made any complaint, almost as if he intended to reflect on Virgil, as Euripides sometimes reflects on Aeschylus, "Iamque iterum moriens non est de coniuge quicquam Questa suo : quid enim nisi se quereretur amatam ? It is 495.] Furor' is the 'dementia' of v. 488. We need not take 'iterum' in the sense of 'rursus,' as Forb. thinks. true that the Fates were not calling Eurydice a second time 'retro,' but they were calling her a second time, and there is nothing strange in supposing Virgil to have combined the two forms of expression, 'vocant retro' and' vocant iterum.' 496.] Natantia lumina,' A. 5. 856. 497.] Ingenti circumdata nocte,' a contrast to the light into which they were just emerging, v. 490, as in non tua we have Invalidasque tibi tendens, heu non tua, palmas! 498.] 'Invalidas palmas: "in umbrae tenuitatem reductas" Serv., the Homeric áμεvηvós. With 'tendens palmas' comp. A. 6. 314, "Tendebantque manus ripae ulterioris amore." 499.] "Tenuis fugit, ceu fumus, in auras" occurs A. 5. 740. The comparison is from Π. 23. 100, ψυχὴ δὲ κατὰ χθονός, ήΰτε καπνός, "Ωιχετο τετριγυῖα. 500.] Wakefield's doubt whether 'tenuis' ought not to go with 'fumus' will hardly be entertained by any one now. 'Tenuis' is not an idle epithet, as it marks that quality in the air which makes the disembodied spirit combine with it. Fugit diversa' like 'quo diversus abis?' A. 5. 166. Med. a m. s. has 'fugit in diversa.' 501.] Umbras' may possibly be the shade of Eurydice, as Forb. thinks, as the pl. seems to recur A. 5. 81 of a single soul (comp. the use of Manes'), while the singular would naturally be avoided on account of 'prensantem;' but it seems better to understand it of the darkness which Orpheus clutches in the hope of embracing his wife. Multa volentem dicere,' A. 4.390. 502.] Praeterea,' A. 1. 49. 'Portitor' of Charon, A. 6. 326. 503.] 'Obiectam' like obiecta flumina' 3. 253. 'Palus' here seems to be Styx. The object of " 'passus must be Orpheus, who, as Keightley says, doubtless attempted to cross the river again. Serv. says of this passage "mysticum est: dicitur enim bis eandem umbram evocari non licere." 500 505 the shades, though from various passages in Here 506.] This verse, like 3. 219, has been thought out of place, when it really adds much to the force and beauty of the passage, serving at once to complete the picture of hopelessness as presented to Orpheus' mind and to balance her fate with his, which is described in the subsequent lines. 'What should he do? even while these thoughts are passing through his mind, she is on her way back over the Styx; and so she doubtless wanders as before on the shores beyond, while he' &c. We may conceive him (see on v. 503) as returning to the bank and being repelled by Charon, who will not admit him, or put back for him, but hurries over the river with his single passenger. The objection that 'illa' is followed not by 'hunc' but by 'illum' may be met if we consider that the contrast is not meant to be so much formally expressed as suggested, her subsequent fate being left to be inferred from her being seen floating over the water. 'Iam' seems to go with 'frigida;' all the warmth of life by this time had left her, and she was a ghost again. Possibly the word may be illustrated by the reason given by Lucian (De Luctu, c. 11) for putting a robe on the dead body, viz. that might not take cold while crossing the Styx. 'Nare' of sailing on board ship seems rare. Forc. quotes Catull. 64 (66). 45, "iuventus Per medium classi barbara navit Athon." Cymba' of Charon's boat, A. 6. 303, 413, Septem illum totos perhibent ex ordine menses Flet noctem, ramoque sedens miserabile carmen 507-527.] He wandered about in wintry solitudes, lamenting his fate like the bereaved nightingale in strains that drew savage beasts and rocks after him, and never admitting the thought of another love a slight resented by the Thracian women, who in one of their Bacchanalian orgies tore him in pieces. As his head floated down the Hebrus, it was heard still to repeat the name of his lost wife.' 508.] Rupe sub aeria,' a picture like E. 10. 14. Comp. ib. 52. 509.] Rom. reads 'flesse sibi,' a rather remarkable variation, as being less usual than 'flevisse.' The same authority with some others gives 'gelidis sub astris,' which is exceedingly plausible. Perhaps however it might be said that with the epithet gelida''astra' could hardly be understood except of the night, and that this would not agree with the effect of the song on wild beasts, who can hardly be supposed to have been kept from their dens to listen. Poets are placed in caves elsewhere, Prop. 4. 1. 5, "Dicite, quo pariter carmen tenuastis in antro," and possibly Hor. 1 Od. 32. 1, "Si quid vacui sub umbra Lusimus," where Bentley from one MS. read 'sub antro.' Thus gelidis' would have force here as reminding us that caves are not merely places of nestling green for poets made,' but have their dreary and uninviting side, which was here the attraction to Orpheus. 'Evolvisse,' recounted his sufferings in order, a metaphor either from spinning or from turning over a book. 510.] The existence of tigers in Thrace is of course a fanciful or mistaken notion. Keightley reminds us that Shakspeare talks of a lioness in the forest of Ardennes. 511.] The celebrated simile which follows is compounded from Od. 19. 518 foll. and ib. 16. 216 foll., the former of which de 510 515 scribes the nightingale singing as if in lamentation for her lost Itylus, while the latter speaks of vultures screaming for the real loss of their young. Germ. finds a possible allusion to the fact, mentioned by Pausanias, that the nightingales near the tomb of Orpheus were more vocal than others of their kind. The quivering motion of the poplar leaves may be intended, as Heyne thinks, to be in keeping with the protracted melancholy singing. 512.] οἶσί τε τέκνα ̓Αγρόται ἐξείλοντο Tápoç TεTEηvà yevéolai, Od. 15. 217. 'Observans' is used loosely, to supply the want of an aor. part., the sense being 'observatos detraxit.' With the fact compare E. 3. 68, which may be said to give the other side, the countryman's view of his action. Arator,' 2. 207, where however the word is used more strictly, as it is for ploughing that the countryman clears the land of trees, birds' nests and all. 513.] δενδρέων ἐν πετάλοισι καθεζομévη πvкivotoi, Od. 19. 520. 515.] Integrat,'' renews,' or 'repeats,' the nightingale constantly recurring to the same notes. Hom. (Od. 1. c.) gives the contrary image, Te Oаμà тршπшσα xéει πολυηχέα φωνήν, thinking probably of the difference of the notes among themselves. 'Maestis. inplet,' perhaps from Lucr. 2. 146, "liquidis loca vocibus opplent," as Cerda suggests. ... 516.] Nulla Venus,' as Eur. (Iph. A. 1264) talks of 'Aopodirη ris for a certain passion.' 'Non ulli,' the reading of Med. and others, was restored by Heins. for the common text 'nullique.' Animum flexere may be illustrated by Catull. 62 (64). 330, "Quae tibi flexanimo mentem perfundat amore." The meaning then, as it would be expressed in prose, seems to be, 'no passion bowed his soul, so that he took on him the yoke of wedlock.' If we choose to press ' non ulli flexere hymenaei,' understanding it of the softening influence of Solus Hyperboreas glacies Tanaimque nivalem Dona querens; spretae Ciconum quo munere matres 520 marriage, we may comp. Lucr. 5. 1017, "puerique parentum Blanditiis facile ingenium fregere superbum," and the whole passage on domestic life of which it forms part. 517.] The places mentioned in this and the following line are doubtless intended by Virgil to be in or bordering on Thrace, as Heyne remarks, as it is not likely that Orpheus should be represented as wandering far north of his own country; so that we must again note the poet's loose handling of geography. Hyperboreas,' see on 3. 197. Trapp says of this and the next line, "Those verses are enough to make one shudder at Midsummer." 518.] 'Rhipaeis,' note on 1. 240. 'Viduata' is similarly used by Lucr. 5. 840, "Orba pedum partim, manuum viduata vicissim." It is possible that Virgil may have chosen 'viduata' with reference to Orpheus' condition, but the thought, even thus slightly hinted at, would be a mere conceit. 6 520.] Munus' is technically used of funeral honours (A. 4. 624., 6. 686., 11. 26, and various instances cited by Forc.), that being, according to one opinion (see Tertullian de Spectat. 12, apud Forc.), the sense which led to another technical application of the word, to games, shows, &c. It does not seem harsh to speak of Orpheus' constancy and suffering sorrow as a ' munus to Eurydice in this sense, especially as quo,' as it were, apologizes for the word with which it is joined, a tribute like this,' or, as we might say, 'this way of honouring his wife,' any more than in A. 4, 1. c., where the Tyrians are charged to be the implacable enemies of the Trojans, as a 'munus' to Dido's ashes. There would be considerable probability in the interpretation of Asper, mentioned by Philarg., ob quam rem, où χάριν (quo munere ='cuius [Orphei] munere '), if it could be supported by ex 525 amples; but though such expressions as vestro munere' (1. 7), munere Divum' (ib. 238), help us to see how the phrase might have arisen, they do not entitle us to assume its existence. 'Spretae munere then will mean slighted by the tribute,' i. e. feeling themselves slighted. Sperno' is specially used of scorned or rejected love, E. 3. 74, A. 1. 27. Thus we may see that spreto,' the reading of some MSS., is a mere correction by those who did not understand the passage. 'Matres' seems at first sight a strange word for the marriageable women of Thrace (Ov. M. 11. 3 has 'nurus Ciconum '), but it seems to be applied to them as Bacchanals, like θύουσαν "Αιδου μŋrép', Aesch. Ag. 1235. 521.] The story as told by Ov. 1. c. is that the Thracian women, while in the midst of their orgies, accidentally saw Orpheus, remembered his scorn, and so tore him in pieces. Some MSS. have 'nocturnaque,' which Pier. defends, supposing' que' to be unelided. 523.] The application of 'marmoreus to the body is as old as Lucilius (28. 47), "Hic corpus solidum invenies, hic stare papillas Pectore marmoreo," where however the reference seems to be to firmness of flesh rather than to colour. Caput a cervice revulsum' is from Enn. Ann. 462. 524.] Oeagrus was the father of Orpheus, so that Oeagrius' here = paternus.' 525.] Vox ipsa,' the mere voice, as if it were a separate organ, like the tongue. 'Frigida,' v. 506. 526.] Vocabat,' not that he invoked her in death, which the mode of the address contradicts, but that he went on lamenting her in death as in life. 527.] Toto flumine,' if pressed, seems to mean over the whole breadth of that part of the stream down which the head floated while it still retained its power of |