For Daedalus and the stories connected with him see Dict. Myth. The bulk of tradition seems to point to Sicily as the place where he took refuge after leaving Crete: but Sardinia was also mentioned as a spot to which he went. Italy as well as the adjoining islands would naturally associate his name with its works of art: and so Sil. 12. 102 makes him the builder of a temple of Apollo at Capua, under circumstances similar to those in the text—one of Silius' many imitations of Virg. ‘Regna’ probably includes the government as well as the kingdom. At any rate ‘Minois’ is significant, as it was on Minos' account that Daedalus fled from Crete.
 Virg. might have spoken of flying as either trusting to wings or trusting to the sky. Here he has chosen the latter, ‘pennis’ being the instrumental abl. This is better than to make ‘pennis’ dat., ‘caelo’ abl., whether ‘caelo’ be connected in that case with what precedes, or, as Heyne suggested and Wakef. punctuates, with what follows. “Credunt caelo” G. 4. 192 is different: see note there. ‘Praepetibus’ here merely means ‘swift,’ and has no augurial reference.
 We have already had ‘nare’ and one of its compounds used of flying, 4. 245, G. 4. 59. But Virg. may have been thinking of Lucr. 3.591, “Quam prolapsa foras enaret in aeris auras,” of the soul quitting the body. ‘Gelidas ad Arctos’ has perplexed the commentators: but Wagn. after Hand. Turs. 1, p. 82, seems right in explaining it as meaning no more than that Daedalus flew northward, which would be the case whether we think of his rising from the ground, or of the position of Cumae as north of Crete.
 Marius Plotius De Metris quotes ‘Chalcidicas—arces,’ and Med. exhibits traces of a reading ‘arcem,’ which Heins. prefers. ‘Chalcidica:’ see above on v. 2. ‘Levis’ of easy motion, 5. 819, = ‘volans.’ ‘Arce:’ “the ancient citadel or arx (still called the Rocca di Cuma), an isolated and precipitous rock, very difficult of access, and on that account regarded as a very strong fortress:” Dict. G. ‘Cumae.’ ‘Adstitit’ 1. 301 note.
 Redditus &c. gives the reason of what follows. This being the place where he alighted, he paid a thank-offering to Apollo here. One MS. gives ‘hic,’ which Burm. prefers and Heyne approves: but Wagn. rightly remarks that ‘his’ is more poetical, as it includes ‘hic.’ Comp. 1. 534 note. With ‘primum’ Wagn. comp. 3. 209, “Servatum ex undis Strophadum me litora primum Accipiunt.”
 Daedalus hangs up his wings, as a mariner rescued from shipwreck hangs up his garments, or a soldier the arms which he has used for the last time. ‘Remigium alarum’ 1. 301 note. Cerda is doubtless right in regarding the temple also as a votive offering. ‘Posuit templa’ G. 3. 13.
 For sculptures on the door of a temple comp. G. 3. 26 note. ‘Letum’ (erat): for Androgeus and the different accounts of his death see Dict. Myth. s. v. For the spelling ‘Androgeo’ or ‘Androgei’ see on 2. 371. Here the majority of MSS. (Med., Pal., Rom., &c.) is for the Latin genitive, ‘Androgeo’ being only found in later copies: but the grammarians are (Serv., Charisius, Priscian, Probus) for the Greek form here, and I have followed Wagn. in restoring it, though with considerable hesitation. ‘Tum’ indicates that the Athenians sending their children to death was a second subject represented. How it was represented may be gathered from v. 22, “stat ductis sortibus urna.” With ‘pendere poenas’ comp. Catull. 62 (64). 173, “Indomito nec dira ferens stipendia tauro,” of the Minotaur.
 Miserum interjectional, like ‘infandum,’ ‘nefas,’ &c. Heinsius' latest notion that it could stand for ‘miserorum’ is contrary to Virg.'s usage: see on 3. 704. ‘Septena:’ the story mentioned seven youths and seven maidens: but Virg. has chosen only to name the former.
 Corpora natorum: see on 2. 18. The force of the periphrasis here is the same as when in the writ of Habeas Corpus the body of a prisoner is required to be produced. ‘Stat ductis sortibus urna’ = ‘stat urna, et sortes inde ducuntur.’ Comp. G. 2. 141, “Invertere satis dentibus.”
 We need not inquire how many of the subjects hinted at by Virg. were separately represented. It is sufficient to say that there was a plurality of sculptures in the Cretan part, as there had been in the Athenian. “Crudelis amor” E. 10. 29. Here the epithet is meant to excite our pity for Pasiphae as a victim, as she actually was, the passion having been Venus' revenge on her for revealing the goddess' adultery with Mars. ‘Furto’ = ‘furtim’ 4. 337. Comp. 7. 283, “Supposita de matre nothos furata creavit.”
 Forb. would make ‘domus’ nom. in apposition to ‘labor,’ like 7. 248, “Iliadumque labor vestes:” but it is doubtless gen., probably to be explained as definitive (Madv. § 286), like ‘opus Academicorum,’ ‘familia Scipionum.’ The labour is that of Daedalus, not, as Heyne thought, that of the wanderers in the labyrinth. Perhaps also ‘domus’ is to be constructed with ‘error’ as in Catull. 62 (64). 115, which Virg. had in his mind, “Tecti frustraretur inobservabilis error,” though the construction would not be quite the same as that with ‘labor.’ “Falleret indeprensus et inremeabilis error” 5. 591.
 Virg. recapitulates the heads of the story briefly, and, to one unacquainted with it, unintelligibly. ‘Magnum reginae amorem’ is not, what it would seem from the context it must mean, the passion of the queen Pasiphae, but that of the princess (comp. 1. 273: so Valerius Flaccus uses it repeatedly of Medea: see Forc.) Ariadne. ‘Sed enim’ 1. 19 note, 2. 164.
 Vestigia, not his own footsteps, but those of Theseus—another instance of Virg.'s ambiguity. The expression is from Catull. v. 113, “Errabunda regens tenui vestigia filo,” where Theseus is the subject of the sentence. Comp. also 3. 659.
 “Opere in tali” Lucr. 6.815. On the construction ‘sineret dolor’ see Madv, § 442 a. obs. 2. ‘Icare, haberes’ is omitted by Rom. and some other MSS., Ribbeck thinks on account of the length of the line.
 Patriae manus like “patrius amor” 1. 643. ‘Protinus,’ successively, G. 4. 1. The choice lies between regarding ‘omnia’ as a dactyl, and compressing it into a spondee by synizesis: a hypermeter, which Macrob. Sat. 5. 14 talks of, is not to be entertained, as in the case of other hypermetric verses in Virg. the following line begins with a vowel. See on G. 2. 69. The hypothesis of a dactylic ending would not be impossible in itself, but becomes highly improbable in the face of the fact that of all the possible instances in Virg. some, like G. 2. 69., 3. 449, may be regarded as hypermeters; others, like the present one and 7. 237, may be resolved by synizesis. ‘Omnia’ then will be a dissyllable, like ‘taeniis’ 5. 269. Copyists sought to get rid of the anomaly by substituting ‘omne’ (actually found in Rom.), ‘omnem,’ ‘omnes,’ as in 7. 237 they substituted ‘precantum,’ ‘precantis’ for ‘precantia.’
 Terentius Scaurus in his treatise De Orthographia contends that Virg. wrote ‘pellegerent,’ a form printed by Ritschl in some passages in Plautus on MS. authority. For the rhetorical use of the imperf. for the pluperf. see Madv. § 347 b. obs. 2. The plural is used because Aeneas had several companions with him: comp. vv. 13, 41, 54. ‘Praemissus,’ sent on by Aeneas, that the Sibyl might be ready for him on his arrival at the temple. “Praemittit Achaten” 1. 644. ‘Iam’ probably with ‘adforet’ rather than with ‘praemissus.’
 “Phoebi Triviaeque sacerdos” 10. 537, of Haemonides. Holdsworth and Spence (Miscellanea Virgiliana, pp. 207 foll.) distinguish between the priestess and the Sibyl, who they say, being a goddess, required some other person to introduce worshippers to her. But their distinction is not really borne out by Virg., who must have intended the same person in vv. 46 foll. and 77 foll., a patent fact which they are compelled to deny. The Sibyl is nowhere called a goddess by Virg., as in v. 258 ‘dea’ is Hecate: she is called a priestess v. 321, as they admit. It is true, as they assert, that in Silius Italicus, Book 13, where Scipio goes down into the shades, he deals in the first instance not with the Sibyl, but with the priestess Autonoe: but Silius' Sibyl is not alive, but dead: she is like Homer's Tiresias, who drinks the blood of the victim, and then acquires the power of speech, and tells the visitor what he wishes to know. They object that Deiphobe the daughter of Glaucus was not the Sibyl's name; but there were several Sibyls, and the Cumaean Sibyl in particular had several names (Dict. Myth. ‘Sibylla’), so that Virg. may have followed some legend unknown to us, or may have thought himself at liberty to invent a name. On the whole subject see Heyne's Excursus. Glaucus, as the commentators remark, is a natural personage to be a Sibyl's father, being himself a prophetic god.
 Poscunt is found in Rom., and is the earlier reading of Med. The editors think it intrinsically inferior to ‘poscit:’ but there is little difference between making the time call for the thing to be done, and making the thing to be done call for the time. It might even be urged that as ‘non’ apparently goes not with ‘ista’ but with ‘hoc,’ the latter is here the more natural expression. ‘Poscit’ however is more likely to have been altered into ‘poscunt’ than vice versa, as copyists are apt to alter the number to make the verb agree with the noun immediately preceding. See Wagn. Q. V. 8. Serv. recommends ‘poscit.’
[40-55] ‘They pass through the temple towards the adytum, when the Sibyl feels the power of the god, and calls on Aeneas to pray fervently, that the doors may open and the response be given.’
 Sacra is a substantive, so that ‘iussa sacra’ is like “iussos honores” 3. 547, “iussos sapores” G. 4. 62. ‘Morantur’ then will mean to delay to execute, or execute slowly, as in Val. F. 7. 60, “Haud ipse morabor Quae petitis,” possibly an imitation of Virg.