183) who takes refuge in the usual remedy of the desperate, the assumption of the loss of a line, for which there is no sort of evidence.

superno Pithoeus; superbo S; superhuc T.

II. marinis Rivinus; maritis S T (from line 4).

Buecheler

omits de marinis imbribus, thinking that the end of the line was lost, and that these words were introduced from line 4 to fill up the gap.

13. floridis Rigler; floribus S T.

15. nodos the conjecture of a friend of Scriverius for notos S, totos T.

feraces, Bachrens; penates S; pentes T. Cp. Catull. 64. 283. Mackail suggests tumentes.

17. en Schultze; et S T.

micant Lipsius; micanat S; mecanat T.

jectured emicant.

19. en Schultze; in S T.

Salmasius con

21. virgineas is genuine; for the metre cp. 17, and see Buecheler, p. 33. Lipsius conjectured virgines.

umenti Pithoeus and Lipsius; tumenti S T; urenti Baehrens.

22. mane ut udae Douza; manet tute S; mane tuae T. Mackail says 'iubere occurs seven times again in the poem, always with an infinitive; nor indeed is it Latin to say iubere ut. Iubeo ut is not very common, but is perfectly good Latin, and means I order.' Here the meaning is 'the fiat of Venus is,' See Plaut. Amph. I. I. 50, Telebois iubet sententiam ut

dicant suam; Cic. II. in Verr. 4. 28, hic tibi in mentem non venit iubere, ut haec quoque referret HS VI milibus I se tibi vendidisse? Hor. Sat. 1. 4. 120, sic me formabat puerum dictis, et sive iubebat ut facerem quid; Liv. 28. 36, Magoni nuntiatum ab Karthagine est iubere senatum ut classem in Italiam traiceret. Indeed it is regular in the phrase populus iubet ut. See Kühner, Ausführlicher Grammatik, 2. 530. Buecheler and Baehrens adopt the prosaic reading of Orelli, ipsa iussit mane totae.

23. facta S; fusta T.

was

Veneris is my conjecture for prius S T. The rose created from blood of Venus and kisses of Love. Buecheler and Baehrens adopt O. Müller's conjecture Cypridis, which introduces a scansion less common in Latin (cp. Anth. Lat. 210. I (Baehrens), candida sidereo fulgebat marmore Cypris; ibid. II temerantur Cypridis artus); though the quantity can be paralleled from Anth. Lat. 314. 4, dulcia vina Cypris; 317. 3, sors fuerat melior, Cypridos cum nomen haberes. Scriverius conjectured Cypris, but there is no authority for such a genitive. Cp. Tiberianus (Baehrens, Poet. Lat. Min. 3. 264) 1. 8, inter ista dona veris gemmeasque gratias omnium regina odorum vel colorum Lucifer aurea forma Diones eminebat flos rosa: so I correct the corrupt reading of the MS. auro flore praeminebat forma dionis rosa.

25. ruborem S T rightly; pudorem Baehrens.

26. unico Pithoeus; unica S T.

voto Bergk and Baehrens.

It means 'marriage,' as in the code of Justinian. noto S; nodo T. Salmasius read unico marita nodo . . . solvere, 'to unfold from its single bud' (Mackail). Buecheler reads unico něto marita (p. 37).

For pudebit, here personal as in Terence Adelph. 754, Lucan 8. 495, Baehrens reads rubebit.

29. it Pithoeus; et S T.

30. vexerit MSS. Vehere is used by poets in the sense of portare; this is illustrated at length by Gronovius Observat. 3. 5, and there is no need with Baehrens to read exuit. The Nymphs are afraid of the arms carried by Cupid; he is therefore bidden to lay them aside.

35. est in armis totus idem Scriverius; totus est inermis (in armis Pithoeus, Baehrens) idem S T. The transposition of words made by Scriverius is necessary on account of the metre, which does not admit a spondee in the third foot. In means 'arrayed in,' 'equipped with,' as in Verg. Aen. 3. 595, patriis ad Troiam missus in armis; 4. 537, horridus in iaculis; Ov. M. 8. 26, in galea formosus erat; 12. 65, in armis hostis adest.

37. conpari S T ; conparis Baehrens, unnecessarily. The meaning is: O Diana, Venus sends to you maids as modest as yourself to pray you to refrain from bloodshed in her wood.

40. This line, which follows 58 in the MSS., is rightly placed here by Buecheler. recentibus Pithoeus; rigentibus S T.

43. feriantis Scriverius ; feriatis S T ; feriatos Pithoeus.

47. detinenda nocte tota est pervigilia canticis is my emendation (detinenda Heinsius).

detinente tota nox est peruiclanda canticis S; detinent et tota nox est peruigila canticis T. I think the pervigila of T gives us the clue required. The poet uses the rare singular form pervigilia, fem.,

found in Iustin. 24. 8. 14 (see Georges, Lexicon), fames et lassitudo super haec maximum pervigiliae malum. For the meaning of detinere 'to while away' see Ov. Trist. 5. 7. 39, detineo studiis animum falloque dolores; ex P. 4. 10. 67, detinui, dicam, tempus curasque fefelli. Buecheler reads detinenda tota nox est, perviglanda canticis, but professes himself not satisfied: Baehrens after Schenkl reads detinenter tota nox est perviclanda canticis: but detinenter does not exist in Latin. The contracted forms perviglanda, perviclanda seem to me highly objectionable, unless supported; though the forms puertia, lamna, surpuerat, caldior, soldo, are found in Horace.

53.s ubde Scriverius ; vestem Salmasius; superestem S; rumpereste T; sume restem Crusius, followed by Buecheler; rumpe vestem Lipsius. Perhaps we should read subde restem (a chain or rope of flowers).

Hennae Lipsius; et nec S; ethne T; (Aetnae Pithoeus).

55. fontes Scriverius; montes S T.

56. Buecheler (p. 13) objects to the anapaest (pueri) in the first place in the third dipodia; but it occurs again 62, which line Buecheler is compelled to eject as an interpolation, while he emends this line, iussit omnes adsidere mater alitis dei. These are violent remedies.

59. cum quom) Buecheler; quo S; qui T.

60. totum Salmasius; totis S T.

creavit S; crearet T.

vernus Pithoeus; vernis S T.

Combining the conjectures of Salmasius and Pithoeus I have reconstructed this line. Vernus seems necessary on account of the metre, as a spondee is here impossible: pater vernus means 'the Father in spring,' like Vergil's Aeneas se matutinus agebat (Aen. 8. 465), or Juvenal's fremeret saeva cum grandine vernus Iuppiter (5. 78). Cp. Verg. G. 2. 323, vere tument terrae et genitalia semina poscunt. tum pater omnipotens fecundis imbribus Aether coniugis in gremium laetae descendit et omnis magnus alit magno conmixtus corpore fetus.

66. Baehrens conjectures praevium sui teporem. See Rhein. Mus. 31, p. 102.

69. For nepotes Buecheler and Baehrens follow Rivinus (not Scriverius) in reading penates; which is an ingenious alteration.

70-74. In the MSS. 71, 72 follow 73, 74; I follow Baehrens in transposing them. In 71 posterum is genitive; see Buecheler, p. 29. In 72 Buecheler conjectures patrem for matrem, which seems superfluous. In 73 de sacello refers to the fact that Silvia was a Vestal Virgin: cp. Ov. Fast. 3. II. In 74 Buecheler argues against Romuleas; his main objections are (1) that it does not mean Roman, but belonging to Romulus; (2) that it does not scan (Romulčas). But I do not see any difficulty in understanding Romuleus or Romulius in a poet as equivalent to Romanus; the Romans were the Romuli nepotes. Cp. Hor. Carm. 4. 5. I optime Romulae custos gentis. Carm. Saec. 47. As regards the quantity, Roman poets of the best period take such liberties with the scansion of proper names (see on Catullus 29. 20), that I think we may accept Romuleas from the author of the Pervigilium. The date of the poem is uncertain; it was written under the Empire, and probably not early. Teuffel (398.5-7) ascribes it to the third century of the Christian era and a poet