² ω = all or most of the recentiores (some twenty) which I take into account, s = some or a few. Sigla, where used, are Munari's, but in most cases I cite the manuscripts by name.

³ Which he seems to have misunderstood, alleging that prendere here = iungere.

⁴ See Palmer's, edition, p. 422Google Scholar and Ullman, , Classical Philology, xxvii. 13. I shall deal with these florilegia more fully later.Google Scholar

¹ As indeed collo ought to be taken if it be read; v. 41 cannot mean that the lady is to be black and blue all over.

² Anaphora instead of copula: cf. Ovid, Her. 1. 73 quaecumque aequor habet, quaecumque pencula tellus, etc.

³ On the construction see further Hosman, , J.P. xxx. 238;Google ScholarBell, , The Latin Du and Poetic Diction, pp. 264–78, though I cannot follow him in every case.Google Scholar

¹ Cf. A.A. 3. 569 nec scindet tunicasue suas tunicasue puellae, Tib. 1. 10. 61 sit satis e membris tenuetn perscindere uestem, Prop. 2. 5. 21 nec tibi periuro scindam de corpore uestis.

² And at Met. 3. 480 read, with the minority of manuscripts, summa uestem diduxit ab ora. Planudes translates = scidit. Magnus's citation of Eur. Hec. 558 is very apposite.

¹ I do not mention the passages where Weise refrained from following such fashions, since he was actuated only by fidelity to the text of the Aldine editions. The chief merit of his text is its punctuation: cf. Housman's remarks in his Lucan, , p. xxxiii.Google Scholar

² Tr. 2. 379 unde nisi indicia magni sciremus Homeri … ? is not a fair parallel, since unde is felt as = ex quo; and Balbus, Att. 9. 13a. 1 breuitate epistulae scire poteris … does not perhaps carry much weight.

³ Since manibus here = ‘arms’. For this common sense (cf. Greek) cf., e.g., Her. 10. 40, 146, 17. 166, 20. 58, al., and see Thes. L.L. 8. 343. 3 ff.

¹ In spite of Met. 3. 302 qua tamen usque potest, uires sibi demcre temptat: it would be the height of pedantry to indicate the hyperbaton by punctuation.

² ‘ex uno Moreti’ = Mus. Plant, lat. 68.

¹ For this construction with metonymy cf. Dilke's note on Statius, , Ach. I. 608.Google Scholar

² Cf. the common captus, laesus aliquo in an erotic sense.

³ Analogous to the ablative with careo, orbus esse, et sim.

⁴ Hau, P., De casuum usu Ovidiano, Münster, 1884.Google Scholar

⁵ I make no rash assertions about the other poets; but see in particular Getty's, R.J. edition of Lucan 1 (², 1955), p. xlviii.Google Scholar

⁶ See Platnauer, , C.Q. xlii. 91 ff.;Google ScholarLatin Elegiac Verse, pp. 78 ff.Google Scholar On A.A. 2. 91 I suspend judgement.Google ScholarPubMed

⁷ Müller's ague ita is palaeographically easier, but ita would have to be taken with sollicito, since one can hardly assume hyperbaton of such a word; andthismakesno sense that I can see. ita owes its existence to that desire of readers for a dactyl at this place in the verse which gives us, for example, uixerit et for uiuent et (or et uiuet? I have never seen this suggested) at 1. 15. 18.

⁸ Not, however, Harder and Marg in their translation (Munich, 1956).

⁹ As Brandt points out, Strabo's editors print at 15. 736. But the most modern critical editions of Strabo are over a century old, and I do not attach much importance to this, in view of the known badness of his manuscripts.

¹ See above, p. 55.

¹ Compare perhaps the errors in Bodleian MS. Auct. F.4.32, the ninth-century fragment of the Ars Amatoria (a Welsh hand): at A.A. 1. 267Google ScholarPubMed it has ubig, at 655 … This seems to point to a confusion between (quod) and (que). Cf. Lindsay, , op. cit., p. 228.Google Scholar

² See further Housman, , C.R. iii. 315;Google ScholarMiller, , Classical Journal, xi. 516 ff.;Google ScholarKroll, , Studien turn Verständnis der römischen Literatur, 263 n. 42.Google Scholar

¹ Mr. G. W. Williams has kindly drawn my attention to Lucan 8. 198 nec idem spectante carina (cf. 179). This goes a little way to allay my uneasiness, but I am not convinced that it is a sufficient parallel.

¹ Answer, I suppose, ‘manuscript authority’ ; but what is the authority of a ninth-century manuscript, or any other, in such a matter ?

² There is little or no implication of vividness, as is sometimes suggested (e.g. by Owen at Tr. 2. 33).

³ At A.A. i. 278 read aget, not agat, as I shall argue.Google Scholar

⁴ Kiihner-Stegmann, ii. I. 342 condemns, for example, meae laudi est as a solecism. The Thes. L.L. nevertheless includes the present passage among the examples of curae esse alicui without remark (4. 1456. 53 f.).Google Scholar

¹ Translating ‘Through this, Homer, hadst thou wished it, she might have been kind to thee’, i.e. hoc ablative. This is possible, and is certainly better than hoc nominative or accusative = ‘this rank’ or ‘this life’, but it leaves uelles more feeble than ever, hoc, it seems to me, must be eliminated.

² Communicated to me by letter and published here by Professor Campbell's generous permission.

³ At Plaut. Bacch. 252 istius hominis ubifit quaque mentio it means, if right, ‘whenever’; Lambinus read cumque. This is the only passage I can find in which the manuscripts give ubiquaque; it was restored by Merkel at A.A. 2. 627, where the manuscripts give scilicet excuties omnis ubiquaeque (or ubicumque) puella est. Read rather omnis, ubi quaeque,puellas (the last word restored by Heinsius ex codd.),Google ScholarPubMed

¹ See now Bailey, Shackleton, Propertiana, p. 203.Google Scholar

² The Scylla of whom Ovid is thinking is not Homer's Scylla of six heads, but a monster with the head and torso of a woman (Met. 13. 733), who seized her victims in her hands (Xen. Mem. 2. 6. 31); she terminated in serpentine fishtails and round her loins a brood of dogs tore the prey consigned to them: Virg. A. 3. 424 ff.,Google ScholarPubMedBuc. 6. 74 ff.Google Scholar (= Ciris 59 ff.),Google Scholar and cf. Ellis on Catullus 60. 2. She is frequently so depicted in ancient art: see Waser ap. Roscher s.v. Skylla, also Waser's treatise, Skylla u. Charybdis in d. Literatur u. Kunst d. Griechen a. Römer, Zurich, 1894, Passim.Google Scholar

³ iubeat Bailey, Shackleton, C.Q., N.s. iv. 166;Google Scholar but Dr. Shackleton Bailey has himself iince drawn my attention to Eur., Bacch. 476 The passages from the Amoves and Ars support each Other.