² See Munari, F., Il Codice Hamilton 471 di Ovidio (Rome, 1965), p. 19.Google Scholar

³ On the picking up of the simple verb by the compound and vice versa (though without specific reference to anaphora), see L.-H.-S. ii, Stilistik, § 49.c.o. On the anaphora of synonyms (though without specific reference to ‘simplex pro composito’ or vice versa), see Lausberg, H., Handbucb der literarischen Rhetorik (Munich, 1960), §§629–30,656.Google Scholar

⁴ It is possible that Ovid is seeking to reproduce a technique of Lucretian argumentation examined by David West in The Imagery and Poetry of Lucretius (Edinburgh 1969), ch. 4, esp. pp. 43 ft. (‘transfusion of terms’). The opening of the tractatio is heavily indebted for its matter to DRN 4. 1063 ff. and 1141 ff.Google Scholar

⁵ Cf. the confusion between I and t at Ovid, , Fast. 3. 726Google Scholar (vitis-, vilis), on which see Maas, P., Textual Criticism (1958), Appendix II ad fin.Google Scholar

⁶ As suggested to me by Professor Kenney.

⁷ See Goold, 96, Geisler, ad loc. To Goold's comment on Met. 9.262 populabile flammae one might add that the epithet itself helps to ‘personalize’ the dative, since populari is essentially a human activity.Google Scholar

⁸ TLL vi, s.v. finis sub fin. (‘iuncturae’).

⁹ For the characteristic hyperbaton cf. 8,641, Trist. 1.1.18.Google Scholar

¹⁰ Met. 15.1 gives little support, as the context is so different (Geisler).Google Scholar

¹¹ Cf., e.g., Cic, . Top. 82Google Scholar (status coniecturalis: ‘coniecturae ratio in quattuor partes distributa est, quarum … tenia [est cum quaeritur] quae id causa effecerit …’), 84 (s. generalis, ‘cum quaeritur quale quid sit’; cf. Or. 45, Part. 61 ff., etc.); Lausberg, , op. cit., §§ 123 ff.Google Scholar

¹² SIFC 29 (1957), 14. Cf. p. 66 of his Paravia text and p. 82 of his Berlin (1969) edition.Google Scholar

¹³ Cf. Kenney, E. J., SIFC 30 (1968), 172 n. 2.Google Scholar

¹⁴ L.-H.-S. ii. 438Google Scholar; K.-S. i. 24 f.Google Scholar

¹⁵ See further Kenney, ‘Notes on Ovid II’, CQ N.S. 9 (1959), 258.Google Scholar

¹⁶ Handford, S. A., The Latin Subjunctive (London, 1947), pp. 117Google Scholar ff.; L.-H.-S. ii. 657.Google Scholar

¹⁷ Kenney, , ‘Notes on Ovid: II’, 259. Cf. id., CQ N.S. 12 (1962), 27.Google Scholar

¹⁸ See Servius on Aen. 1.165atrum nemus for an explanation of this trope. A similar interaction between epithets can be seen at Rem. 251–2 ‘ista veneficii VETUS est via; noster Apollo/INNOCUAM sacro carmine monstrat opem’, where vetus = ‘(harmful) old’ and innocuam = ‘harmless (new)’.Google Scholar

¹⁹ See Nutting, H. C., ‘The form si sit … erit’, UCPCP 8. 2 (1926), 187 ff.Google Scholar

²⁰ As may conveniently be seen, e.g., in the index of Owen's, S. G. O. C. T. (1915) of the Tristia etc.Google Scholar

²¹ Handford, , op. cit., pp. 133 f. (§149).Google Scholar

²² Op. cit. [n. 17], p. 260. The relationship of id/tu caveas to the rest of the couplet is not made wholly clear; presumably it is to stand as an independent prohibition with reference back to 757.

²³ i.e. a reported form of quid caveam? See Handford, , op. cit., §§72 f. and n. 1 on p. 65.Google Scholar

²⁴ H, in Lenz, , 1965Google Scholar, Ha in Lenz, , 1969.Google Scholar

²⁵ Deleted by Damsté, , Mnem. 39 (1911), 445.Google Scholar

²⁶ Deleted by Heinsius.

²⁷ Deleted by Müller, L., RHM 17 (1862), 541.Google Scholar

²⁸ Bracketed by Bornecque, (1929) without comment or explanation.Google Scholar

²⁹ Liverpool Classical Monthly 3 (1978), 185–7, which also deals with the Amores ind Ars.Google Scholar

³⁰ WS 36 (1914), 39.Google Scholar

³¹ Cf. Lenz, (1965), app.crit. ad loc: ‘distichon mihi quoque valde suspectum’; see however his note in the 1969 edition (pp. 78 f.).Google Scholar

³¹ Hermes 92 (1964), 185 f.Google Scholar

³³ For the morass into which this leads, see Luck, G., Philol. 106 (1962), 147CrossRefGoogle Scholar f. In his Untersuchungen zur Textgeschichte Ovids (Heidelberg, 1969), pp. 45 f., Luck attributes the couplet to a Carolingian copy ist who found 25–6 illegible through wear and tear.Google Scholar

³⁴ Cf. Geisler, (p. 150): ‘Die Harmonie des Aufbaus leidet unter der Athetese nicht.’Google Scholar

³⁵ One is reminded of the old joke ‘If we had some bacon, we could have bacon- and-cggs – if we had some eggs.’

³⁴ Were it worth considering emendation then perhaps concipiunt or concipient mighi be offered. But both ‘my (remaining) years hold many poems’ and ‘my years (will) conceive many poems’ seem foreign to Ovid's idiom.

³⁷ Cf., e.g., Cic, . Cat. 4Google Scholar.6, Ad Fam. 13.64.1. The word does not occur elsewhere in Ovid's works; he has sustinere in a similar context at A. A. 2.690, but meaning producere (‘utque morer meme sustineamque, rogat’).Google Scholar

³⁸ Cf. Prop. 4.9.62 and 70, Tib. 1.3.78. The poetic idiom exemplified by sitis Herculis – Hercules sitiens is of course both ancient and universal, and may readily be found in Ovid (e.g. Met. 1.58 cura dei = deus curans, 1.74 ira Iovis = Iuppiter iratus but he does not appear to employ it where physical ) like hunger, thirst, or pain ar involved.

³⁹ With the former Vollmer, F. (Hermes 52 (1917), 468Google Scholar) compared the Senecan dolori adesse (De Cons. 7.2Google Scholar.), which = dolorem nutrire (see further TLL ii. 925Google Scholar, s.v. adsum i.q.faveo. But what fatumnutrire would mean is hard to guess; perhaps mortem parare, or (closer to the original dative) exitio imminere (cf. Met. 1.146). Reading obesse, Professor Kenney would translate (loosely) ‘is lousing up his entire life’, but without much conviction.Google Scholar

⁴⁰ Cf., e.g., Plaut, . Merc. 702Google Scholar ff., or Mart. 8.12 (Kenney). The previous couplet (563–4) introduces the Comic figures of the durus pater and (by implication) adulescens flagitiosus; 565–6 perhaps owe their genesis to that.Google Scholar

⁴¹ Lausberg, , op. cit., §§643 and 647.Google Scholar

⁴² Platnauer, M., Latin Elegiac Verse (Hamden, Connecticut, 1971), pp. 64 ff.Google Scholar

⁴³ ‘divitiis pereat luxuriosa suis.’

⁴⁴ RHM 17 (1862), 541.Google Scholar

⁴⁵ LAC 35 (1966), 581 ff.Google Scholar

⁴⁶ Senecae Tragoediae i (1878), 174.Google Scholar

⁴⁷ Pasiphae precedes Phaedra in the catalogue of love's casualties at 63 f.

⁴⁸ The hexameter occurs on a wall at Pompeii (Bücheler, CLE 354.2)Google Scholar; cf. Am. 1.8.77–8Google Scholar (CLE 1785)Google Scholar, A.A. 1.475–6 (CLE 936.1–2).Google Scholar

⁴⁹ Cf. Fast. 6.796 ff. (796 ‘cum data sunt trabeae templa, Quirine, tuae’, 809 Caesar, 812 Alcides).Google Scholar

⁵⁰ See Lee, A. G., ‘The Authorship of the Nux’, in Ovidiana, ed. Herescu, N. I. (Paris, 1958), pp. 468 f. The Halieutica was accepted as by Ovid in Pliny's day (and afterwards).Google Scholar

⁵¹ The archetype of our manuscripts contained all four works in the order Ars, Remedia, Amoves, Heroides. See Goold, 3 f. (Luck, , Untersuchungen zur Text geschichte Ovids, pp. 11 ff. and 45 f., argues otherwise, unconvincingiy.)Google Scholar

⁵² The positions of the interpolated verses are, with very few exceptions, consistent with their having been added at the foot of the page in a codex having 13 lines to the page (allowing appropriate space for titles etc.). This would be a great rarity, of course, though not an impossibility (cf., e.g., Lowe, , CLA i. 23Google Scholar, Vat. Pal. Lat. 24 of Geilius; fourth century, capitalis rustica, 13 lines). We have nothing nearly as old as the second century to provide a control. But calculations of this kind in these circum-stances are unprofitable, since too many unverifiable assumptions require to be made.Google Scholar