¹ Frivolity and light irony: Campbell, A. Y., Horace (London, 1924), p. 225Google Scholar, Commager, H. Steele, The Odes of Horace (New Haven and London, 1962), pp. 16, 111, 142Google Scholar. Pasquali, G., Orazio Lirico (Florence, 1920), pp. 463–70 provides an interesting treatment, comparing Propertius 3.12 but seeing no Odyssean aspect to Horace's poem. No mention of Odysseus either in the useful accounts ofGoogle ScholarSyndikus, H. P., Die Lyrik des Horaz (Darmstadt, 1973), 2.98–102Google Scholar and Mutschler, F.-H., SO 53 (1978), 111–32CrossRefGoogle Scholar. For other articles on the poem to 1977, see Mutschler 126 n. 4 and Kissel, W. in ANRW 31.3 (1981), 1505Google Scholar; add Bradshaw, A., Hermes 106 (1978), 158–61Google Scholar. The standard commentaries of Kiessling/Heinze and Gordon Williams are cited below by author's name only.

² In that poem, however, the poet attempts to seduce a ‘puella’ left in Rome by her lover on a trading trip to Africa.

³ So the name of Pompeius, is postponed after a descriptive clause to the head of the second stanza in Odes 2.7Google Scholar; for the technique of name-postponement cf. Pearce, T. E. V., CQ 18 (1968), 338–40CrossRefGoogle Scholar.

⁴ Cf. Mutschler, art. cit. (n. 1), 113 n. 9; this allusion to Γὑγης πολὑχυσος (Archilochus, fr. 19.1 West – cf. Herodotus 1.14) appeared in the second edition of Kiessling's Horace commentary (1890), but was later cut out by Heinze (it does not appear in the final Kiessling/ Heinze of 1930).

⁵ Noted by Kiessling/Heinze; for this τóπος cf. Page, Denys, Further Greek Epigrams (Cambridge, 1981), p. 161Google Scholar. Asterie's name is also mythological, being that of a sister of Leto and thence an earlier name of the island of Delos (where she was said to have leapt into the sea to escape the embraces of Zeus) – cf. Hesiod, , Theog. 409Google Scholar with West's note, Callimachus, H. 4.37 with Mineur's note.

⁶ The ‘Capra’ or she-goat may also symbolize the problems of passion which retain Gyges, for this animal is associated with Venus – cf. Keller, O., Die antike Tierwelt (Leipzig, 1913)Google Scholar, i.306.

⁷ The change from winter storm-winds to soft spring breezes is naturally enough a τóπος of the spring poem – cf. Catullus 46.1–3, Meleager, , AP 9.363.1, 10Google Scholar, Agathias, , AP 10.14Google Scholar.

⁸ A point noted by Syndikus, , op. cit. (n. 1), p. 99 n. 8Google Scholar.

⁹ Analogous is the only other archaic form of ‘fides’ in Horace, dative ‘fide’ at Sat. 1.3.95 ‘commissa fide’, where Horace modifies the legal technical term ‘fidei commissa’ for dactylic verse but retains an archaic and legalistic flavour by using the old form.

¹⁰ Cf. conveniently Syndikus, , op. cit. (n. 1), p. 99 n. 9Google Scholar.

¹¹ Penelope, exemplum pudicitiae: Propertius 2.8.17, 2.14.27, 2.24.35, 2.34.93, 3.3.17, 3.10.15, 3.12 passim, 3.13.10, 24, 4.1.71, 4.7.49; Ovid, Am. 1.8.47, 2.18.21, 29, 3.4.23, Ars 1.478, 2.103, 139, Rem. 66.

¹² Cf. esp. Jacobson, H., Ovid's Heroides (Princeton, 1974), pp. 263ffGoogle Scholar.

¹³ Odes 3.10.11, Sat. 2.5.76ff.

¹⁴ So Kiessling/Heinze and Nisbet/Hubbard ad loc.

¹⁵ So Anaereon himself re-works the lament of Helen, in Iliad 6.345ffGoogle Scholar. at PMG 347.1 Iff. (cf. Führer, R., Formproblem – Untersuchungen zu den Reden in der frühgriechischer Lyrik [Zetemata, 44] (Munich, 1967), pp. 130–2Google Scholar, Emley, M. L. B., CR 21 [1971], 169)Google Scholar; Alcman, uses Circe at PMG 80Google Scholar, and Alcaeus' concern with Helen is well-known (frr. 42 and 283 L/P).

¹⁶ So ‘atqui’ articulates a similar contrast at Odes 3.5.49.

¹⁷ Cf. Pichon, R., De Sermone Amatorio Apud Latinos Elegiarum Scriptores (Paris, 1902), pp. 166 and 265Google Scholar.

¹⁸ So Gordon Williams ad loc.; the quotation is also from his commentary.

¹⁹ Mutschler, art. cit. (n. 1), 115 n. 5 connects Chloe with the similar cult-title of Demeter, but this seems a long shot. On another Horatian Chloe in vegetative surroundings cf. Odes 1.23.1 with Nisbet/Hubbard's note.

²⁰ For the Bellerophon, story cf. Iliad 6. 144ffGoogle Scholar. (it was also told in lost plays – Sophocles' Iobates and Euripides' Stheneboia and Bellerophon); for the Peleus story cf. Pindar, , Nem. 4.57ff., 5.25ffGoogle Scholar.

²¹ For magnets from Magnesia cf. Lucretius 6.908–9 ‘quem magneta vocant patrio de nomine Grai, / Magnetum quia fit patriis in finibus ortus’, Plato, , Ion 533bGoogle Scholar, Pliny, , HN 36.128Google Scholar, Pease on Cicero, , Div. 1.86Google Scholar. This connection is Professor Nisbet's, and will be published in a forthcoming article of his. A speaker at Achilles Tatius 1.17.2 quotes the ‘attraction’ of the magnet as an example of the power of love, perhaps indicating that ‘magnetic love’ was a literary τóπος.

²² For ‘peccare’ cf. Pichon, , op. cit. (n. 17), p. 227Google Scholar.

²³ Bentley (ad loc.) regards both readings as possible (‘utrumque recte’) but prefers the second: ‘illud tamen magis placet ut figuratus et ποιητικώτερον’.

²⁴ The sequence ‘frustra, nam’ is taken from Catullus 21.7, as Kiessling/Heinze note: for the emphatic enjambement in the Odes of a single word at the end of its clause and the beginning of a stanza cf. 1.2.49 and 3.1.37.

²⁵ Cf. Euripides, , Medea 28Google Scholar (cited by Kiessling/Heinze) with Page's note, Vergil, , Aen. 7.586ffGoogle Scholar.

²⁶ On storms in the Icarian Sea cf. Odes 1.1.15 with Nisbet/Hubbard ad loc.

²⁷ Propertius 3.13.10; Ovid, Pont. 3.1.113; cf. Culex 265.

²⁸ Mr Hollis suggests that the choice of name may owe something to Callimachus, , Aetia fr. 23.2–3Google Scholar Pf. ώς έλς κοὑει | Σελλς ν Tμαροις οὐρεσιν 'Iκαης, a similar exemplum of paying no heed (in that case, to curses).

²⁹ In the Odyssey (by contrast) Odysseus is of course tied up previously by his own orders, but tries to escape under the spell of the Sirens (12.192ff.)

³⁰ On the (generally Stoic) moralizing interpretation of Odysseus' resistance to the Sirens and of his other adventures cf. Stanford, W. B., The Ulysses Theme (Oxford, 1954), pp. 121–7Google Scholar, Buffière, F., Les Mythes d'Homère et la pensée grecque (Paris, 1956), pp. 369ffGoogle Scholar., Rahner, H., Greek Myths and Christian Mystery [tr. Battershaw, B.] (London, 1963), pp. 328ffGoogle Scholar.

³¹ Odysseus was at least technically unfaithful to Penelope with Circe and Calypso in the Odyssey – a suggestion perhaps as in the matter of freely resisting the Sirens (cf. n. 29 above) that Gyges follows the moralized rather than the strictly Homeric Odysseus.

³² Antinous, (along with Eurymachus) is said to be the ‘best man’ amongst the suitors (Odyssey 21.185–6)Google Scholar, and his father Eupeithes is clearly an Ithacan (24.420ff.). Propertius (4.5.8) seems to regard him as the most likely/attractive of the suitors.

³³ Enipeus' name may also appropriately recall the use of that river as a disguise by Zeus, in his seduction of the heroine Tyro, an episode mentioned in the Odyssey (11.235ffGoogle Scholar.; cf. Apollodorus, Bibl. 1.9.8). On swimming in the Tiber as an athletic exercise in Rome cf. Nisbet/ Hubbard on Odes 1.8.8 and Griffin, J., Latin Poets and Roman Life (London, 1986), p. 89Google Scholar.

³⁴ The tibia: Propertius 2.7.11; Ovid, Am. 3.13.11.

³⁵ ‘Querulus’ recalls the traditional ‘queri’, ‘querela’ and ‘questus’ of the lover's complaint (Pichon, , op. cit. [n. 17], p. 248Google Scholar), while ‘despicere’ and ‘dura’ commonly describe the heartlessness of the ‘puella’ – cf. Pichon, pp. 128, 136 and Commager, , op. cit. (n. 1), p. 293 n 46Google Scholar.

³⁶ For ‘difficilis’ in the amatory sense of ‘unresponsive’ cf. Syndikus, , op. cit. (n. 1), p. 111 n. 17Google Scholar, Pichon, , op. cit. (n. 17), p. 130Google Scholar.