1. Ov. Met. 7.350-93. On the flight, see esp. Schubert, W., ‘Medeas Flucht aus Iolcos (Ovid, Met. 7,350-393)’, WJA 15 (1989), 175–81Google Scholar; Binroth-Bank, C., Medea in den Metamorphosen Ovids (Frankfurt am Main 1994), 137–42Google Scholar; McKinley, K.L., Reading the Ovidian Heroine: Metamorphoses Commentaries, 1100-1618 (Leiden 2001), 21–27CrossRefGoogle Scholar; Pavlock, B., The Image of the Poet in Ovid's Metamorphoses (Madison 2009), 49–60Google Scholar. On Medea in the Met. more generally, see Rosner-Siegel, J.A., ‘Amor, Metamorphosis and Magic: Ovid's Medea (Met. 7.1-424)’, CJ 77 (1982), 231–43Google Scholar; Hinds, S., ‘Medea in Ovid: Scenes from the Life of an Intertextual Heroine’, MD 30 (1993), 9–47Google Scholar; Binroth-Bank, op. cit., passim; Newlands, C.E., ‘The Metamorphosis of Ovid's Medea’, in Clauss, J J. and Johnston, S.I. (eds.), Medea: Essays on Medea in Myth, Literature, Philosophy, and Art (Princeton 1997), 178–208Google Scholar; Pavlock, op. cit., 38-60. For allusions to the chronological future in the textual past, see Barchiesi, A., ‘Future Reflexive: Two Modes of Allusion and Ovid's Heroides’, HSCP 95 (1993), 333–65Google Scholar.

2. A partial exception is the story of Hyrie and Cycnus (7.371-81), which, while brief, is actually narrated rather than merely referred to.

3. On the reification of figurative language in general, see esp. Rosati, G., Narciso e Pigmalione: illusione e spettacolo nelle Metamorfosi di Ovidio (Firenze 1983), 166–70Google Scholar; Kaufhold, S.D., ‘Ovid's Tereus: Fire, Birds, and the Reification of Figurative Language’, CP 92 (1997), 66–71Google Scholar; in metamor-phosis: Tissol, G., The Face of Nature (Princeton 1997), 11–26Google Scholar; in personifications: Tissol, op. cit., 61-88; Hardie, P., Ovid's Poetics of Illusion (Cambridge 2002), 231–36Google Scholar.

4. Calypso: Ap. Rhod. 4.572-76, with Nelis, D., Vergil's Aeneid and the Argonautica of Apollonius Rhodius (Leeds 2001), 60f.Google Scholar; Ithaca: Virg. Aen. 3.273, with Nelis, loc. cit., and Horsfall, N., Virgil, Aeneid 3: A Commentary (Leiden 2006), 218CrossRefGoogle Scholarad loc.; Phaeacians: Virg. Aen. 3.291, with Nelis, loc. cit., and Horsfall, op. cit., 231 ad loc.; Circe: Virg. Aen. 7.5-24, with Basto, R., ‘The Grazing of Circe's Shore: A Note on Aeneid 7.10’, CW 76 (1982), 42f.Google Scholar, and Horsfall, N., Virgil, Aeneid 7: A Commentary (Leiden 2000), 50Google Scholarad loc. Cf. Gowers, E., ‘Virgil's Sibyl and the “Many Mouths” Cliché’, CQ 55 (2005), 170-82, at 176CrossRefGoogle Scholar: ‘[The Sibyl's] style of exegesis, half suggestive and half brisk, makes the tour of the Underworld literally an extended praeteritio, one that opens up tantalizing half-glimpses and closes off potential vistas.’ Horsfall's use of the formulation ‘literal praeteritio’ is anticipated by Miller, A.M., From Delos to Delphi: A Literary Study of the Homeric Hymn to Apollo (Leiden 1986), 31Google Scholar, on Leto's passing of possible sites for Apollo's birth at H.H.Ap. 30-44, and Feeney, D.C., The Gods in Epic (Oxford 1991), 324Google Scholar, with reference to the Argonauts’ passing of the land of the Amazons at V.FI. 5.120-39.

5. Though many scholars refer to aspects of the passage which reflect its status as a praeteritio, I am not aware of any who have explicitly identified it as such or made the connection with Medea's ‘passing over’. Closest, albeit I think inadvertently, is David Raeburn's own dismissive praeteritio in the notes to his Penguin translation (Harmondsworth 2004) as he declines to annotate the ‘fifteen minor metamorphoses and a few other mythological references, which need hardly detain the reader any more than they detained Medea’ (650).

6. First Minyeis: 4.43-54, with G. Rosati, in A. Barchiesi and G. Rosati (eds.), Ovidio, Meta-morfosi Vol. 2 (Libri III-IV) (Milano 2007), 254 ad loc.; Alcithoe: 4.273-84 (using the actual word praetereo at 4.284), with Myers, K.S., Ovid's Causes (Ann Arbor 1994), 80CrossRefGoogle Scholar, inter multos alios; Arachne: 6.108-14, with Wheeler, S.M., Narrative Dynamics in Ovid's Metamorphoses (Tübingen 2000), 99Google Scholar.

7. On ‘untold stories’ in general, see Tarrant, R., ‘Roads Not Taken: Untold Stories in Ovid's Metamorphoses’, MD 54 (2005), 65–89Google Scholar. For an instance of a palimpsestic myth, see Gildenhard, I. and Zissos, A., ‘Ovid's Narcissus (Met. 3.339-510): Echoes of Oedipus’, AJP 121 (2000), 129–47Google Scholar.

8. Hill, D.E. (ed.), Ovid Metamorphoses V-VIII (Warminster 1992), 204Google Scholar.

9. Ov. Met. 7.368-70; Ant. Lib. 1.

10. That Ovid invented many of the Met.'s myths (or versions of and details within existing myths) is widely accepted. For example, Galinsky, G.K., Ovid's Metamorphoses: An Introduction to the Basic Aspects (Berkeley 1975)Google Scholar, assumes Ovidian invention of Echo's petrification (55), the Thracian Maenads’ arborification (90), and the metamorphoses of Cyane and Caeneus (181). For the ‘ironic use of praeteritio to highlight mythological innovations’, see Peirano, I., ‘Mutati Artus: Scylla, Philomela and the End of Silenus’ Song in Virgil Eclogue 6’, CQ 59 (2009), 187-95, at 193 η33 and Tarrant (n.7 above), 66f.CrossRefGoogle Scholar

11. ‘[Q]uesto…è un volo di pura invenzione, che offre l'occasione per uno sfoggio d'erudizione in forma di rapida rassegna di storie che per qualche motivo Ovidio ha deciso di non raccontare in dettaglio.’ (‘This is a flight of pure invention, which offers the opportunity for a display of erudition in the form of a rapid survey of stories which for some reason Ovid has decided not to narrate in detail.’) Kenney, E.J. (ed.), Ovidio, Metamorfosi Vol. 4 (Libri VII-IX) (Milano 2011), 258Google Scholarad 7.350.

12. ‘Bei aller Vielfalt und Heterogenität der Sagen zeigt sich…daß Ovid um eine, wenn auch sehr allgemein assoziative, motivische Verknüpfung bemüht ist.’ Binroth-Bank (n.1 above), 142.

13. ‘[B]ei der Mehrzahl dieser Sagen ein thematischer Zusammenhang mit der Medeasage auf-grund der Identität einzelner Mythologeme besteht, der den topographischen überlagert.… [Some stories] sind mit den anderen vielleicht eher assoziativ verbunden.’ Schubert (n.1 above). 181.

14. McKinley (n.1 above), 27.

15. Pavlock (n.1 above), 50.

16. ‘Wegen unserer mangelnden Vertrautheit mit solch entlegenen Mythen läßt sich das tertium comparationis, das diese Mythen möglicherweise miteinander verbindet, nicht ohne weiteres erk-ennen; dabei besteht die Gefahr, daß man ein uns ersichtliches Motiv, hier z.B. das der Topographie, allein (als das nächstliegende) assoziiert und als hermeneutischen ‘Rettungsanker’ verwendet, während dieses Motiv für den Dichter selbst vielleicht nur ein oberflächliches Band war und das tatsachliche tertium comparationis eher auf einer anderen Ebene zu suchen ist.’ Schubert (n.1 above), 178, original emphasis.

17. The term coined by Solodow, J.B., The World of Ovid's Metamorphoses (Chapel Hill 1988), 43f.Google Scholar; the quote from Morgan, LI., ‘Child's Play: Ovid and his Critics’, JRS 93 (2003), 66–91, at 88Google Scholar. Cf. Wheeler, S.M., A Discourse of Wonders: Audience and Performance in Ovid's Metamorphoses (Philadelphia 1999), 119Google Scholar.

18. Contra the assumption that Hellenistic metamorphosis poems were necessarily obscure to Augustan readers, see Hutchinson, G.O., ‘The Metamorphosis of Metamorphosis: P. Oxy. 4711 and Ovid’, ZPE 155 (2006), 71–84Google Scholar = Talking Books (Oxford 2008), 200–27CrossRefGoogle Scholar.

19. Tarrant (n.7 above), 66.

20. First Minyeis (of Pyramis and Thisbe in clear antithesis to the rejected myths): quoniam uulgaris fabula non est (‘since the story is not commonly known’, 4.53); Alcithoe: uulgatos ta-ceo…mnores (‘I remain silent about love-affairs which have been made commonly-known’, 4.276).

21. On ‘Ovid's Aeneid’, see esp. Solodow (n.17 above), 110-56; Tissol, G., ‘Ovid's Little Aeneid and the Thematic Integrity of the Metamorphoses’, Helios 20 (1993), 69–79Google Scholar; Papaioannou, S., Epic Succession and Dissension: Ovid, Metamorphoses 13.623-14.582, and the Reinvention of the Aeneid (Berlin and New York 2005)CrossRefGoogle Scholar. The events of Euripides’ tragedy are given in four lines (7.394-97), with the possible addition of the first five feet of 7.398 (hinc Titaniacis ablata draconibus, ‘borne away from this place by the sun-god's dragons’) which form a bridge into the Athenian world of Sophocles’ Aegeus.

22. Cf. Kenney (n.11 above), 258 ad 7.353-56, on this aspect of the Cerambus notice: ‘discre-panze e contraddizzioni di questo tipo, prodotte da un'ostentata erudizione’ (‘discrepancies and contradictions of this kind, produced by ostentatious erudition’).

23. The most thorough study of the role of the reader in the Met. is Wheeler (n.17 above). On Roman reading practices, see esp. Parker, H.N., ‘Books and Reading Latin Poetry’, in Johnson, W.A. and Parker, H.N. (eds.), Ancient Literacies: The Culture of Reading in Greece and Rome (Oxford 2009), 186–229Google Scholar, and Johnson, W.A., Readers and Reading Culture in the High Roman Empire (Oxford 2010)CrossRefGoogle Scholar.

24. Though not among the tropes discussed by Hinds, S., Allusion and Intertext (Cambridge 1998), 1–16Google Scholar, or Barchiesi, A., Speaking Volumes (London 2001), 129–40Google Scholar, one might compare Gale, M., ‘Poetry and the Backward Glance in Virgil's Georgics and Aeneid’, TAPA 133 (2003), 323-52, at 334Google Scholar (‘poets in general face backward towards the past whence they derive their inspiration’), though her emphasis is on the role of the poet rather than on Alexandrian footnotes.

25. ‘Es ist nicht zu entscheiden, ob der attische oder der phokisch-boiotische Fluß oder Flußgott gemeint ist’, Bömer, F., P. Ovidius Naso, Metamorphosen: Kommentar Bd. 3: Buch VI-VII (Heidelberg 1976), 295ad 7.388Google Scholar. Argive Cephisos: RE 11, with Paus. 2.15.5 on Hera and Poseidon.

26. ‘Flußgott in Phokis und Böotien. Seine Enkelin Praxithea gebar dem attischen Könige Erechtheus acht Kinder; s. 6, 679. Die Verwandlung eines dieser Kinder ist sonst unbekannt.’ Haupt, M. and Müller, H.J. (eds.), Die Metamorphosen des P. Ovidius Naso (Berlin, 1885), 240f. ad 7.388Google Scholar. van Proosdij, B.A., P. Ovidii Nasonis Metmorphoseon libri I-XV: textus et commentarius naar de editie van D.E. Bosselaar (Leiden 1951)Google Scholar, also cites Met. 6.639 and explicitly (as Haupt-Müller implicitly) takes nepos in the extended sense of ‘afstammeling’ (descendant) on the basis of Apollodorus’ assertion that Praxithea was Cephisos’ granddaughter. None of these seem aware of the Euripidean version that she was his daughter, which would go some way to strengthening their case.

27. Lycurg, . Leocr. 98–101Google Scholar = Eur. Erechtheus test, ii, Apollod. 3.15.1. Barrett convincingly conjectured ΚηΦισοῦ for the gap in the papyrus before ϰόϱη in Athena's exodos address to Praxithea at Erechtheus fr. 370.63 TGrF.

28. ‘Die Annahme von Haupt-Ehwald…hat nur wegen der vielen Kinder aus dieser Ehe eine zahlenmäßige Wahrscheinlichkeit’, Bömer (n.25 above), 295 ad7.388.

29. ‘Der boiotisch-phokische K[ephisos] ist mit einer reichen genealogischen Tradition ver-knüpft’, Mantis, A., ‘Kephisos IV’, LIMC VI, 13Google Scholar.

30. On the daughters see below. Eteocles: Paus. 934.9.

31. P.M.C. Forbes Irving, Metamorphosis in Greek Myths (Oxford 1990), 313; Paus. 10.6.4.

32. Mantis (n.29 above), 13. The passage of Epaphrodirus is cited by Σ Aesch. Eum. 16b, not 2 as claimed.

33. ΈπαΦϱόδιτος ἐν ύπομνήματι Καλλιμάχου αιτίων β’ Φησί Μελανθοῦς τῆς Δευϰαλί-ωνος ϰαὶ ΚηΦισοῦ τοῦ ποταμοῦ γίνεται Μέλαινα τοὒνομα, Μελαινης δἐ ϰαὶ Ποσειδῶνος ΔελΦός, ἀΦ’ οὖ οί ΔελΦοί (‘Epaphroditus in his commentary on Callimachus’ Aetia 2 says: “To Melantho the daughter of Deucalion and the river Cephisos was born [a daughter] Melaena by name, and to Melaena and Poseidon, Delphus, from whom the Delphians [are descended]’”). Epaphrod. fr. 56 Braswell-Billerbeck = Schal, vet. (M) Aesch. Eum. 16b = Callim. fr. 52 Pf.

34. Thyia, Castalius and Delphus: Paus. 10.6.4 (cf. 10.29.12 and 10.6.3) also records a tradition whereby Delphus is the son of Apollo and one Celaeno. Thyia and Cephisos: Hdt. 7.178.2. Thyia's name and the testimony of Pausanias (10.6.4 again; cf. the Elean festival at 6.26.1) intriguingly link her with Dionysus, but there is no trace of how any tension between this and her Apolline connection might have played out.

35. Pavlock (n.1 above), 50; cf. Binroth-Bank (n.1 above), 142, on ‘die Trauer der Eltern über das Schicksal ihrer Kinder’ (‘the grief of parents over the fate of their children’).

36. An intriguing further possibility is that Ovid's polemical replacement of Delphus the dolphin with Phocus the seal is facilitated by the intermediate figure of the Φώϰαινα, or porpoise, which Aristotle describes as ὃμοιον δελϕῖνι μιϰϱᾧ (‘like a small dolphin’, HA 566b). More intriguing still is Hesychius’ listing of ϕῶϰος as a variant term for Φώϰαινα.

37. ‘Kephisosenkel (Phokos?) Robbe, wohl nur Ov. Met. 7,3.38 f.’ Zgoll, C., Phänomenologie der Metamorphose: Verwandlungen und Verwandtes in der augusteischen Dichtung (Tübingen 2004), 337Google Scholar.

38. On explicit etymologising in the Met., see Michalopoulos, A., Ancient Etymologies in Ovid's Metamorphoses (Leeds 2001), 3Google Scholar; more generally in the ‘Alexandrian’ tradition, O'Hara, J.J., True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay (Ann Arbor 1996), 24–41, 73–75Google Scholar. On ‘markers’ and ‘signposts’, Maltby, R., ‘The Limits of Etymologizing’, Aevum(ant) 6 (1993), 257-75, at 268–70Google Scholar; O'Hara, op. cit., 75-79; Michalopoulos, op. cit., 4f.

39. O'Hara (n.38 above), 79f., with discussion and an extensive list of Virgilian examples on 79-82; on suppressed etymologising in the Met., see Tissol (n.3 above), 173f.; Michalopoulos (n.38 above), 3f.

40. Virg. Aen. 12.872, with Barchiesi, A., ‘Il lamento di Giuturna’, MD 1 (1978), 99–121, at 103 n.4Google Scholar; O'Hara (n.38 above), 241.

41. One need hardly say that ἀϱάχνη is ‘derived’ from Άϱάχνη only in the aetiologising fiction, and in reality the etymology operates in the opposite direction.

42. regis Romani primam qui legibus urbemlfundabit (‘[I recognise the white hair and chin] of the Roman king who will establish the city for the first time with laws’, Virg. Aen. 6.810f.)

43. Kenney, E.J., ‘Explanatory Notes’, in Melville, A.D. (tr.), Ovid's Metamorphoses (Oxford 1986), 417Google Scholar (cf. Kenney [n.11 above] 264 ad 7.290), sees a suppressed bilingual wordplay in the very next myth of the praeteritio, where in aere natum (‘son in the air’) supposedly calques the Boeotian name for bee-eater, άέϱοπ1ος (‘air-face’), into which Eumelus’ son was transformed (Boios Ornith. 3 ap. Ant. Lib. 18; Papathomopoulos, M. [ed.], Antoninus Liberalis: Les Métamorphoses [Paris 1968]Google Scholarad loc., insists that Boios would have used the Ionic form ἠέϱοπος). This would be a very close parallel for Phocus (though ἀέϱοπος was only the name of the bird; the boy's human name was Botres) but the case for an etymology is weakened both by absence of any parallelism between natus (‘son’) and ὤψ (‘face’) and by the recent use of the same or very similar phrases in the same sedes to allude to the metamorphosis into birds of Cerambus (in aera, ‘into the air’, 7.354) and Cycnus (in aere, ‘in the air’, 7.379); cf. Combe's trepidantibus alis (‘with quivering wings’, 7.382.

44. Virg. Aen. 8337-41, with O'Hara (n.38 above), 209.

45. One might compare Heraclid. Lemb. fr. 35, which etymologises Phocaea either from an eponymous hero Phocus or from the number of seals in the area. For the connection in Latin poetry, Holmes, N., ‘Notes on Lucan’, CQ 41 (1991), 272-74, at 273CrossRefGoogle Scholar, suggests that the name (or possibly ethnonym) of the Massilian diver Phoceus at Luc. 3.696f. also evokesphoca.

46. Φωϰίς• χώϱα πεϱὶ τὸν Παϱνασσόν. Έϰαταῖος Εὐϱώπηι. ἀπὀ Φώϰου τινός. Hecat. FrGH 1 F 114.2.

47. RE (1) XX.1.497.21-498.60; LIMC II, 396. Son of Ornytion: Paus. 2.43 (Ποσειδῶνος δἐ έπίϰλησιν, ‘but nominally the son of Poseidon’), 2.293; Ornytus: Scymn. 487. According to Σ b Il. 2.517, Phocus was the son of Ornytus and the father of Ornytion, and it was the former who settled in the region which later took its name from his son.

48. Met. 11.345-409. The connection between this Phocus and Phocis is also made, albeit obliquely, since it is a Phocian herdsman (armenti custos Phoceus Onetor, 348) who reports the wolf's attack.

49. τὀ δέ ὂνομα πϱοϋπῆϱχεν ἤδη τᾔ χώϱα, Φώϰου τοῦ Όϱνυτίωνος γενεᾷ πϱότεϱον ἐς αὐτὴν ἐλθόντος. ἐπί μἐν δὴ Φώϰου τούτου ή πεϱί Τιθοϱέαν τε ϰαὶ Παϱνασσὸν έϰαλεῖτο Φωϰίς• έπὶ δὲ τοῦ Αἰαϰοῦ ϰαὶ πάσιν ἐξενίϰησεν, ὃσοι Μινὑαις τέ εἰσιν Όϱχομενίοις ὃμοϱοι ϰαὶ ἐπὶ Σϰάϱϕειαν τὴν Λοϰϱῶν ϰαθήϰουσι (‘The name [Phocis] already pre-existed in the land, since Phocus the son of Ornytion had come to it a generation earlier. In the time of this latter Phocus, then, the area around Tithorea and Parnassus was called Phocis; but in the time of [Phocus] the son of Aeacus it also came to prevail in all areas, both those which border the Minyae of Orchomenus and those which reach to Locrian Scarpheia.’, Paus. 2.29.3).

50. μίγνυται δὲ αὖθις Αἰαϰὀς Ψαμάθῃ τῇ Νηϱέως εις ϕώϰην ἠλλαγμένῃ διὰ τὸ μὴ βούλεσθαι συνελθεῖν, ϰαὶ τεϰνοῖ παῖδα Φῶϰον (‘Aeacus had sex with another partner, Psamathe the daughter of Nereus, transformed into a seal because she did not want to sleep with him, and fathered a child Phocus’, Apollod. 3.12.6); cf. the almost identically worded Σ A. Eur. Arulr. 687). In these passages, the etymology is only implicit, but heavily so.

51. McInerney, J., The Folds of Parnassos: Land and Ethnicity in Ancient Phokis (Austin 1999), 136Google Scholar.

52. Cf. Ahl, F., Metaformations: Soundplay and Wordplay in Ovid and Other Classical Poets (Ithaca 1985), 101–03Google Scholar, who links the dolphins and seals (delphines and phocae) featured in the flood at Met. 1.299-303 with the setting of the subsequent Deucalion and Pyrrha episode at Delphi in Phocis. One of Ramus's anonymous readers makes the attractive suggestion that there ‘may be an extra grammatological fillip’ in the way that (to adopt Ahl's typography of wordplay) PHoKE ‘looks back’ to KEPHisos.

53. I thought I had coined this term in my own undergraduate lectures on Ovid, but found my-self anticipated by Barchiesi in, appropriately enough, ‘Future Reflexive’ (n.1 above, 350), with his formulation ars patet arte sua.

54. Barthes, R., ‘L'effet de réel’, Communications 11 (1968), 84–89CrossRefGoogle Scholar.

55. OLD s.v. 2 (‘rounded and protuberant in outline’) mainly qualifies inanimate objects. Cels. 8.1.18 on the shape of the ends of the humerus is not really a parallel for Proteus’ well-fed seals. ζατϱεϕής of seals: Od. 4.451; pigs: 14.19; goats: 14.106; oxen: Il. 7.223.

56. This interpretation is supported by the sharp antithesis set up with the graciles…capellae (‘slender goats’, 1.299) who grazed the grass before the flood.

57. Faunus: F. 2.346; Ilia: Am. 2.14.15; sea: e.g. F. 3.595, Tr. 1.2.24; anger (OLD s.v. 4): Met. 2.602,8.437,13.559; pride (OLD s.v. 5): 1.754,8.396,495. Cf. tumentem at 15.755.

58. One other possible case may be the description of Neptune's (and Apollo's) taking human form to build the walls of Troy, with a play on the swell of the sea and the pride of the god: cumque tridentigero tumidi genitore profundi/mortalem induitur formam (‘and with the trident-wielding father of the swelling deep put on mortal form’, 11.202f.).

59. tumidus of snakes: Cic. Vat. 4, Virg. Aen. 2.472, Ov. Met. 10.313, Tr. 5.2.14, Sen. Med. 688.

60. Commentators such as Allen and Richardson, and the standard studies of the hymn, are curiously silent on the odd usage, but interestingly for Ovid's calquing of it as tumidus, both Evelyn-White and West render it ‘bloated’. The humorous tone of W. Beck's entry at Lfrg E s.v. ζατϱε-ϕής—‘of domest, animals tended by herders…and extended to serpent (Python) feeding on herds (and herders)’—rather suggests that he is reneging on the challenge of accounting for this anomalous usage.

61. Aglauros: nec lapis albus erat: sua mens infecerat ollam (‘nor was the stone white: her heart had stained it’, 2.832); oleaster: arbor enim est, sucoque licet cognoscere mores./quippe notam linguae bacis oleaster amaris/exhibet: asperitas uerborum cessit in ilia (‘For he is a tree, and you can recognise his character from its juice. For the oleaster shows the mark of his tongue in its bitter berries: the harshness of his words passed into them’, 14.524-26); Anaxarete: paulatimque occupat artus,/quod fuit in duro iam pridem pectore, saxum (‘gradually there took over her limbs that which was already long since in her hard breast: stone’, 14.757f.).

62. As well as the works listed in n.3 above, see esp. Häge, H., Terminologie und Typologie des Verwandlungsvorgangs in den Metamorphosen Ovids (Göppingen 1976), 85–99Google Scholar; Solodow (n.17 above), 174-86.

63. Hyg. fab. 253 mentions incest with both his mother and his daughter Cyllene. Since Cyllene is the specified location of Ovid's myth, there may be a hint at another metamorphosis. Bömer (n.25 above) ad loc. canvasses some possible parallels, including a possible doublet from Boios’ Ornithogonia, preserved at Ant. Lib. 5, in the story of Aegypius and his mother Bulis. The latter's name would not be a difficult emendation of Blias, the name of Menephron's mother in Hyginus (Papathomopoulos [n.43 above], 81). For the thematic significance of the story, see Pavlock (n.1 above), 59.

64. See n.31 above.

65. Niobe: Met. 6.146-312, with her ‘arrogant eyes’ (oculos…superbos) at 169.

66. Pavlock (n.1 above), 59.

67. On the distinction between Alexandrian and Callimachean poetics and the way in which they were tendentiously re-shaped by neoteric and Augustan poets, see esp. Hunter, R., The Shadow of Callimachus (Cambridge 2006)CrossRefGoogle Scholar; Knox, P.E., ‘Cicero as a Hellenistic Poet’, CQ 61 (2011), 192–204CrossRefGoogle Scholar.

68. Achelous: Met. 8.548f., with Hinds, S., ‘Generalizing about Ovid’, Ramus 18 (1987), 4–31, at 19Google Scholar, and Barchiesi (n.24 above), 50-55; Glaucus: Met. 14.4, with Myers, K.S. (ed.) Ovid Metamorphoses Book XIV (Cambridge 2009), 53f.Google Scholar; storm: Tr. 1.2.23f., with Ingleheart, J., ‘Ovid Tristia 1.2: High Drama on the High Seas’, G&R 53 (2006), 73–91, at 87fGoogle Scholar.

69. Of the numerous generic analyses of this passage and its relationship with the Aetia prologue and Am. 1.1, see esp. the seminal account of Nicoll, W.S.M., ‘Cupid, Apollo and Daphne (Ovid, Met. 1. 452 ff.)’ CQ 30 (1980), 174–82CrossRefGoogle Scholar, and the innovative perspective of Morgan (n.17 above).

70. On Ovid's revisionist depiction of Apollo in the amatory works, see Armstrong, R., ‘Retiring Apollo: Ovid on the Politics and Poetics of Self-Sufficiency’, CQ 54 (2004), 528–50CrossRefGoogle Scholar. On Apollo in the Met., Le Bonniec, H., ‘Apollon dans les Métamorphoses d'Ovide’, in Frécaut, J.M. and Porte, D. (eds.), Journées ovidiennes de Parménie (Brussels 1985), 145–74Google Scholar.

71. On such contextual glosses, see esp. Rengakos, A., ‘Homerische Wörter bei Kallimachos‘, ZPE 94 (1992), 21–47Google Scholar; Sistakou, E., ‘Glossing Homer: Homeric Exegesis in Early Third Century Epigram’, in Bing, P. and Bruss, J.S. (eds.), Brill's Companion to Hellenistic Epigram Down to Philip (Leiden 2007), 391–408CrossRefGoogle Scholar. On νέποδες (not only in Callimachus): Rengakos, op. cit., 41.

72. Presumably this refers to fins or flippers as quasi-feet, rather than to actual webbed feet, pace LSJ, which would be appropriate to neither seals nor fish.

73. Nie. Alex. 468,485; Satyrius AP 6.11.6; Opp. Hal. 3.441,4,468; Cyn. 1.384. A third alternative, only attested in the fragments of Apion's first-century CE Glossae Homericae but probably deriving from much earlier sources, most likely Aristarchus, is to derive it from νη- privative + πούς to mean ‘those without feet’. Apollonius the Sophist, who preserves Apion's gloss, also dismisses it curtly: τὸ μὲν οὖν ἂποδες ψεῦδος• ἒχουσι γὰϱ πόδας αί ϕῶϰαι (‘the gloss “footless” is certainly false; for seals do have feet’, Apollon. Lex. s.v. νέποδες).

74. E.g. εἶμι δ’ ές αὐγὰς/ήελίου μετόπισθε, τεοῖς νεπόδεσσιν έτοίμη (‘I shall go afterwards to the sun's rays, ready for your descendants’, Ap. Rhod. 4.1744f.); ἀθάνατοι δὲ ϰαλεῦνται έοὶ νέποδες (‘immortal are called his [Heracles’] descendants’, Theoc. 17.25); ό Κεῖος Ύλίχου νέπους (‘the Cean descendant of Hylichus’ [i.e. Simonides], Callim. Iamb. fr. 222 Pf.); τοῦτο μὲν οὖν ϱέξαντες ἀολλέες ἠγεϱέθοντο I βαυϱιόθεν βϱιαϱοὶ Γοϱγοϕόνου νέποδες (‘after doing this, they gathered in a throng from their homes, the mighty offspring of the Gorgon-slayer’, Cleon. Sic.fr. 1).

75. πουλὺ θαλασσαίων μυνδότεϱοι νεπόδων (‘much more dumb than the marine flippered ones/offspring of the sea’, inc. sed. fr.295.2 SH = 533Pf.). The even rarer μυνδός qualifies ίχθῦς (‘fish’) at Soph. fr. 1072 Radt and ἒλλοψ (‘fish’) at Lyc. 1375, which would encourage the reader to take νέποδες as meaning ‘fish’. Yet the otherwise redundant θαλασσαίων might encourage her to understand ‘offspring of the sea’ rather than the banal ‘fish of the sea’. If Lloyd-Jones and Parsons are right to place fr. 496 Pf. (λαοὶ Δευϰαλίωνος ὃσοι γενόμεθα, ‘the people of Deucalion as many as we are’) immediately before it and accept Naeke's supplement γενέθλης (‘of the stock’) at line-end, then the antithesis with both the latter and γενόμεθα might further encourage the reader towards this interpretation. Interestingly, LSJ unhesitatingly cites this passage as an instance of the meaning ‘fish’, while Rengakos (n.71 above, 41) with equal confidence groups it with the other Callimachean passages where it means ἀπόγονοι (‘descendants’). Pfeiffer documents the rival interpretations, but glosses ‘progenies maris’ (‘offspring of the sea’).

76. Paul Roche suggests to me (pers. comm.) that Ovid may, like Callimachus, be playing on the ambiguity of νέπους, since he does connect it with a seal. This is possible, and it is true that Ovid does not studiedly separate nepos from any association with the sea, as Theocritus and Cleon do with νέπους. However, unlike Greek poets, Ovid needs some reference to a seal to activate the association of nepos with νέπους, which the reader probably would not otherwise make. Moreover, unlike νέπους, nepos simply cannot mean ‘flippered’, even in such a bilingual context. By the very act of associating the two words, I feel Ovid is quite strongly aligning himself with the ‘offspring’ camp.

77. Our passage is the only instance of a Greek ending given in TLL X.1.2046.49-2047.14. It seems very unlikely that the several instances of the accusative plural phocas (Virg. Geo. 4.395, etc.) would be ‘felt’ as transliterating ϕώϰας. It should be acknowledged that uitulus (with or without the marinus) for ‘seal’ is not attested before Col. 63.2 or possibly (assuming a Neronian date) Calp. Eel. 7.66. Even if it did occur in earlier, non-extant texts, it is overwhelmingly likely that phoca was the poetic vox propria and hence Ovid's use of it would not be especially marked. His use of phoce, however, certainly is.

78. The most sustained, albeit rather overstated, argument for the use of invention by the writers of ancient commentaries, mythographic handbooks and other paratextual aids is that in Cameron, A., Greek Mythography in the Roman World (Oxford 2004)Google Scholar.

79. Nisbet, R.G.M. and Hubbard, M., A Commentary on Horace Odes Book II (Oxford 1978), 93Google Scholar. The claim that Septimius was an eques Romanus cannot in any obvious way be deduced from the context and hence suggests that the commentator has additional knowledge, but it may well be that it is intended to produce precisely that effect as only an ex nihilo invention could, a commentator's own effet de réel.