¹ P. Oxy. 2369 (Carden 52 ff., Sophocles F 269a Radt). The identification would in any case be plausible in view of the subject-matter, and is put beyond reasonable doubt by the phrase ἰώ, γ⋯, θε⋯ν [(1.51) in view of Philodemus' statement (de piet. p. 23 Gomperz; cf. S. F 615) that Sophocles ⋯[ν 'Iν⋯]χῳ τ⋯ν Γ⋯ν μη[τ⋯]ρα τ⋯ν θε⋯ε φη[ϲιν].

² Pearson, A. C., The Fragments of Sophocles, I (Cambridge, 1917), 197 ff.Google Scholar

³ The only evidence for the presence of Iris is the comment of the scholiast on Ar. Av. 1203 comparing Pisthetaerus' question to Iris (ὂνομα δ⋯ ϲο⋯ τί ⋯ϲτι; πλοῖον ἢ κμν⋯;) with Hermes' words in the Inachus ⋯πἱ τ⋯ϲ Ἲριδοϲ (F 272) ‘γυν⋯ τ⋯ϲ ἢδε, ϲυλ⋯ϲ ’Αρκ⋯δοϲ κυν;⋯ϲ;’ (?). M. Mayor (Roscher ii. 347, 12) argued that Ἲριδοϲ was a mistake, natural enough given the Aristophanic context, for Ἰοῡϲ; his grounds seem somewhat flimsy, since so little is known about the play, but Iris' function is not obvious, and his doubts should be borne in mind.

⁴ In his edition of Aristophanes', Plutus (Harlingen, 1744), p. 248Google Scholar (on sch. Ar. Pl. 727); he did not reveal his reasons. Calder well emphasises the weakness of the arguments which others have adduced in support of this hypothesis, but overstates the case against it. Wilamowitz's view that the Inachus was pro-satyric, like the Alcestis, depends on his own decidedly speculative reconstruction (Euripides Herakles, I (Berlin, 1889), 88 n. 53)Google Scholar. For a survey of the controversy see Sutton 25 ff. Radt, (‘Sophokles in seinen Fragmenten’, Sophocle: Entretiens sur l'antiquité classique 29 [Fondation Hardt, Vandoeuvres-Geneva, 1982], 186 ff.)Google Scholar urges a measure of positive discrimination in favour of Sophoclean satyrs: out of the (probable) total of 122 Sophoclean titles known to us only 13 are expressly designated as satyr-plays, less than half the figure we should expect (given a ratio of three tragedies to one satyr-play), and it is much more likely that some of the missing satyr-plays lurk among the transmitted titles than that they failed to reach the Alexandrian library.

⁵ P. Tebt. 692, second century B.C. (Carden 52 ff., Sophocles F 269c Radt). We have no direct proof that this fragment is satyric; the main grounds for so regarding it are the brevity and excitement of the choral songs and the chorus' disorderly reaction to the invisible Hermes playing on the syrinx.

⁶ We cannot pretend to be completely informed about the range of dramatic compositions dealing with this theme which might have circulated in Ptolemaic Egypt; Körte, suggested (APF 11 [1935], 253–4) that the papyrus might represent that long-felt want of students of the Ars Poetica, a Hellenistic satyr-playGoogle Scholar.

⁷ Sutton, though himself convinced that the play was satyric, well adverts (53–4) to ‘the seriousness of the thematic issues raised’ and emphasises the real suffering of the characters; he also (51, 78) calls attention to stylistic features which seem nearer to tragedy than to satyric drama.

⁸ For the sake of brevity I shall assume this to be so; for other possibilities see Carden's note on 29 ff. (pp. 61–2).

⁹ εἰϲελθ⋯ντοϲ: ⋯θ⋯λοντοϲ Sutton, perperam.

¹⁰ 2. 1. 3: τα⋯την (sc. Ἰώ) ίερωϲ⋯νην “Ηραϲ ἒχουϲαν Ζεὺϲ ἓφθειρε, φωραθε⋯ϲ δ⋯ ὑφ’ ‘Ηραϲ τ⋯ϲ μ⋯ν κ⋯ρηϲ ⋯ψ⋯μενοϲ εἱϲ βο⋯ν μετε⋯ρφωϲε λευκ⋯ν, ⋯πωμ⋯ϲατο δ⋯ ταὑτῃ μ⋯ ϲυνελθεῖν. Cf. Hyg, . fab. 145: hanc Iuppiter dilectam compressit et in vaccae flguram convertit, ne Iuno eam cognosceretGoogle Scholar.

¹¹ ‘Negro’ is convenient, though possibly too specific; however, black people of non-negroid type must have been much less familiar to fifth-century Athenians.

¹² Sutton 56. Against the view that Hermes accompanied Zeus on his visit to Inachus see Carden's note on 34 sqq. (pp. 63–4), Seaford 23 f.

¹³ I assume that Plautus would have neither expected in his audience a knowledge of Euripidean tragedy in its authentic form nor mechanically reproduced a joke from his Diphilean original without regard to its potential effectiveness in the Roman theatre. See further Stewart, Z., ‘The Amphitruo of Plautus and Euripides' Bacchae’, TAPA 89 (1958), 348–73, esp. 358 ff.Google Scholar, Webster, T. B. L., The Tragedies of Euripides (London, 1967), 92–3Google Scholar, Trendall, A. D. and Webster, T. B. L., Illustrations of Greek Drama (London, 1971), 76Google Scholar.

¹⁴ We might wonder whether the legend of Croesus' miraculous deliverance (B. iii 53 ff., Hdt. 1. 87. 2) suggested this development to Euripides. We may note that though Herodotus professes uncertainty about Cyrus' motives in committing Croesus to the flames (1. 86. 2 ⋯ν ν⋯ῳ ἒχων εἲτε δ⋯ ⋯κροθ⋯νια τα⋯τα καταγιεῖν θε⋯ν ὂτεῳ δ⋯, εἲτε κα⋯ εὐχ⋯ν ⋯πιτελ⋯ϲαι θ⋯λων εἲτε κα⋯ πυθ⋯μενοϲ τ⋯ν Κροῖϲον εἶναι θεοϲεβ⋯α το⋯δε εἲνεκεν ⋯νεβ⋯βαϲε ⋯π⋯ τ⋯ν πυρ⋯ν, βουλ⋯μενοϲ εἰδ⋯ναι εἲ τ⋯ϲ μιν δαιμ⋯νων ῥ⋯ϲεται το⋯ μ⋯ ζ⋯ντα κατακαυθ⋯ναι) his subsequent narrative rules out the first two possibilities, since Cyrus could hardly repent of what he had undertaken as a religious duty; this cruel and unusual punishment is thus seen as a quasi-judicial ordeal by fire. This connection of ideas explains Thucydides' air of scepticism in reporting the downpour which frustrated Archidamus' attempt to fire Plataea (2. 77. 6); this timely storm could easily be construed as supernatural vindication of the Plataean cause.

¹⁵ The case for supposing Zeus to have appeared in Aeschylus' Psychostasia is shaky: see further Taplin, O., The Stagecraft of Aeschylus (Oxford, 1977), 431–2Google Scholar.

¹⁶ On the descendants of Zeus and Io cf. A. Su. 312 ff., Apollod. 2. 1. 4.

¹⁷ A. Su. 497 f.

¹⁸ Su. 154, 279 ff., 719, 745, 887.

¹⁹ Cf. 2. 55. 1; 57. 2. This misconception is a long-standing embarrassment to Herodotus' interpreters; the hypothesis that he was confused by the presence of negro slaves and half-breeds (so How and Wells) does little credit to his common sense and powers of observation. He compounds the error by remarking on the strong physical resemblance of Egyptians and Colchians; as with Pindar's κελαινώπεϲϲι Κ⋯λχοιϲι (P. 4. 212) we evidently have here a reflection of the old conception of Colchis as the most easterly point on a flat earth, where men are toasted by the fierce rays of the rising sun (cf. Hdt. 3. 104. 2); but contemporary reality may be found in Hippocrates, who speaks of the yellow skin of those who live on the Phasis, (Aer. 15)Google Scholar. See further Fehling, D., Die Quellenangaben bei Herodot (Berlin, 1971), 15–17CrossRefGoogle Scholar.

²⁰ TGrF Adespota F 161 Kannicht-Snell. Vase-paintings of Heracles and Busiris commonly represent negroes in the king's retinue, and sometimes it is difficult to avoid the suspicion that the king himself is supposed to be one; but since negro slaves were common enough in Egypt and there are obvious artistic advantages in thus accentuating the differences between Heracles and his opponent, it would be unfair to treat this material as evidence of real misconception. See further Froidefond 88 ff.

²¹ See further Snowden, F. M. Jr, Blacks in Antiquity (Cambridge, Mass., 1970)Google Scholar, Before Color-Prejudice (Cambridge, Mass., 1983)Google Scholar.

²² Il. 1. 423 ff., 23. 205–7, Od. 1. 22–26. That the poet and his audience understood Αἰθ⋯οψ to mean ‘Burnt-face’ can hardly be doubted: see further LfgrE, Frisk, GEW, Chantraine, Dict, ëtym. s.v.

²³ Lyc, . Alex. 537Google Scholar with sch. ad loc.; cf. Eust. 1385. 62(on Od. 1. 22); see further Cook, A. B., Zeus, I (Cambridge, 1914), 195, 289 f.Google Scholar I imagine Lycophron derived his knowledge of this obscure title from literature rather than from personal study of the antiquities of Chios.

²⁴ The persistence of the Ethiopian character for piety may not be unconnected with the variable location of Ethiopia; see further Berger, E. H., Mythische Kosmographie der Griechen (Leipzig, 1994), 22–24Google Scholar, Lesky, A., ‘Aithiopika’, Hermes 97 (1959), 27 ff.Google Scholar, Dihle, A., Umstrittene Daten (Cologne, 1965), 65 ff.CrossRefGoogle Scholar

²⁵ Seaford has argued that there is a further, theological, significance, and that Zeus is here Zeus Chthonius, his sombre hue deriving from his role as lord of the dead; in developing this conception here Sophocles may, he argues, have been influenced by what he knew of Osiris, with whose spouse, Isis, Io was certainly identified in the Hellenistic age (Call. Ep. 57. 1, cf., perhaps, fr. 383. 12 ff.). Obviously we cannot rule out the possibility that Sophocles' mind moved on some such lines, but I find it hard to believe that he could have expected his audience to follow him. If we may take Herodotus as representing general Greek conceptions about Egyptian religion, we may note that (a) he does not equate Osiris with Zeus or Hades, but with Dionysus (2. 42. 2; 144. 2), (b) he does not suggest any connection between Apis (= Epaphus) and either Osiris or Isis, (c) though he alludes (2. 41. 2) to the iconographical resemblance which favoured the identification of Io and Isis (and which, almost certainly, accounts for the extension of Io's wanderings from Euboea to Egypt), he is clearly unaware that they might be identified; for him Isis is Demeter. That this theologically much more satisfactory equation was already well established is suggested by the presence of a figurine of Isis in the grave of a rich lady buried at Eleusis c. 800 B.C. (the so-called ‘Isis-grave’), where the unusually copious grave-offerings have been thought to suggest that the lady herself served Demeter as priestess of the local cult (see AEphem. 1898, pi. 6, BSA 64 [1969], 145 f., pi. 35b, Coldstream, J. N., Geometric Greece [London, 1977], 79 f.)CrossRefGoogle Scholar. See further Roscher ii. 270 s.v. Io (Engelmann), RE ix. 1737 (Eitrem).

²⁶ We may compare Herodotus' observation that Egypt is full of doctors (2. 84); see further Froidefond 58 ff.

²⁷ Fr. 726 Mette; the trait is well illustrated by Herodotus' tale of Rhampsinitus and the dishonest architect (2. 121); see further Gow on Theoc. 15. 49, Froidefond 93 ff., 225 f.

²⁸ It would be pleasant to suppose that the condescending and sententious couplet F 282 was addressed by Inachus to his divine guest.

²⁹ Cf. Od. 1. 169 ff., 3. 69 ff., 7. 236 ff.

³⁰ Cf. Od. 1. 319 f., 410 (οἷον ⋯ναῗξαϲ ἂφαρ οἲχεται), 3. 371 ff.; for cases where there is no motive for haste cf. Il. 13. 63–72, 18. 616, 24. 121.

³¹ See above, p. 293.

³² We have perhaps another parallel for this miraculous bounty in the unexpected wealth which the golden shower brought to the palace of Acrisius in Euripides' Danae (frr. 324–8); see further Schmid-Stähelin i, 3. 596.

³³ Both τοι⋯νδ' and ⋯μόν have been suspected.

³⁴ Or, perhaps, ⋯νενφε⋯αϲ χ⋯ριν means ‘so that Inachus will have no cause to find fault’. The effect of this miracle would be enhanced, as would the merit of Inachus' hospitality, if Wilamowitz (loc. cit. n. 4) was right in supposing that at some point in the play the land of Argos was afflicted by a plague or blight which brought famine (cf. F 276, 284, 286, 294); he put this disaster after Zeus' union with Io, but it might be worth considering the possibility that Argos was gripped by famine at the opening of the play. But the evidence for this blight is not absolutely unambiguous.

³⁵ Cf. Od. 17. 485–7. The New Testament well illustrates the prevalence of such beliefs. Apart from the reception accorded to Paul and Barnabas at Lystra in Phrygia Galatica (Act. Ap. 14. 11 f.), we may note that the author of the Epistle to the Hebrews supports his (or her) injunctions about hospitality (13. 2) not by an appeal to general humanitarian considerations nor, as we might expect in a Christian context, by reference to the teaching embodied in the parable of the sheep and the goats (Ev. Matt. 25. 31 ff.), but by an allusion to a strange episode in the life of Abraham, (Gen. 18–19)Google Scholar: δι⋯ τα⋯τηϲ (sc. τ⋯ϲ φιλοξεν⋯αϲ) ἒλαθ⋯ν τινεϲ ξεν⋯ϲαντεϲ ⋯γγ⋯λονϲ.

³⁶ For specific references to the divine visitors as foreign magicians see E. Ba. 234, Pl. Am. 1043. Friedrich, W. H., Euripides u. Diphilos, Zetemata 5 (1953), 271Google Scholar, offers some suggestive observations on the resemblances between Pentheus and Amphitruo; see also Stewart op. cit. (n. 13).

³⁷ ‘The transformation of Io, Ox. Pap. xxiii 2369’, CR n.s. 10 (1960), 194–5 (= Collected Papers [Cambridge, 1969], 137–8)Google Scholar.

³⁸ She compared E. Hel. 375–80, where Callisto, having been transformed into a bear, is said to have a ϲχ⋯μα λεα⋯νηϲ. On this interpretation Carden well comments ‘(i) The Euripides passage is corrupt, and therefore shoddy “support” for our passage, (ii) it is hard to accept this shift in meaning’.

³⁹ Il, 4. 111, 15. 261, Od. 8. 260.

⁴⁰ A.P. 6. 295. 5 = HE 2982. See also Frisk, GEW, Chantraine, Dict. étym. s.v. λεῖοϲ.

⁴¹ So Miss Dale, who supposed that Io had been spinning (cf. λινεργ[ (43)). This meets a difficulty raised by Lobel, who, finding it ‘perhaps unreasonable to suppose that Io, even after her change, climbed on the furniture’, suggested that θρ⋯νοϲ was here used with the unusual sense of ‘floorboard’ (cf. LSJ s.v. θρ⋯νοϲ ii 1).

⁴² So Seaford, though he adopts a rather different approach otherwise.

⁴³ Thus Apuleius, (Met. 3. 24)Google Scholar anticipates any queries on this point by making his hero undress before applying the ointment which effected his transformation. With outdoor metamorphoses we easily imagine wind, water or briars removing the victim's clothing if it is not explicitly said to be dissolved as part of the transformation process.

⁴⁴ See further Chantraine, P., ‘A propos de l'adverbe ionien λεἰωϲ, λ⋯ωϲ’, Glotta 33 (1954), 25 ff., esp. 35 f.Google Scholar

⁴⁵ κ 8 Nachmanson.

⁴⁶ [Hp.] Epid. 2. 3. 1. ⋯γ⋯νοντο δ⋯ ⋯ν τοῖϲι θερινοῖϲι πυρετοῖϲι…τρηχ⋯ϲματα ⋯ν τῷ χρωτ⋯, κερχνώδεα, τοῖϲιν ὑπ⋯ κων μ⋯λιϲα εἒκελα ⋯ναδ⋯γμαϲιν.

⁴⁷ The text is far from secure. Ἰοῡϲ is Elmsley's emendation for ἰχθ⋯ϲ; he also deleted ῷ to restore a trimeter. The slightly surprising word-order τ⋯ϲ ⋯ποταυρυν⋯νηϲ φηϲ⋯ν Ἰοῡϲ, though easily remediable, may be a symptom of something more seriously amiss; χελώνηϲ has also been suspected. See further Carden (pp. 66–8), Luppe, W., ‘Die Verwandlung der Io in einem Sophoklesverweis bei Erotian (Fr. 279 P.)’, Philologus 120 (1976), 296–9CrossRefGoogle Scholar.

^47a For an ingenious, but to my mind unconvincing, attempt to enlist Tryphiodorus ’description of the Trojan horse (87–9) in support of Carden's interpretation see Livrea, E., ZPE 52 (1983), 40 ff.Google Scholar

⁴⁸ Lobel's account of the dilemma could not be bettered: ‘I do not myself find particularly convincing the interpretation of these words as “a harsh excrescence of a tortoise (i.e. like the carapace of a tortoise) rises up”, with reference to the sprouting of horns from Io's head, but to take them as meaning “a harsh croaking of the lyre resounds” is to reject the part of Erotian's testimony which it is inconvenient to accept while retaining the part which falls in with our preconceptions’.

⁴⁹ Op. cit. n. 47.

⁵⁰ Contrast the scholiastic agonising over Lycophron's application to Io of the epithet ταυροπ⋯ρθενοϲ (Alex. 1292); ϲ⋯λοικον φα⋯νεται…οὐ γ⋯ρ ⋯ρρεν⋯θηλυϲ ἦν ὣϲτε λ⋯γειν αὐτ⋯ν ταυροπ⋯ρθενοϲ' οὐ γ⋯ρ τα⋯ροϲ ⋯γ⋯νετο, ⋯λλ⋯ βο⋯ϲ. Lycophron's Io is to some extent assimilated to Isis and his odd coinage (as it must surely be) may be justified as an allusion to Isis' hermaphroditic role in the procreation of Harpocrates; see further S. West, ‘Lycophron on Isis’, JEA (forthcoming).

⁵¹ E. Med. 187–8 κα⋯τοι τοκ⋯δοϲ δ⋯ργμα λεα⋯νηϲ⋯ποταυρο⋯ται δμωϲ⋯ν (picking up ὂμμα ταυρουν⋯νην 92). The verb is also used figuratively in the sense of ‘rage like a bull’ by Cyril of Alexandria (Jo. 3. 6 [4. 311c], glaph. Gen. [1. 16b], Soph. 41 [3. 619c]).

⁵² See above, n. 47.

⁵³ Cf. Ar. V. 429, 1292.

⁵⁴ Mentioned among Io's haunts at A. Pr. 676.

⁵⁵ Cf. P. Nem. 10. 5, P. 4. 14, 56; see further Froidefond 107 ff. We may feel rather more sympathetic towards Herodotus' insistence on the unrivalled antiquity of Egypt, based though his view is on false premises (see in particular 2. 142), if we bear in mind the obvious appeal of this wildly unhistorical alternative sanctioned by the authority of great poets.