¹ Euripides, Orestes (Warminster: Aris and Phillips, 1987).

² See the apparatus and/or notes in the edition at lines 164, 201, 349–51, 367, 528, 609, 613, 827, 838, 839, 986–7, 1015, 1046, 1135, 1227–30, 1305, 1395, 1397, 1431, 1469, 1473, 1512, 1589–90, 1608.I discussed a few of these passages, and certain others, in BICS2S (1981), 68–70. Some of my proposals were communicated to Sir Charles Willink in time for him to comment on them in his admirable edition (O.U.P. 1986).Google Scholar

³ For the application of this concept to Homer see Schadewaldt's, W.Iliasstudien(2nd edn. 1943) passim;Google ScholarGoold, G. P., Illinois Classical Studies 2 (1977), 1–34;Google Scholarfor Hesiod, , my Hesiod, Theogony (1966), 206, 289, and Hesiod, Works and Days (1978), 44, 55, 58, 268, 326,Google Scholar and my remarks in C, Brillante and others (ed.), Ipoemi epici rapsodici non omerici e la tradizione orale(Padova, 1981), 65–7;Google Scholar for Sophocles (OC 1300), BICS 31 (1984), 188; for Euripides (El. 520–3+527–4), BICS 27 (1980), 17–20.

⁴ The sort of citizens' doctor described in PI. Leg. 720d, who

⁵ Willink in his note on 421–3 says that after 402 ‘Menelaus saw no pressing need to pose the further question “how long ago was that?” Orestes' replies diverted him to other matters of interest’. But we are concerned not with Menelaos’ mind but with Euripides': why did he not complete the questioning about the duration of Orestes’ madness before going on to the ‘other matters of interest’?

⁶ Willink ingeniously emends 424 to but is still forced to assume a dislocation after 423. He suggests placing 412–13 here. 412 certainly goes quite well after 423, but I cannot see that 424 becomes any more relevant if preceded by 413. I speculate that Euripides' first draft may have contained the sequence 401–2, 421–3, 412–17, 424.

⁷ According to one version (Dictys 6.2, Hyg. Fab. 117) it was Oeax who prompted Clytaemestra to murder Agamemnon, and Paus. 1.22.6 describes a fifth-century painting on the Acropolis which showed Pylades killing ‘the sons of Nauplius’ as they came to Aegisthus’ aid.

⁸ See Willink, 178f.

⁹ I see no necessity for the transposition of 585–90 to follow 578 (Willink, following a suggestion from J. Diggle so to transpose 585–7 with deletion of 588–90). The thought of Clytaemestra's wickedness is there throughout 572–84, even if less central in 579–84, and 585 follows well enough. The renewed address to Tyndareos in 585 leads on to 588 591

¹⁰ Otherwise in drama only at Soph. Tr. 210, and there not guaranteed by metre. See Stinton, T. C. W., BICS 22 (1975), 90.Google Scholar

¹¹ No real parallel in tragedy. Cf. however Pind. Isth. 7.8–9 (dub.), Bacchyl. 3.22

¹² Porson read (nominativus pendens; the is omitted by some manuscripts), and Hermann Another possibility is is suitable (after 588 - if 588–90 are accepted as genuine), as the appeal to Apollo's authority is the climax of Orestes’ argument. Denniston's reservations about continuative in tragedy (Greek Panicles², 155) seem excessive, and he has not noted all the instances: add at least Or. 111.

¹³ Its absence from the very abbreviated report of his speech to the assembly in 932–42 is not significant; the arguments he uses to Tyndareos in 551–6 and 572–84 are also omitted there.

¹⁴ Not as reported in Prinz-Wecklein.

¹⁵ Weil, H., Sept tragédies d' Euripide (3rd edn, 1905), 745, ‘une espèce de superlatif;Google ScholarWecklein, N., Eur. Orestes mit erklärenden Anmerkungen (1906), 63; L. Méridier in the Budé translation (1959); and apparently V. Di Benedetto, Euripides Orestes (1965), 160. Cf. Verrall on Aesch. Eum. 394 (his 397):Wieseler, perhaps ratherGoogle Scholar

¹⁶ Ar. Av.921. This example sounds paratragic, and anadiplosis is a familiar device in tragic, especially Euripidean lyric. Other material, however, suggests that it was at home in colloquial speech: Ar. Equ. 1155 Nub. 1288 ‘more and more’, Ran. 1001 (so also Eur. IT 1406); Soph. Phil. 1197 and OC 210 are urgent rather than elevated. Cf. also Homeric later Anacreontea 58.4 9.3 Mod. Gk. etc.; F. T. Gignac, A Grammar of the Greek Papyri of the Roman and Byzantine Periods (1981), ii.159, on the phrase and more generally K. Brugmann, Kurze vergleichende Grammatik der indogerm. Sprachen (1902), 287,638f.; L. Radermacher, Indog. Forsch. 31,1912/13, Anzeiger 8f.; E. Hofmann, Ausdrucksverstärkung (Zeitschr. f. vgl. Sprachforschung, Ergänzungsheft 9, 1930), 12–48, ‘Die Doppelung’; E. Schwyzer and A. Debrunner, Griech. Grammatik, ii (1950). 700.

¹⁷ See J. Wackernagel, Kl. Schr. (1953), ii.1098 n. 1, who argues that the is an old instrumental. This presupposes, of course, that is a prehistoric survival, not an Aristophanic invention. Some may query this.

¹⁸ PMG 976(c); ci. Wilamowitz on Ar. Lys. 350.

¹⁹ See especially A. C. Pearson, The Fragments of Sophocles (1917), i.92f. and iii.5f

²⁰ Cf. also IT 193, 816; PI. Polit. 269a. J. S. Morrison, PCPS 16 (1970), 85–90, approved by Willink, interprets the Euripidean passages differently, of the tilting of the celestial pole from the zenith to its present position. He supposes dawn and evening to be mentioned as novel features, the sun having previously travelled round the edge of the sky. But can Euripides have thought there were no days and nights before Thyestes?

²¹ Oenopides DK 41 A 10 had a theory that the Milky Way marks the sun's earlier course.

²² The gloss in Hesychius/Photius/Swda/Bachmann's Anecdota perhaps suggests that some ancient commentators, expecting a preposition after the first accusative, understood as if it were They naturally had to read etmtpov (so also BOLC), which could be construed as an adjective.

²³ Di Benedetto and Biehl wrongly infer it as the lemma presupposed by the interlinear gloss in M (i. 199.25 Schw.), The Homeric phrase (II. 8.455) is an example of a masculine participial form used for the feminine. What was intended, therefore, was

²⁴ One sets horses to chariots rather than vice versa. Cf. II. 16.148 al. 24.14 Rhes.27

²⁵ Cf. perhaps Ov. M. 4.629f. dum Lúcifer ignés éuocel Aurórae, currús Auróra diurnós.

²⁶ Wilamowitz, EM. in d. gr. Trag., 152–4. It may be noted in addition that the discreet tone of is matched in the verbatim quotation from Callistratus in schol. 434,

²⁷ Also cited at schol. 12; cf. Wilamowitz, 156 n.72. I have not traced him elsewhere; he is unknown to Pauly-Wissowa.

²⁸ Collected Papers (1969, 127.

²⁹ Willink says that in Ion 1320–1 the word denotes the ‘ lofty, architectural masonry’ of the skene facade, as also in Hel. 430. ‘The Phrygian has no more come “over the triglyphs” than the Priestess in Ion has come “over the cornice”.’ He ignores not only the textual uncertainties of the Ion passage is transmitted), but the usual interpretation of it: ‘a low kerb protecting the adytum from intrusion’ (Owen, following Hermann, Paley, Wecklein; and so LSJ). It is probably something that the priestess steps through, not over, but not something high up on the building.

³⁰ IT14, 113, 128f., 1159, Ion 156, 172, 184ff., 1321, Hel. 70, 430, Or. 1569f., 1620, Hyps. fr. 764 (p. 24 Bond), Ba. 591, 1214; Pickard-Cambridge, A. W., The Theatre of Dionysus at Athens (1946), 125;Google ScholarHourmouziades, N. C., Production and Imagination in Euripides (1965), 29.Google Scholar

³¹ Cf. Wasps 379ff.; Dover, K. J., Aristophanic Comedy (1972), 25.Google Scholar

³² The only definitely indicated leap in tragedy is Euadne's suicide in Euripides' Supplices. It is from a high eminence, but her landing is concealed from the spectators, so it is no evidence of the height from which an actor might be required to jump. It is evidence, nevertheless, of Euripides' imaginative use of upper levels of the skene.

³³ In Wasps I.e. Philocleon lets himself down from the window by means of a rope. So the idea was available for Euripides' use.

³⁴ That the Phrygian got out between the triglyphs was the view of Paley, Weil, Wedd (all of whom, however, believed that this was in the inner courtyard, not visible to the audience), and Wecklein; so too England and Platnauer on IT 113; Pickard-Cambridge, op. cit. (n. 30), 53 n. 1. The hypothesis has been neglected by recent commentators. A further possibility, which would make the actor's task less demanding, is that there was a Doric frieze with open metopes between the lower and upper storeys. But Dr J. J. Coulton, whom I consulted on the question, is of the opinion that this would not be expected, as in monumental architecture true two-storeyed facades do not appear before the fourth century. He remarks that there are no certain examples of open metopes in stone architecture, but that some representations on vases suggest that they may have existed in non-monumental buildings (cf. B. Dunkley, BSA 36 [1935/6], 166f., fig. 8 and nn.). ‘I should think’ he writes ‘it would be hard for an actor to emerge from a metope opening unless the frieze was on quite a large scale (the fairly monumental Stoa of Zeus on the Athenian Agora has metopes two feet square), but I suppose it could be done.’

³⁵ I cannot see any force in M. D. Reeve's argument (GRBS 13, 1972, 263f.) that with the removal of 1366–8 and 1503–36 (Grüninger, Gredley) the scene becomes ‘perfectly symmetrical’, sc. strophe - monody - antistrophe. We find no such ‘symmetry’ in the other tragic examples of separated strophe and antistrophe, unless we count those cases where only dialogue intervenes (Soph. Phil. 391–02 - 507–18, Rhes. 131–6 ˜ 195–200; the others are Hipp. 362–72 ˜ 669–79 and Rhes. 454–66 ˜ 820–32). The monody itself is no marvel of symmetry.

³⁶ See Hourmouziades, op. cit., 142.

³⁷ Willink (303) thinks that a pendanfclose is less likely before a dochmiac period, and he assumes to be somehow corrupt for But see IT 889, Hyps. p. 47.86 Bond. It is risky to alter the strophe when it is the antistrophe that is obviously corrupt.

³⁸ So already Hartung and Weil, whose conjectures, however, are unacceptable: Hartung; Weil.