References to the editions of Davies (Epicorum Graecorum Fragmenta, Göttingen 1988) and Bernabé (Poetae Epici Graeci. Testimonia et Fragmenta I, Stuttgart – Leipzig 21996) can be found in his concordance (pp. 299–308).
All tragic fragments are quoted from these, unless otherwise indicated.
That Iliad 10 does not belong to to the original poem is ‘the almost unanimous (and certainly correct) view of modern scholars’ (West, Making of the Iliad, 233[-5]; cf. Hainsworth, The Iliad: A Commentary III, 151–5). The notion goes back to antiquity: ΣT Il. 10.0 (III 1.4–5 Erbse) ϕασὶ τὴν 
ὑϕ᾽ Ὁμήρου ἰδίᾳ τετάχθαι καὶ μὴ εἶναι μέρος τῆς 
, ὑπὸ δὲ Πεισιστράτου τετάχθαι εἰς τὴν ποίησιν (~ Eust. 785.42–4).
E.g. H. D. F. Kitto, YCS 25 (1977), 317–50 and Burnett, ‘Smiles’ (who regards it as a youthful ‘joke’ of Euripides). The most recent attack on Rhesus is Liapis’ commentary (see the reviews by D. Zuckerberg, CR n.s. 63 [2013], 29–31 and A. Fries, Mnemosyne IV 66 [2013], 814–21 + ‘Corrigendum’, Mnemosyne IV 67 [2014], 179).
S. Lakainai and Ion Phrouroi, which dramatised the Palladion theft and Ptôcheia respectively (501–2, 503–7a nn.), may be parallels (Ritchie 136–7), and also A. Phrygians = 
(+ Dionysius’ tragedy of that title: TrGF 76 F 2a?), if the poet(s) followed Iliad 24 strictly. S. Syndeipnoi = Ἀχαιῶν Σύλλογος perhaps started in the evening (fr. 143); cf. Sommerstein, in Shards from Kolonos, 359–60 (with n. 16) and Sophocles. Selected Fragmentary Plays I, 89, 92.
Strohm 260–1.
Cf. Ritchie 137 with n. 2 (where add 45, 765, 774, 788, 824 and 852) and Liapis, xxxiv–xxxv.
Strohm 257–66, 274, G. Paduano, Maia n.s. 25 (1973), 13–15 and, on the ironical interplay of light and darkness in the drama, H. Parry, Phoenix 18 (1964), 283–93.
B. Fenik, The Influence of Euripides on Vergil’s Aeneid, diss. Princeton 1960, 84–93 (especially 91–3). He largely depends on Strohm and does not take sides in the authenticity debate (cf. 54–5).
For further inversions between scenes see 85–148, 467–526 and 642–74nn.
Strohm 261.
On both motifs see Fenik (n. 8), 93 n. 1
Cf. also Strohm 264–5 and, in comparison with Iliadic patterns, R. S. Bond, AJPh 117 (1996), 262–3.
The chorus express some scepticism in 332 and particularly 882–4 and 995–6 (nn.).
A careful examination of the external evidence suggests a dating in the first quarter or third of the fourth century BC (ch. III.1).
Note Hector’s censure of the Shepherd (264–77), the speed with which he is persuaded to accept Rhesus as an ally (264–341, 333–41nn.) and the topic of especially the first agon (388–526n.).
See ch. III.2.
A. Lesky, Die tragische Dichtung der Hellenen, Göttingen 31972, 529. Cf. T. Thum, Philologus 149 (2005), 228.
Tr. S. D. Olson (ed.), Athenaeus. The Learned Banqueters IV, Cambridge (Mass.) –London 2008, 425.
Among fragmentary plays, enough of Sophocles’ satyric Ichneutae survives to define its relationship to the Homeric Hymn to Hermes (cf. Richardson, Three Homeric Hymns, 25–6).
Cf. G. Paduano, Maia n.s. 25 (1973), 15. On the sources of the plot see Ritchie 64–81, 83–6, R. S. Bond, AJPh 117 (1996), 255–73 and Liapis, xvii–xxvii, xxx–xxxi. To view our poet’s habit of modelling scenes and characters on others in the Iliad as an attempt to mislead the audience or reader (Fantuzzi, in I luoghi, 244–56, Entretiens Hardt LII, 148–51) is to credit him with the qualities of a Hellenistic or Roman poet.
Including the betrayal of the watchword, which is not in ‘Homer’ (cf. 572–3n.).
To repeat schematically (and a little more fully): Rh. 1–148 ~ Il. 10.1–179, Rh. 149–223 ~ Il. 10.299–337, Rh. 264–387 ~ Il. 10.436–41, Rh. 388–526 (end) ~ Il. 10.434, Rh. 527–64 ~ Il. 10.251–3 (time of night), 428–31 (allies), 561–3 (Dolon’s killing), Rh. 565–94 ~ Il. 10.339–468, Rh. 595–641 ~ Il. 433–41, 463–4, 474–5, 479–81, Rh. 642–74 (end) ~ Il. 10.509–11, Rh. 675–727 ~ Il. 10.523–4, Rh. 728–881 ~ Il. 10.515–21 (Charioteer ~ Hippocoon), Rh. 756–803 ~ Il. 10.471–97 (narrated in retrospect).
Cf. R. S. Bond, AJPh 117 (1996), 271–2 (although he probably goes too far in regarding Rhesus as an attempt ‘to create a dramatic equivalent of the Iliad’) and ch. I above.
Il. 10.245, 274–95, 366–8, 460–4, 482, 496–7, 507–12 + 516, 552–3, 570–1, 578–9. Some of these passages are reflected in Rhesus (ch. II.1(a)).
Thus also the second spurious prologue (Hyp. (b) Rh. 65.37–47 = 431.34–44 Diggle = fr. tr. adesp. 8l = TrGF V.2 (60) ΡΗΣΟΣ 642.12–22) is a dialogue between Hera and Athena. In the part cited, which betrays itself as just the ‘un-Euripidean’ pastiche the ancient critics took it for (pp. 26, 38 with n. 61), Hera worries about the hard-pressed Achaeans and would probably have turned to mentioning Rhesus and sending Athena on her ‘errand’ (Ritchie 110; cf. Fenik, Iliad X, 36–7 on the similarities with Il. 5.714–18 and 8.352–6).
Contrast e.g. Ritchie 63, Fenik, Iliad X, 30 and Jouan 54 n. 275. On Rhesus’ parentage see below.
Serv. on Verg. Aen. 1.469 ~ Myth. Vat. 1.200 Kulcsár (CCSL 91 C) = 203 Mai. Cf. Serv. on Aen. 2.13, Serv. auct. on Aen. 12.347 equos fatales, and the scholia to Ov. Ib. 629. On the many conditional oracles relating to the fate of Troy (most notably the fetching of Philoctetes and Neoptolemus) see W. Kullmann, Die Quellen der Ilias, Wiesbaden 1960, 221–3 and, in his wake, Fenik, Iliad X, 10–12.
W. Leaf, JHS 35 (1915), 2 n. 5, J. Rempe, De Rheso Thracum heroe, diss. Münster 1927, 38–40 (+ 53). Euterpe features most frequently (Heraclid. Pont. fr. 159 Wehrli + Apollod. FGrHist 244 F 146 = ΣV Rh. 346 [II 335.13–25 Schwartz = 94.7–95.20 Merro], Serv. auct. on Verg. Aen. 1.469, Eust. 817.27; cf. ‘Apollod.’ 1.3.4 [1.18]: Euterpe or Calliope, ΣV Rh. 393 [II 336.21–2 Schwartz = 97 Merro]: Clio or Euterpe, ΣAD Il. 10.435 [p. 356 van Thiel = I 364.17–18 Dindorf]: Terpsichore or Euterpe). Alone also Clio (ΣV Rh. 346 [II 335.8–13 Schwartz = 94.2–7 Merro] ~ Marsyas II FGrHist 136 F 7), Calliope (Hyp. (a) Rh. 64.20 = 430.19 Diggle) and Terpsichore (Ar. Byz. Hyp. (c) Rh. 65.48, 53 = 431.45, 432.50 Diggle, ΣTz Lyc. 831 [II 266.10–11 Scheer], Tz. Carm. Il. p. 65 Schirach). Strymon (cf. 279n.) became canonical, except for Dict. Cret. 2.45.1 (‘Eioneus’ as in Iliad 10) and Conon’s attempt to reconcile the two genealogies by having a mortal king Strymon give his name to the river Eioneus (FGrHist 26 F 1 IV; cf. [Plut.] Fluv. 11.1).
Achilles: Iliad, Aethiopis, A. Nereids, Cycnus: Cypria, S. Poimenes, Penthesileia: Aethiopis, Memnon: Aethiopis, A. Memnon, Psychostasia, S. Aethiopes (perhaps identical with his Memnon; only two fragments survive), Eurypylus: Little Iliad, S. Eurypylus, Sarpedon: Iliad, A. (?) Cares = Europa.
Penthesileia, Memnon and presumably Eurypylus also came in the evening to join battle on the following day (West, Epic Cycle, 138–9, 143, 190).
At length Ritchie 79–81, who suggests that some of these details were already attached to the Rhesus myth in the fifth century. Pindar certainly could have noted the parallels with Penthesileia, Memnon and Eurypylus and transferred the aristeia-motif himself.
Fantuzzi, in Entretiens Hardt LII, 143–5, Eikasmos 18 (2007), 186–7.
See M. L. West, CQ n.s. 50 (2000), 347–9 = Hellenica II, 241–3. The analogy would be reinforced if Cares = Europa had also contained an aition of Sarpedon’s cult at home.
Contrast Thucydides’ secularised account of this expedition (4.102.3) and also Cimon’s invasion of Scyrus (1.98.2) below.
Cf. Paus. 3.3.5–7, 3.11.10, 8.54.4 and see D. Boedeker, in C. Dougherty – L. Kurke (eds.), Cultural Poetics in Archaic Greece. Cult, Performance, Politics, Cambridge 1993, 164–77, I. Malkin, Myth and Territory in the Spartan Mediterranean, Cambridge 1994, 26–33 (especially 26–8 and 136–7).
Plut. Thes. 36 and Cim. 8.3–7, who unlike Pausanias (3.3.7) does not specify the causal nexus between the oracle (476/5 BC) and the capture of the island (cf. A. J. Podlecki, JHS 91 [1971], 141–3).
D. Lazaridis, Ergon 6 (1959), 42–3, Ἀμϕίπολις καί Ἄργιλος (Ἀρχαίες Ἑλληνικές Πόλεις 13), Athens 1972, 60 (no. 257), Amphipolis, Athens 1997, 44–5, 88 (plt. 49), G. 
13), Athens 1972, 60 (no. 257), Amphipolis, Athens 1997, 44-5, 88 (pit. 49), G. Daux, BCH 84 (1960), 797–8 (with plts. 8, 9).
I. Malkin, Religion and Colonization in Ancient Greece, Leiden 1987, 82. Cf. e.g. P. Perdrizet, Cultes et mythes du Pangée, Paris 1910, 15 and W. Leaf, JHS 35 (1915), 6–7, who by combination with the Muse’s prophecy saw in Rhesus a pièce d’occasion for the re-foundation of the colony. Similarly P. Perdrizet, in In Memoria lui Vasile Pârvan, Bucharest 1934, 284–90 and still tentatively Z. H. Archibald, The Odrysian Kingdom of Thrace. Orpheus Unmasked, Oxford 1998, 101.
The name Ῥῆσος is mostly recognised as Thracian for ‘king’, from the same IE root *(H)rēĝ- as e.g. Vedic rāj- and Latin rēx. For an extensive bibliography, including some critical voices, see V. Liapis, Kernos 24 (2011), 99 n. 26.
Rather than Aenia on the Chalkidike (cf. St. Byz. α 132 Billerbeck Αἴνεια· τόπος Θρᾴκης … τὸ ἐθνικὸν Αἰνειεύς). Thus Degani, most recently, reads Αἰνίων with ten Brink (Philologus 6 [1851], 39), who cites St. Byz. 52.8 Meineke = α 135 Billerbeck Αἶνος· … τὸ ἐθνικὸν Αἴνιος, ὡς Τήνιος (add e.g. Thuc. 7.57.5, IG I3 261 IV.2, FD III.1 497.6, Call. Iamb. 7 fr. 197.1 Pf. 
ὁ Π]ερϕεραῖος, Αἰνίων θεός). Some also made Rhesus a son of Hebrus (Serv. auct. on Verg. Aen. 1.469).
Cf. B. ten Brink, Philologus 6 (1851), 40, E. Degani (ed.), Hipponax. Testimonia et Fragmenta, Stuttgart – Leipzig 21991, 87 and id. (ed.), Ipponatte. Frammenti. Introduzione, traduzione e note (ed. G. Burzacchini, with additions by A. Nicolosi), Bologna 2007, 116.
Cf. Exc. Salm., An. Par. II 390.31–2 = [Ioan. Antioch.] fr. 24.6 FHG IV 551 Ῥῆσος 
οἰκήσεις ἔχων ἐν Βυζαντίῳ ἐν τῷ λεγομένῳ Ῥηάτῳ, ὅπου νῦν ἐστιν ὁ τοῦ μεγαλομάρτυρος ἁγίου Θεοδώρου ναός, Procop. Aed. I 4 (26.20–2 Haury-Wirth). One of Byzantium’s gates was called πόρτη τοῦ Ῥησίου (Theoph. Conf. Chronogr. 229.30–230.3, 231.20 de Boor). In general see V. J. Liapis, CQ n.s. 57 (2007), 396–7.
A neuter adjective of the IE ‘king’ root (n. 21). The type has several parallels from Ireland to the east of Europe, including the well-known R(h)egium in Calabria (West, IEPM 413).
On the often fictional character of Philostratus’ works see G. Anderson, Philostratus. Biography and Belles Lettres in the Third Century AD, London et al. 1986, 253–4, E. L. Bowie, in J. R. Morgan – R. Stoneman (eds.), Greek Fiction. The Greek Novel in Context, London – New York 1994, 181–99 (especially 184). P. Grossardt (Einführung, Übersetzung und Kommentar zum Heroikos von Flavius Philostrat, 2 vols., Basel 2006) seems more sceptical in his commentary (II, pp. 435–441) than in the introduction (I, pp. 35 with n. 8, 115 with n. 362) about Rhesus’ mountainous existence.
Earlier discoveries are listed by G. I. Kazarow, Die Denkmäler des thrakischen Reiter-gottes in Bulgarien, I Textband, II Tafelband, Budapest – Leipzig 1938. For additions to these and the Hero’s place in Thracian religion see e.g. A. Foll et al., Légendes Thra-ces, Sofia 1977, 7–40, V. Velkov – V. Gerassimova-Tomova, ANRW II 18.2, 1322–30 and Hoddinott, Thracians, 169–75. LIMC VI.1/2 s.v. Heros Equitans (A. Cermanović-Kuzmanović et al.) treats the Thracian Horseman in the wider context of mounted heroes.
First proposed by G. Seure, RPh 54 (1928), 106–39. The issue has recently been revisited (with cautious approval) by V. Liapis, Kernos 24 (2011), 95–104.
Cf. e.g. LIMC VI.1/2 s.v. Heros Equitans 170 (Varna, late II AD), 501 (Nesebăr, late II/ early III AD).
See V. Liapis, Kernos 24 (2011), 102–4.
Cultes et mythes du Pangée, 19, 27–8, 39. Cf. A. D. Nock, CR 40 (1926), 186 and Rh. 972–3n.
Possibly after Asclepius of Myrlea’s Bithyniaca (FGrHist 697 F 2). But see Lightfoot, Parthenius, 249–50, 253 on our inability to prove that those manuscript notices or ‘man-chettes’ name the actual source in any one case.
The location of this river Rhesus, which is mentioned already in Il. 12.20 (with Hainsworth on 20–2, 20) and Hes. Th. 340 (with West), was disputed in classical times. Demetrius of Scepsis (fr. 31 Gaede = Strab. 13.1.44) assumed it was the Rhoeites (cf. ΣT Il. 12.20 [III 299.35–7 Erbse] ~ Eust. 889.59–61) – εἰ μὴ 
ὁ εἰς τὸν Γρανικὸν ἐμβάλλων * * Ῥησός ἐστιν. The Elder Pliny (NH 5.124) could not find it, but gives ‘Rhesus’ as an alternative name for the Bithynian Rhebas (NH 6.4). Cf. Solin. 43.1 and, further on the Trojan river, D. Detschew, Die thrakischen Sprachreste, Vienna 1957, 397.
Lightfoot, Parthenius, 553. St. Byz. β 97 Billerbeck also declares Bithys, the ancestor of the Thracian Bithyans, to be a son of Ares and Rhesus’ sister Sete. The relationship of the anonymous SH 939 (II AD) with our tale is unclear.
Cf. V. Liapis, Kernos 24 (2011), 96. Ustinova (Kernos 15 [2002], 283 n. 154) assumes that Cicero ignored the fringes of the known world.
The main publications are in Θυμέλη, 159–88 and JHS 129 (2009), 71–88. See also CQ n.s. 57 (2007), 408–11 (on the non-Greek character of the cult envisaged by the Muse) and his commentary passim. He suggests as author the tragic actor Neoptolemus, who spent some time at the court of Philip II.
For Liapis (p. 77) the ‘Macedonian’ interpretation would do so.
M. Fantuzzi, JHS 127 (2007), 161. Cf. R. D. Milns, Historia 20 (1971), 187.
M. Manili Astronomicon …, Leiden 1600, 6–7: ‘Nam ecce auctor Rhesi Tragoediae ve-tustissimus, qui sine dubio non est Euripides …’ (cf. 8). Seven years before M. A. Del Río (Delrius) had already attributed the play to ‘Euripides II’ (Syntagma tragoediae latinae in tres partes distinctum, Antwerp 1593, 22). For a brief review of the debate since then see Jouan, X–XV.
Ritchie 11–15. To his cases of an author’s χαρακτήρ being invoked in questions of authenticity (13) add ΣA Il. 18.39–49 (IV 443.19–21 Erbse) ὁ τῶν Νηρεΐδων χορὸς προηθέτηται καὶ 
Ζηνοδότῳ ὡς Ἡσιόδ<ε>ιον ἔχων χαρακτῆρα.
Between these years Euripides competed in the 83rd Olympiad (448/7–445/4 BC) and won his first victory in 442/1 BC (DID D 1 60, D 3 TrGF), but we do not know with which plays.
Taken from Aristotle’s compilation or a full version of Aristophanes of Byzantium’s hypothesis (65.48–58 = 431.45–432.53 Diggle). Cf. Pfeiffer, History of Classical Scholarship I, 242 and Kannicht on TrGF V.2 (60) ΡΗΣΟΣ.3–4.
Cf. ΣV Rh. 251 (II 332.7–8 Schwartz = 89.5 Merro) κέχρηται δὲ καὶ νῦν Εὐριπίδης τῇ παροιμίᾳ παρὰ τοὺς χρόνους and ΣV Rh. 430 (II 337.14–19 Schwartz = 100 Merro), where Or. 220 is quoted as if by the same poet.
Ritchie 47–55.
De Rhesi scholiis, 10–12 = KS I, 9–12 and Einleitung, 155–6. The theory rests mainly on his unwarranted emendation of the probably misplaced (821–3n. with n. 292) ΣV Rh. 41 (II 330.6–7 Schwartz = 82 Merro) τὸ χ ̄ ὅτι συνθέτως ἀναγινώσκεται (sc. πυραίθει) καὶ ὅτι οὐκ ἔστιν Εὐριπίδειος [ὁ στίχος] (del. Wilamowitz, i.e. ‘because the play is not by Euripides’) and has rightly been refuted by Porter (Hermathena 17 [1913], 366–8) and Ritchie (49–52).
So E. Theristai, a satyr-play (DID C 12 TrGF = Ar. Byz. Hyp. Med. 90.40–3 Diggle), Cratin. Χειμαζόμενοι (Hyp. I Ar. Ach. 2.4 Wilson [PCG IV, 244]), all but one or two of Lysippus’ comedies (Lysipp. T 3 PCG = IG 14.1097.7–9 = IG Urb. Rom. 1.216.7–9) and an unknown, perhaps satyric, play connected in some way with E. Oenomaus and Chrysippus (Hyp. (g) Phoen. 81.6–7 Diggle).
Cf. ΣΣvett. Ar. Ran. 1269a/b/c (III 1a 143 Chantry), Σrec. 1269a (III 1b 211 Chantry).
Cratin. Satyroi: Hyp. A5 3.10–12 Jones–Wilson (PCG IV, 232). Eup. Noumeniai: Hyp. I Ar. Ach. 2.4–5 Wilson (PCG V, 424). In general see Fraenkel, Rev. 229–30.
Including four of Aristophanes (Poiesis, Dionysos Nauagos, Nesoi, Niobos), which ‘some said were by [his contemporary] Archippus’ (Vit. Aristoph. 1.59–61 PCG). Cf. Ritchie 24.
In the latter case the other fragments must follow, since ‘it is … most unlikely that both Sisyphoi survived the fifth century BC and were then quietly transmitted side by side’ (R. Kannicht, in C. Mueller-Goldingen – K. Sier (eds.), ΛΗΝΑΙΚΑ: Festschrift für Carl Werner Müller, Stuttgart – Leipzig 1996, 27).
Whether a tetralogy Tennes, Rhadamanthys, Pirithous, Sisyphus should be given to Critias (Wilamowitz, Analecta Euripidea, 161–6; cf. Einleitung, 40–2, ‘Memorandum’, edd. J. M. Bremer – W. M. Calder III, Mnemosyne IV 47 [1994], 211–16) cannot be discussed here nor indeed decided from the extant data. ‘Suffice to say that acceptance of Critias as author of the 42 lines considered above does not necessarily entail acceptance of either his further authorship of the disputed Peirithous or of the notion that the Sisyphus was written to accompany the above-mentioned tragedies’ (M. Davies, BICS 36 [1989], 32). Cf. C. Collard, in J. A. López Férez (ed.), Da Homero a Libanio …, Madrid 1995, 183–93 ~ Selected Papers, 56–68 and Collard–Cropp, Euripides VIII. Fragments, 629–35.
If, as West plausibly suggested (Studies, 67–72, CQ n.s. 50 [2000], 339 = Hellenica II, 229–30), Aeschylus’ son Euphorion wrote and staged the play, or in fact the entire Prometheus trilogy, under his father’s name and it was so entered in the didascaliae.
Apart from Aristotle and his pupil Dicaearchus. Previous efforts had primarily been directed at tragic mythography, e.g. Asclepiades of Tragilus’ six books of Τραγῳδούμενα (below) and Philochorus’ Περὶ 
Σοϕοκλέους 
(cf. J. Rusten, GRBS 23 [1982], 361–2 with n. 22).
Cf. J. S. Scullion, in D. Cairns – V. J. Liapis (eds.), Dionysalexandros. Essays … in Honour of Alexander F. Garvie, Swansea 2006, 188. According to Kannicht’s (n. 12) and Scullion’s (197–8 n. 7) calculations, only two (or three) of the fifteen (or sixteen) dramas by Euripides that did not reach Alexandria were tragedies, the others satyr-plays.
[E.] fr. 1108 N.2 = E. fr. 660a N.2-Sn. = TrGF V.2 (60) ΡΗΣΟΣ, 642.8. A supplement for the next line, more suitable to the action of Rhesus than 
διώκει … (Snell, 
Diggle) may be something to do with Nyx (J. Rusten, GRBS 23 [1982], 360 n. 17). For ‘Night’ driving a chariot cf. Cho. 660–1, A. fr. 69.6–7 and Ion 1150–1 (A. Fries, Mnemosyne IV 66 [2013], 816 with n. 4).
A. Nauck, Aristophanis Byzantii Grammatici Alexandrini Fragmenta, Halle 1848, 254. Recent attempts to defend the paradosis (A. Tuilier, Sileno 9 [1983], 11–28 [21–3], P. Carrara, ZPE 90 [1992], 35–44 [40–3]) do not convince. See V. Liapis, GRBS 42 (2001), 313–16, who then, however, follows Kirchhoff (Philologus 7 [1852], 563–4) and others in regarding the whole end of Hyp. (b) Rh. as Dicaearchan, on the basis of which he would reconstruct a lost work including both plot narratives and more theoretical dicus-sions.
Zuntz, Political Plays, 135, although Wilamowitz (Analecta Euripidea, 183–4, Einleitung, 134 n. 19, 170) had already deduced the existence of such a collection and compared it to Lamb’s Tales from Shakespeare. As Sophoclean examples have also come to light (P. Oxy. 3653, 3013), ‘narrative hypotheses’ may now be a better title (van Rossum-Steenbeek, Greek Readers’ Digests?, 1–2). Dicaearchan authorship was particularly defended by M. W. Haslam, GRBS 16 (1975) 149–74 (150–6) and W. Luppe, in J. Wiesner (ed.), Aristoteles. Werk und Wirkung I, Berlin – New York 1985, 610–12 and in W. W. Fortenbaugh – E. Schütrumpf (eds.), Dicaearchus of Messana. Text, Translation, and Discussion, New Brunswick – London 2001, 329–32.
J. Rusten, GRBS 23 (1982), 357–67, R. Kassel, in W. J. Aerts et al. (eds.), Σχόλια. Studia … D. Holwerda oblata, Groningen 1985, 53–9 = KS, 207–14, Scullion (n. 16), 198 n. 9.
J. Diggle, in G. Bastianini – A. Casanova (eds.), Euripide e i papiri …, Florence 2005, 27–67. A terminus ante quem is provided by the first-century AD P. Mil. Vogl. II 44 (Pack2 = Mertens–Pack3 398).
Pfeiffer, History of Classical Scholarship I, 113–15, 173–4. Cf. V. Liapis, GRBS 42 (2001), 318–19, also against the theory of Carrara (n. 18) that Aristophanes’ note on the προλογίζων is a later addition. For the formulation cf. e.g. Ar. Byz. Hyp. (g) Phoen. 81.7–8 Diggle ὁ χορὸς συνέστηκεν ἐκ 
γυναικῶν. προλογίζει δὲ 
and Hyp. Pers. 3.7–8 West 
δὲ προλογίζει χορὸς πρεσβυτῶν.
LIMC VIII.1/2 s.v. Rhesos 3, 4, 6. Cf. L. Giuliani, Tragik, Trauer und Trost. Bildervasen für eine apulische Totenfeier, Berlin 1995, 31–3, 94–102, BICS 41 (1996), 71–86 (with plts. 16–20), Jouan, LXI–LXIII, O. P. Taplin, Pots & Plays. Interactions between Tragedy and Greek Vase-Painting of the Fourth Century B.C., Los Angeles 2007, 160–5, 283 (notes).
V. Liapis, ZPE 143 (2003), 20.
Kaibel on Epich. fr. 205, M. L. West, in Hansen’s apparatus on Hsch. ρ 272 (of which mine is an abridged and adapted version). More hesitantly V. Liapis ZPE 143 (2003), 20–2.
V. Liapis, GRBS 42 (2001), 328, in Θυμέλη, 173–7 ~ JHS 129 (2009), 85–6 and his commentary on Hyp. (b) Rh. (pp. 59–60, 62–5).
Ritchie 35–7 and also A. Fries, Mnemosyne IV 66 (2013), 815–16. In addition to Asclepiades and the vase painting, the possible reminiscence of Rh. 161–83 in Men. Peric. 271–91 (cf. p. 44 with n. 87) would have to be explained away.
It has long been shown that neither the stichic nor the lyric metres afford conclusive evidence to the dating and authorship of the play (cf. Fraenkel, Rev. 236–7, Liapis, lxiv–lxvii).
G&R 6 (1959), 96–108.
Ritchie 143, 150–1, 170–1, 175, Fraenkel, Rev. 230. Cf. Rh. 214–15, 551–3 (ἰάν), 736–7nn.
M. Parry (HSPh 41 [1930], 96–7, 98 = The Making of Homeric Verse, 284–5, 286) notes how the irregular metres of choral song inhibit the formation of set expressions. He also warns against unthinkingly equating tragic iterations with traditional epic formulae (80–4, 97–114 = 272–5, 285–98).
See Ritchie 218–25, who compares to Rhesus collections of recurrent junctures in Hippolytus and Bacchae, without, however, specifiying their metrical positions or any artistic effect they might have.
Cf. Ritchie 218–19. The plentiful allusions to night and darkness, so intrinsic to the setting and atmosphere of our play, can be related to the frequency with which the similarly repetitive author of Prometheus Bound refers to the skene as ‘rock’, ‘hill’ or ‘crag’ (Griffith on PV 15).
Cf. e.g. Rh. 57 (56–8n.) … Ἀργείων στρατόν, 577 (n.) τί 
ἂν εἴη; 669 (668–9n.) 
παῖ, 915 (915–16n.) … 
μολών.
A term coined by P. T. Stevens, CR n.s. 15 (1965), 270 (cf. H. D. F. Kitto, YCS 25 [1977], 318–19, Liapis, lviii–lix). Few of the ‘fifth-century’ half-lines which recur in the minor tragedians (Th. K. Stephanopoulos, ZPE 75 [1988], 3–38) may be identified as (semi-) conscious contextual borrowings: e.g. Chaerem. TrGF 71 F 30 ~ Andr. 1008, Theodect. TrGF 72 F 2.1 ~ E. El. 1278 (two hated mothers) and Theodect. TrGF 72 F 10.1–2 ~ Phoen. 3. Among the adespota (Th. K. Stephanopoulos, ZPE 73 [1988], 207–47) attribution to one of the ‘big three’ is often a possibility.
See P. T. Stevens, CR n.s. 15 (1965), 270. As regards diction, Rh. 42 (41–2n.) ὄρϕναν, 164, 271 … 
λέγω, 351–2, 393, 923 μελῳδοῦ, μελῳδίας, 509 θάσσων, 625 κομψά, 662 (n.) ϕροῦδος, 722 
μ᾽ ἔχει, 750 
and 974 … 
θεοῦ are sanctioned by comic parody (especially Aristophanes). Cf. D. Sansone, BMCR 2013.03.15.
E.g. Rh. 97 
εὐσέλμων νεῶν, 255 οὐτάσει, 825 ἔβριξ᾽, 827 
μοι 
… 
and 932–3 ϕιλαιμάτους / 
κορύσσοντ᾽, where see the commentary notes. Cf. A. C. Pearson, CR 35 (1921), 55.
Cf. A. C. Pearson, CR 35 (1921), 56 with n. 1, Fraenkel, Rev. 238.
Cf. Fraenkel, Rev. 239 and 445b–6n. (against the inclusion of 
… ἐξ ἡμέρας).
In Rh. 82 (n.) … ἐν 
δορός + 819 τὸ 
εἶναι … ~ Ai. 1275 
τὸ 
ὄντας ἐν 
the second ‘part’ comes too late and is too common a phrase to be significant (cf. Ritchie 202–3).
Fraenkel, Rev. 233.
On Myrmidons see below (pp. 34–5 with n. 53). Another welcome reference to a lost drama which seems to have wider connections with our play is Rh. 967–9 (n.) ~ A. (?) fr. 99.13–14 (Cares = Europa). Cf. above (p. 14).
With the help also of a choral-lyric reminiscence at Rh. 554–6 (n.) ~ Pi. Pyth. 9.23–5, Bacch. Pae. 4.76–8 and probably Alcm. 3 fr. 1.7 PMGF.
De Rhesi scholiis, 12 = KS I, 13 and, more reserved, Hermes 61 (1926), 282–3 n. 3, 284, 288 = KS IV, 409–10 n. 3, 411, 415.
A stronger linguistic connection would be established if S. fr. 859 ~ Rh. 33, 383–4 (nn.) could be firmly ascribed to Poimenes. See also S. fr. 515 ~ Rh. 380 (380–1n.).
So Wilamowitz (De Rhesi scholiis = KS I, 13, Hermes 61 [1926], 282 = KS IV, 409). It can hardly have been the Greeks, as Welcker and others surmised, since they did not arrive over land (Sommerstein, in Sophocles. Selected Fragmentary Plays II, 181, 203 [on fr. 502]).
Cf. Sommerstein (n. 46), 180–1, 204–6 (on frr. 500, 501, 507) and 388–526n. with n. 147.
See also ch. III.3 (p. 41) and 380–7n. on a possible chariot entry in A. Memnon and Rhesus and, for the overall pattern behind the theatrical representations of Cycnus, Memnon and Rhesus, Ritchie 99–100, M. P. Pattoni, PapLup 9 (2001), 313–31 and ch. II.1 (pp. 13–14).
From only just outside the prologue note Rh. 290 (n.) ~ Sept. 79–80 and otherwise Rh. 49–51 (n.) ~ Sept. 651–2, Rh. 122 (n.) ~ Sept. 447–8 (cf. p. 35), Rh. 158 (158–9a n.) ~ Sept. 658 (+ Eum. 90), Rh. 306b–8 + 383–4 (nn.) ~ Sept. 385–6, Rh. 568b–9 (n.) ~ Sept. 245 + 249 (cf. p. 37), Rh. 770 (770–2n.) ~ Sept. 287 + 288–9 (cf. p. 37) and Rh. 796 (795b–6n.) ~ Sept. 593.
Rh. 30 (n.) 
ἔϕοροι ~ Pers. 25 
ἔϕοροι, Rh. 58 (56–8n.) 
~ Pers. 54, Rh. 117 (n.) 
+ 254, 763 (253–5a, 763b–4a nn.) 
(-εῖ) ~ Pers. 126–7 
… ἱππηλάτας / καὶ πεδοστιβὴς λεώς (?), Rh. 311 
(‘divisions’) ~ Pers. 47, Rh. 375 (375b–7n.) σὲ γὰρ 
~ Pers. 87 δόκιμος δ᾽ οὔτις ὑποστάς, Rh. 741 (n.) 
~ Pers. 44 
δίοποι. For further reminiscences of Persians see below (p. 35–6) and 53–5, 72–3, 96–8, 290, 436–7, 471–2, 734–5nn.
The prologue question is reviewed by W. Stockert (ed.), Euripides. Iphigenie in Aulis I. Einleitung und Text, Vienna 1992, 66–79, to which add D. Kovacs, JHS 123 (2003), 77–103 (80–3, 101–2).
Cf. A. Fries, CQ n.s. 60 (2010), 348 with nn. 16–18 (also on Seven against Thebes).
On the first passage see further below and for the other correspondences 312–13, 814–15nn.
At the outset of the contest between ‘Aeschylus’ and ‘Euripides’ (Ar. Ran. 911–20) Myrmidons and Niobe are ridiculed for their silent ‘opening-tableaux’. Cf. ‘Scene and Setting’, 114 and 1–51n. (p. 116).
The verse-end is varied again at 833 … 
τε βαρβάρου, where the nouns belong to different syntactical units.
Less obvious cases from other scenes are Rh. 690–1 (690, 691nn.) ~ Or. 1353–5 + Cho. 288–9 and Rh. 699–701 (699–701, 701nn.) ~ Tro. 187–9, 241–2 + Hcld. 84, although the latter may rather be a question of stock language used for a particular topic (cf. pp. 31–2). Similarly, thematically linked ‘half-verse formulae’ are joined at Rh. 871 
τράπωμαι 
μονούμενος ~ Ba. 1366 (
…) + Alc. 380 (… 
μονούμενος) and perhaps Rh. 894 (893b–4n.) δόλιος Ὀδυσσεὺς 
τείσει δίκην ~ Phil. 608 (δόλιος Ὀδυσσεύς …) + Med. 802 (… τείσει δίκην).
Cf. Fraenkel, Rev. 235.
Carc. II TrGF 70 T 1 = Suda κ 394 Adler Καρκίνος … 
… ἤκμαζε 
ὀλυμπιάδα (380/79–377/6 BC), πρὸ τῆς Φιλίππου βασιλείας τοῦ Μακεδόνος.
See Xanthakis-Karamanos, Studies in Fourth-Century Tragedy, 87–9, Th. K. Stephano-See Xanthakis-Karamanos, Studies in Fourth-Century Tragedy, 87-9, Th. K. Stephanopoulos, ZPE 75 (1988), 6–7 and Kannicht on Hel. 1306–7, 1321–2. Carcinus repeatedly visited Sicily (Timae. FGrHist 566 F 164 = D. S. 5.5.1), and our fragment is the first explicit witness for this relocation of the legend (Richardson, Homeric Hymn to Demeter, 76–7).
Cf. Rh. 962 
εἶσι γαίας ἐς 
πέδον. The adjective 
does not occur elsewhere.
Th. K. Stephanopoulos, ZPE 73 (1988), 208. Cf. above (p. 26).
Fraenkel, Rev. 233. Cf. Zu den Phoenissen, especially 76–81, 82–3, 101–13, 117–18 and Rh. 149–94 (pp. 175–6 with n. 74), 728–55 (+ ‘Metre’) nn.
Cf. Sept. [1018–19] θεῶν πατρῴων, οὓς 
ὅδε / στράτευμ᾽ 
ἐμβαλὼν 
~ Sept. 582–3 πόλιν 
καὶ θεοὺς 
ἐγγενεῖς / 
, στράτευμ᾽ 
ἐμβεβληκότα and Sept. [1036–7] 
/ λύκοι ~ Sept. 496 
(of Hippomedon’s shield).
Hutchinson (on Sept. 1005–78) persuasively argues for the priority of the Phoenissae scene, which he thinks genuine. But even an interpolation could have been used by a subsequent writer.
See Sept. [1007–8] 
… / 
ἔδοξε γῆς 
+ [1037–8] 
γὰρ 
καὶ 
ἐγώ / … μηχανήσομαι ~ Ant. 920 ζῶσ᾽ 
θανόντων ἔρχομαι 
+ 891–2 
… 
/ οἴκησις (Barrett, Collected Papers, 328–9), Sept. [1013–14] ~ Phoen. 1628–30 (1629b = Ant. 26b) + E. El. 896, Sept. [1028] + [1052–3] ~ Phoen. 1657 and Sept. [1045] ~ Phoen. 1656.
See 149–94n. (Dolon’s entry and Hector’s silence in 191–265), 264–341n. (the Shepherd staying on after delivering his message), 595–674n. (Paris played by a fourth actor) and 804–81n. (Hector’s return from an unspecified direction and the Charioteer remaining on stage). The term ‘unconventional procedures’ is taken from J. P. Poe, Philologus 148 (2004), 21–33.
G. Björck, Arctos n.s. 1 (1954), 16–17, Eranos 55 (1957), 14, Ritchie 94–5.
Wilamowitz, Hermes 61 (1926), 286 = KS IV, 414. Cf. Burlando, Reso, 35 n. 25, Liapis, xli with n. 117.
Hector’s accusation that the sentries let the enemies slip by (804–81, 808–19nn.), which is foreshadowed by their fear in 722–7 (692–727n.) and immediately followed by self-defence (820–32, 821–3nn.), perhaps hints at the ‘problem’.
Cf. M. Fantuzzi, Entretiens Hardt LII, 142 n. 22, JHS 127 (2007), 162.
Cf. O. P. Taplin, LCM 1 (1976), 50.
Xanthakis-Karamanos, Studies in Fourth-Century Tragedy, 11, 12–13, 123.
See most recently S. Perris, G&R 59 (2012), 151–64 (especially 151–7, 160–1, 163–4). To his list of earlier literature (153 n. 11) add Wilamowitz, Hermes 61 (1926), 286 = KS IV, 413, G. Björck, Arctos n.s. 1 (1954), 16, P. Arnott, Greek Scenic Conventions in the Fifth Century B.C., Oxford 1962, 100 and Klyve 104.
Perris (n. 73), 152 also considers Euripides’ and Sophocles’ Palamedes, which are set in the Greek camp. In Aeschylus’ Achilles trilogy the acting area represents the interior of the hero’s hut.
Sommerstein, Eumenides, 33 (although see Garvie, Choephori, xlvi–xlvii against his claim that in the first half of that play the building was also ignored). West’s case for a skene in Aigyptioi remains speculative (Studies, 169–70), and A. fr. 58 (Edonoi) ἐνθουσιᾷ 
δῶμα, 
need not mean that, contrary to Bassarai, Lycurgus’ palace formed the background.
Cf. Taplin, Stagecraft, 43, 76.
So e.g. Fraenkel, Rev. 240–1, Liapis, xxxix.
For useful reviews of how the scene could have been staged see Taplin, Stagecraft, 252–60, M. L. West, JHS 99 (1979), 136–9 ~ Hellenica II, 262–7 (who proposed several cranes) and Griffith on PV 128–92.
Phoenissae was known as 
or 
already in antiquity (Hyp. (b) and (c) Phoen. 77.7, 78.1–2 Diggle). Helen has nine speaking roles (like Andromache), if at 1627–41 we reject the second servant in favour of the coryphaeus (cf. 149–94n. [p. 174]), while the other extant ‘three-actor’ plays waver between five (Eumenides, Philoctetes) and eight.
Mastronarde on Phoen. 1308–479 (p. 511).
Taplin, Stagecraft, 39–49, 477–9. Among the most telling evidence is the comic ridicule of Xenocles, son of Carcinus, as mechanically adventurous (TrGF 21 TT 3a, c, 33 T 4b), Aristotle’s famous censure of poets who seek to arouse ‘tragic’ emotions only through 
(Poet. 1453b7–11) and the repeated references in the Life of Aeschylus to the monstrous effects he obtained (A. T 1 §§ 7, 9, 14 TrGF; cf. 675–91n. [p. 371 n. 250]). Note also the kind of interpolation mentioned in 380–7n.
Apart from Bacchae, which is only carried by the ‘alphabetic’ branch (LP). See p. 49 and p. 53 n. 39.
Barrett, Hippolytos, 50–3. For Wilamowitz’ view see Einleitung, 173–203 (especially 173–4, 195–203).
In particular Aeschylus’ Achilles trilogy (Myrmidons, Phrygians, Nereids) and A. (?) Cares = Europa, on which see pp. 13–14. Also A. Psychagogoi, Penelope, Ostologoi and S. Nausicaa = Plyntriai (all based on the Odyssey). The latter has sometimes been regarded as a satyr-play, like A. Proteus and Kirke.
Cf. Wilamowitz, Einleitung, 195.
West, Homeric Hymns, 235. Cf. H. Wölke, Untersuchungen zur Batrachomyomachie, Meisenheim am Glan 1978, 28–44.
M. Fantuzzi – D. Konstan, in E. Bakola et al. (eds.), Greek Comedy and the Discourse of Genres, Cambridge 2013, 256–74, especially 265–73. Cf. 149–94n. (p. 174 n. 71).
Ritchie 45–7 against, by implication, G. Pacitti, Maia n.s. 15 (1963), 184–98. See also M. Dangel, Accius. Oeuvres (fragments), Paris 1995, 295–6.
B. Fenik, The Influence of Euripides on Vergil’s Aeneid, diss. Princeton 1960, 54–96, (from where all parallels are taken), B. Pavlock, TAPA 115 (1985), 207–24 and, more reserved, A. König, Die Aeneis und die griechische Tragödie – Studien zur imitatio- Technik Vergils, diss. Berlin 1970, 89–108.
For further details see M. P. Pattoni, MD 53 (2005), 83–123 (100–5) and Liapis on 208–15.
Only Rh. 105–8 (Stob. 4.13.8), Rh. 206 (Stob. 2.31.14 = Orio Anth. 1.7 [p. 42 Schneidewin] = [Men.] Mon. 718 Jäkel) and later Rh. 106–7 (Apostol. 13.51g [CPG II 588.18–19]), Rh. 182b–3 (Apostol. 18.34h [CPG II 727.20–1]) and the Byzantine gnomologies gV, gB and gE, which belong to the direct manuscript tradition (p. 52–3).
Since of the extant tragedies Plutarch does not cite Agamemnon, Eumenides, Philoctetes and Helen either (K. Ziegler, RE XXI.1 s.v. Plutarchos 2, col. 917), it is unsafe to assume with Klyve (28) that he regarded Rhesus as spurious or that, on account of such misgivings as voiced by the ἔνιοι (p. 22), it was absent from his Euripides texts.
ΣA Il. 6.479–80 (II 212.19–23 Erbse), where Rh. 390–1a and Il. 13.352–3 are not quite rightly adduced to explain the loose accusative participle 
after εἴποι.
Ap(h)thonius GL VI 54.9–12 Keil (the metrical treatise that was combined with the Ars Grammatica of Marius Victorinus): Rh. 211, Bas. Caes. leg. libr. gent. 31.576 B–C Migne: Rh. 84 (garbled).
Orus B 77 Alpers = Et. Gen. (cod. B) s.v. 
(cf. p. 55). Codex A of the Etymologicum Genuinum omits the second half of Rh. 855 (ἀλλὰ 
τάδε), and in that form the passage is also found in the Suda (η 573 Adler) and the Etymologicum Magnum (439. 3–4).
Cf. Eust. ad Dion. Perieg. 270 (I 138.9–11 Bernhardy = GGM II 264.17–19) ~ Σ Dion. Perieg. 270 (GGM II 442.20–1): Rh. 29, ad Il. 10.519–25 (822.2–5): Rh. 338? (n.), 802–3, 833–55, Tz. Hist. 4.969–70 Leone: Rh. 510–11, Carm. Il. (commentary) p. 65 Schirach: Rh. 618.
On the early popularity of the three plays (with slightly lower scores for Hecuba) see M. Heath, BICS 34 (1987), 41–3 ~ J. Mossman (ed.), Oxford Readings in Classical Studies: Euripides, Oxford 2003, 220–3 and Collard, Hecuba, 37–8.
See Diggle, Euripidea, 517 n. 27 (for the letter α) and my testimonial apparatus, the principles of which are explained in ch. V.
For the special case of Bacchae see p. 43 with n. 82.
A. Turyn, The Byzantine Manuscript Tradition of the Tragedies of Euripides, Urbana (Ill.) 1957 (repr. Rome 1970), 333–5, K. Matthiessen, Studien zur Textüberlieferung der Hekabe des Euripides, Heidelberg 1974, 39, D. J. Mastronarde – J. M. Bremer, The Textual Tradition of Euripides’ Phoinissai, Berkeley 1982, 3.
N. G. Wilson, Scrittura e Civiltà 7 (1983), 161–76. In essence already CR n.s. 28 (1978), 336 and JHS 100 (1980), 219.
G. Vuillemin-Diem – M. Rashed, Recherches de Théologie et Philosophie médiévales 64 (1997), 136–98 (especially 157–80).
Before the invention of spectacles, Nigel Wilson reminds me, the career of an average scribe spanned a maximum of thirty years. See the dated manuscripts under individual names in M. Vogel – V. Gardthausen, Die griechischen Schreiber des Mittelalters und der Renaissance, Leipzig 1909.
E.g. 17 λόχος V: 
OQ, Tr3P2: δοῦλος <LP>, 161 μὲν V: με OΛgVgB, Chr. Pat. 1964, 296 
V: 
OΛ, 601 ἀχιλλεὺς Va: -έως OΛ, 702 
VΛ: 
O (cf. Klyve 86 n. 181). Note also Ioannikios’ propensity for committing the vitium Byzantinum (170, 218, 220, 265, 269, 290, 313, 331, 421, 426, 433, 503, 506, 606, 618, 635, 636).
Cf. Wecklein, Textkritische Studien, 68–70 (who wrote before the discovery of Π2 and the gnomologies).
Turyn 90; cf. N. G. Wilson, Gnomon 38 (1966), 342, CR n.s. 16 (1966), 288 on the somewhat earlier date. Further descriptions in Matthiessen 46–7, Mastronarde–Bremer 3–4 (+ D. J. Mastronarde, GRBS 26 [1985], 106–8), Diggle, Textual Tradition, 6 and H.-C. Günther, The Manuscripts and the Transmission of the Paleologan Scholia on the Euripidean Triad, Stuttgart 1995, 225.
See above (n. 7) and Barrett, Collected Papers, 427.
The last folio (Rh. 899–940) had survived in a set of loose fragments and was reunited with the manuscript upon its discovery by H. Rabe (RhM N.F. 63 [1908], 419–22).
Turyn 91–2, Matthiessen 45–6.
Matthiessen 126–8 (for Hecuba). Examples in Rhesus (without rhythmical significance) are 398 (396–8n.) πολεμίῳ, 607 (n.) ἥξει and probably 887 
(νεόχμ- V: νεόκμ- Λ, recte), which also occurs in Chr. Pat. 1456. The metrical interests of our scribe are evident also from his insertion of pertinent notes and a brief treatise copied on foll. 1r–3v.
H. Rabe, RhM N.F. 63 (1908), 421–2. On the text source see Zanetto, ed. Rhesus, IX–X and, by implication, Diggle, Euripidis Fabulae III, 428, against Turyn (91), who had envisaged a congener of L.
Cf. Rh. 951 ἕκτορ Va: -ωρ Λ, 952 
Va: -εως Λ, 960 ξυμ- Va: συμ- Λ, 976 
Va: -ους Λ, 987 πληροῦν Va: -οῦ Λ: -ου[ Af, 993 στείχωμεν Va: -ομεν Λ. The tendency of the V scribe to write single instead of double consonants is relevant for the constitution of the text at 461–3 (n.).
See R. Prinz, RhM N.F. 30 (1875), 129–33, Turyn 329–33, Matthiessen 42 and B. Schar-tau, Codices Graeci Haunienses. Ein deskriptiver Katalog des griechischen Handschrif-tenbestandes der Königlichen Bibliothek Kopenhagen, Copenhagen 1994, 99–100.
For both phenomena in Hn and other late Euripides manuscripts see Diggle, Euripidea, 239, 243–4, 270–2.
Cf. Diggle, Euripidea, 324 with n. 8.
Turyn 222–58, Zuntz, Inquiry, 126–35, 144–51, 192, Matthiessen 39–40, Mastronarde–Bremer 7 and on the date also O. L. Smith, C&M 43 (1992), 219 n. 92.
Two: Turyn 229, less certain Zuntz, Inquiry, 103–4, 127–8, 134, 178; cf. N. G. Wilson, Gnomon 38 (1966), 336. Three: J. A. Spranger, SIFC n.s. 10 (1932), 321–4, P. G. Mason, CQ n.s. 4 (1954), 57–8, C. Collard, SIFC n.s. 35 (1963), 107–11 (= Selected Papers, 109–114, with ‘Endnote 2006’), Matthiessen 39.
In some of these cases, notably 712, which is corrupt only in <L>P, the truth may go back to L’s exemplar (cf. below on the importance of Q in this respect).
Turyn 258–64, Zuntz, Inquiry, 135–40 (with N. G. Wilson, Gnomon 38 [1966], 336–7), Matthiessen 41, Mastronarde–Bremer 8. For the dating see A. Turyn, Codices graeci Vaticani saeculis XIII et XIV scripti annorumque notis instructi, Vatican City 1964, 127 and O. L. Smith, Mnemosyne IV 35 (1982), 328, C&M 43 (1992), 198, 219 (among new information on the scribes and the advanced metrical notes on E. Supplices and the two Iphigeneia plays). Rhesus is found in the Vatican part.
Zuntz, Inquiry, 144–51, Diggle, Euripidea, 508–13.
Zuntz, Inquiry, 140, 148. Cf. Rh. 11, 49, 266, 345, 348, 531, 702, 706, 711, 733a, 734, 827, 907.
Cf. Zuntz, Inquiry, 136, 139, 289 and Diggle, Euripidea, 510–12, also on his other activities in P.
Turyn 288–98 (with Barrett, Collected Papers, 425), Zuntz, Inquiry, 144–51. Turyn’s dating to about 1475 was revised by Wilson (Gnomon 38 [1966], 337), who identified the scribe as Gian Francesco Burana, born in 1474.
Cf. Diggle, Euripidis Fabulae II, vi–viii (on Troades).
Zuntz, Inquiry, 147–50 with 150–1 n. ¶. Cf. e.g. Rh. 262, 263, 682, 828, 830.
Diggle, Euripidea, 324–5, where add 43 (ναῶν VQ et L1c vel Tr1: 
Luv: 
O), 215 (εἶμι Q et Tr2?: 
L?P: εἰμι Δ), 354 (σὰν 
Q et Tr3: σὰν ἐϕύτευσ᾽ L: 
O: -σαν V), 412 
OQgV: -ον VLgE), 686 (ἦ Q: ἢ ΔL) and 994 (καὶ ξυμμαχίᾳ Q: καὶ συμμ- L: 
Va).
Turyn 341–2.
Turyn 296–8.
Zuntz, Inquiry, 144–5.
In general A. Meschini, Helikon 13–14 (1973–74), 349–62 (349–55).
K. Matthiessen, Hermes 94 (1966), 409–10.
G. A. Longman, BICS 4 (1957), 60–1, CQ n.s. 9 (1959), 129–41 (analysis and collation of the Euripidean part) and E. Lamberz, Katalog der griechischen Handschriften des Athosklosters Vatopedi I, Thessaloniki 2006, 156–62, who on the basis of the principal script (161) supports Christodoulou’s dating to the middle of the eleventh century (᾿Επιστημονικὴ 
Σχολῆς τοῦ 
Ἀθηνῶν 26 [1977/8], 322–3 = Σύμμικτα κριτικά, Athens 1986, 30–1). After examining the samples on the CD-ROM accompanying Lamberz’ catalogue, Nigel Wilson informs me that the mid- to late eleventh century is ‘perfectly possible’.
In 482 (n.) the correct 
νυν (coni. Scaliger) for … 
(ΩgBgE) is barely a variant.
On both compilations see Turyn 93–4 n. 151, Mastronarde–Bremer 173–4. Collations in K. Matthiessen, Hermes 93 (1965), 148–58 (gB) and Hermes 94 (1966), 398–410 (gE).
Including one on Ba. 344 in gB, the final proof that this play belonged to the ‘Selection’.
Edited most recently by A. Tuilier, La passion du Christ. Tragédie, Paris 1969, who still falsely attributes it to Gregory of Nazianzus. On the dating to the eleventh or twelfth century see W. Hörandner, in E. Trapp et al. (eds.), Studien zur Byzantinischen Lexiko-graphie, Vienna 1988, 183–202, also against the attempt of A. Garzya to prove that the extant manuscripts of the cento derive from an uncial source (Sileno 10 [1984] = Studi … Adelmo Barigazzi I, 237–40, ByzZ 82 [1989], 110–13), and N. Vakonakis, Das griechische Drama auf dem Weg nach Byzanz. Der euripideische Cento Christos Paschon, Tübingen 2011, 97–103 (preceded by an overview of the discussion since the editio princeps of 1542).
Cf. 164, 285, 532–3, 875–6, 938–40 (n. 330) nn. and see F. Jouan, in U. Criscuolo – R. Maisano (eds.), Synodia. Studia humanitatis Antonio Garzya … dicata, Naples 1997, 495–509. A. Döring (Philologus 23 [1866], 577–91) takes too many divergences for true variants.
Pack2 428 = Mertens–Pack3 454. Add F. Sisti, BPEC 27 (1979), 105–11 (109–11), W. Luppe, Anagennesis 2 (1982), 74–82 and van Rossum-Steenbeek, Greek Readers’ Digests?, 17–18, 201–2.
Pack2 = Mertens–Pack3 427. Cf. Wilamowitz, Einleitung, 215 with n. 186, P. Collart, BIFAO 31 (1931), 52–5, Turyn 97 n. 156, 313 n. 301.
By D. Obbink, who edited the piece in R. A. Coles et al. (eds.), The Oxyrhynchus Papyri LXVII, London 2001, 62. Cf. Mertens–Pack3 427.01.
(Re-)edited recently by Merro. Schwartz did not publish the notes in Q.
Alcestis, Medea, Hippolytus and Andromache had already been printed by Janus Lascaris in Florence (ca. 1495). Aldus omitted the Electra.
RhM N.F. 118 (1975), 205–25, partly superseded by M. Sicherl, Griechische Erstausgaben des Aldus Manutius. Druckvorlagen, Stellenwert, kultureller Hintergrund, Paderborn et al. 1997, 291–309.
Cf. Hec. 1100, where 
in the exemplar Par. suppl. gr. 212 was corrected to the metrically required ἀμπτάμενος (M. Sicherl, RhM N.F. 118 [1975], 218–19 = Griechische Erstausgaben, 301–2). In Rh. 429 πορθμεῦσαι, also found in Q 
L: 
Δ), has some claim to being right (428b–9n.).
J. A. Spranger (ed.), Euripidis quae inveniuntur in codice Laurentiano Pl. XXXII, 2, Florence 1920 and id. (ed.), Euripidis quae in codicibus Palatino Graeco inter Vaticanos 287 et Laurentiano Conv. soppr. 172 (olim Abbatiae florentinae 2664) inveniuntur, Florence 1939–46.
At http://opac.bmlonline.it/ (via ‘manoscritti’).
In high resolution via http://www.papyrology.ox.ac.uk/POxy/.
CQ n.s. 9 (1959), 135–6.
Cf. W. S. Barrett, CQ n.s. 15 (1965), 63 = Collected Papers, 444. Of the ‘alphabetical’ plays, complete or partial narrative hypotheses survive for Cyclops, Heraclidae, Heracles, Iphigenia in Tauris, Ion and Helen (preceded by Byzantine elaborations and added in P only by Ioannes Catrares). On the transmission of the material see Zuntz, Inquiry, 140–4.
The fact that in Rhesus the wolf-costume is not part of Hector’s orders (Liapis, ‘Notes’, 48) need not be an objection, since the hypotheses are not always faithful in their reports (cf. below on ll. 14–15, and see e.g. Barrett, Hippolytos, 153, Parker, Alcestis, 47).
See H. Grégoire, AC 2 (1933), 97–8 for the apparent dependence of the astronomical remark (ll. 22–3) on commentaries resembling our scholia (cf. ΣV Rh. 528 [II 340.5–341.13 Schwartz = 105–7 Merro]).
See Pfeiffer, History of Classical Scholarship I, 192–6, Barrett, Hippolytos, 153, van Rossum-Steenbeek 32–4. The best-preserved examples of Aristophanic hypotheses are those to Seven against Thebes, Philoctetes, Alcestis and Medea (Zuntz, Political Plays, 131, 139–40 with n. 6).
Cf. A. L. Brown, CQ n.s. 37 (1987), 430 (who otherwise is too critical of the synopses). Liapis (on Hyp. (c) Rh. 50–1) quotes LSJ s.v. διαλαμβάνω III 7, which is different (‘state distinctly’).
Nevertheless, the chorus behave almost as if somebody had to be called out of the stage-building (2–3, 11–14nn.).
See Ritchie 64–5, R. S. Bond, AJPh 117 (1996), 265–6, Fantuzzi, in I luoghi, 244–56 and Entretiens Hardt LII, 148–51.
Strohm (273 n. 1) gives as a parallel Med. 1122–3 Μήδεια, 
ϕεῦγε, 
ναΐαν / λιποῦσ᾽ 
μήτ᾽ ὄχον πεδοστιβῆ.
On our poet’s probable acquaintance with the whole passage see Introduction, 34.
See Radt’s note and add M. L. West, CQ n.s. 50 (2000), 342 = Hellenica II, 234 to those who would assign the fragment to the prologue.
Rhesus’ entry is prepared for by the Shepherd scene (264–341), but those of Odysseus and Diomedes (565), Athena (595) and Paris (641) ‘are all deliberately surprising’ (Taplin, Stagecraft, 63–4 n. 4).
In Textkritische Studien (1921/22) he returned to the paradosis, but with 5–6 ὃς … 
for allegedly better grammatical reference to the chorus.
Diggle, after Reeve (GRBS 13 [1972], 253–4 n. 21), deleted the lines. They are defended by Mastronarde on Phoen. 1070–1.
For the accent see 401–3n.
All or part of the Phoenissae exodos (1582–1766) has been considered spurious, and 1604–7 in particular were deleted by Hartung. Cf. Fraenkel, Zu den Phoenissen, 89–90, M. D. Reeve, GRBS 13 (1972), 463 and, for the defence, Mastronarde on Phoen. 1604.
Diggle follows him, whereas Mastronarde rather feebly defends the lines (Contact and Discontinuity, 121–4 ~ on Phoen. 376–8, 376).
See Kamerbeek on OC 1059–64, Ll-J/W, Sophoclea, 247 and, in favour of the change, Jebb on OC 1059 ff. At Phil. 1146–50 read 
θῆραι 
οὓς ὅδ᾽ ἔχει / χῶρος οὐρεσιβώτας, / 
αὐλίων / ἐλᾶτ᾽ 
Auratus: μ᾽ 
codd., ἐλᾶτ᾽ Canter: 
codd.).
Taplin (PCPS n.s. 23 [1977], 126) refuted the idea of οὐκ ἔστι being an intrusive stage-direction (Murray, apparatus 17, followed by Zanetto, Ciclope, Reso, 136).
PV 172 καί μ᾽ οὔτι μελι | γλώσσοις πειθοῦς ἐπαοιδῆσιν and [A.] fr. 192.4 (P. Lyom.) λίμναν παντο | τρόϕον Αἰθιόπων, if sound, have a ‘quasi-caesura’ within the compound (West, GM 95 n. 56). For Ar. Pax 1002, Av. 523 and 536 see Dunbar on Av. 523 and 534–8.
It is possible that νυκτηγορία for the Greek meeting in ‘Arist.’ fr. 159 Rose (Aporemata Homerica) = Porph. ad Il. 10.194 ff. (I 145.22–146.4 Schrader) = ΣB* Il. 10.198 (III 431.25–432.3 Dindorf) was inspired by our play (Fantuzzi, in Ancient Scholarship, 43 n. 10, 52–3). But Fantuzzi goes too far in suggesting that ‘Rhesus adopts a pre-‘Aristotelian’ interpretation of the assembly of Iliad 10 in terms of anti-panic caution’ (53). See also 138–9 (n.).
Willink (‘Cantica’, 23 n. 8 = Collected Papers, 562 n. 8) accordingly suggests ‘more staccato punctuation’ than printed by e.g. Diggle, ‘with colons after χέρα (23), εὐνάς (24), αἴρειν (25a), ἀϕύπνισον (25b), λόχον (26) and ἵππους (27)’.
There is no argument here about the date of the play, as Willink wishes to see in Ritchie. Stinton’s observations on the rarity of 
- and its virtual restriction to Aeschylus and later Euripides (BICS 22 [1974], 88–95 = Collected Papers, 119–28) mainly concern verse-openings and such exceptional cases of synartesis as listed by Parker (20 n. 17).
Fully resolved e (
) sometimes occurs in Pindar’s ‘freer dactylo-epitrite’, as described by K. Itsumi, Pindaric Metre. The ‘Other Half’, Oxford 2009 (especially 50, 98, 150–1, 354, 389, 434–6).
As do Wecklein and Dale (MATC I, 95) without it. The result is an awkward ‘hendecasyl-lable’ (
), which entails (possible) -ῠπν- in 25 and Attic νεῶν in 43 (Willink, ‘Cantica’, 24 n. 11 = Collected Papers, 563 n. 11).
Ant. 417–18 καὶ τότ᾽ ἐξαίϕνης χθονός / τυϕὼς ἀείρας σκηπτόν, the only example in iambic trimeters, may just add another Homeric touch to the Guard’s account (Griffith on 417–18). Lloyd-Jones and Wilson adopt Radermacher’s ἀγείρας, but see Hyps. fr. I ii.39 Bond = E. fr. 752f.39 
χθ[ον (with Bond on 39).
At Od. 20.49–50 εἴ περ πεντήκοντα λόχοι μερόπων ἀνθρώπων / νῶϊ περισταῖεν ‘ambush’ fits as well as elsewhere in epic (Björck, Alpha Impurum, 291–2; LfgrE s.v. B 3).
Panthous’ third son, Hyperenor, is ‘a mere cipher’ (Janko on Il. 14.511–2), whose death at Menelaus’ hands (Il. 14.516–19) becomes a convenient motive for Euphorbus’ fatal challenge to the Greek chief (Il. 17.24–42 [with Edwards on 24–8, 33–42]).
Cf. ΣT Il. 12.292–3 (III 359.49–50 Erbse) with apparatus, ΣA Il. 12.307 (III 362.20–1 Erbse).
Contrast individual ‘Panic’ frenzy at e.g. Med. 1171–3 καί τις γεραιὰ προσπόλων, δόξασά που / ἢ Πανὸς ὀργὰς ἤ τινος θεῶν μολεῖν, / ἀνωλόλυξε and Hipp. 141–4.
On Hp. Mul. 1.24 (VIII 64.5–6 Littré) ἢν δὲ ὁ γόνος ἀπορρέῃ διιπετής, καὶ μὴ λήγῃ, οὐ μίσγεται ἀσπασίως τῷ ἀνδρί. Cf. Emp. 31 B 100.8–9 DK ὥσπερ ὅταν παῖς / κλεψύδρῃ παίζουσα διειπετέος (Diels: διι- ZMil.) χαλκοῖο, where, despite Bollack (Empédo-cle III.2, Paris 1969, 485–6), the better-attested δι᾽ εὐπετέος seems impossible.
There may be a fifth-century precedent, if A. Henrichs (HSPh 79 [1975], 100–2 with n. 37) is right to find Democritean wording in Phld. Piet. P. Herc. 1428 fr. 16 (1–6 
[ | θέρος εν . . . [. . . . .] | χε[ι]μὼν καὶ ἔ[αρ καὶ] | 
ἄν<ω>θεν δι- | ειπετῆ γε<ί>νεται …).
Cf. B. P. Grenfell – A. S. Hunt, The Oxyrhynchus Papyri VI, London 1908, 21: ‘… ει and ι are unusually correctly written …’ This proves nothing for the original text, but rules out simple iotacism, which presumably accounts for διειπετῆ in Phld. Piet. P. Herc. 1428 fr. 16.5–6 Henrichs (n. 26) and one MS at Hp. Mul. 1.24 (n. 25).
OT 766 ἀλλὰ 
ἐϕίεσαι; differs by involving a neuter pronoun (KG I 310 n. 5, 352 n. 10), and at Xen. Ages. 11.14 
γοῦν 
μεγάλην καὶ 
ἐϕιέμενος <δόξαν, ἔστε> (suppl. Marchant) τὸ σῶμα 
ἐδύνατο 
ῥώμην the corruption may exceed the lacuna.
Cf. R. Goossens, Euripide et Athènes, Brussels 1962, 299 n. 71, M. Fantuzzi, CPh 100 (2005), 269 with n. 3, in I luoghi, 245 with n. 10, Ancient Scholarship, 45 n. 14, Liapis on 41 ff. (p. 84).
So already Hermann (Opuscula III, 285): ‘Oratio Hectoris v. 57 seqq. adumbrata est ex Iliad. θ 498 seqq.’ Cf. Ritchie 65.
For the text see Dawe, STS III, 52, L1-J/W, Sophoclea, 184, Second Thoughts, 105.
Previously among the marginalia in his working copy of Euripidis tragoediae … (31871). Cf. J. Družinina, Hyperboreus 5 (1999), 254.
Cf. ΣA Il. 10.45 (III 12.50–1 Erbse) καὶ ὅτι τὸ 
ὁ Ζηνόδοτος 
τὸν 
λόγον 
κόλον μάχην.
Unless the corrupt Phoen. 710–11 is emended in such a way as to eliminate Ἀργείων στρατόν (see Diggle’s apparatus and Mastronarde on 710–11).
On Homeric river similes see H. Fränkel, Die homerischen Gleichnisse, Göttingen 1921, 25–8, especially Il. 5.87–94 (Diomedes advances like a winter torrent) and 5.597–600 (Diomedes turns away from Hector like a man from a fast-flowing river).
Thus already Il. 8.485–6 ἐν δ᾽ 
ἠελίοιο, / ἕλκον 
ἄρουραν.
Note also Il. 15.718–25 (with Janko), where he blames unnamed ‘elders’ for keeping him from the ships for so long.
Often in fact 
is opposed to the veracity of the gods (Il. 24.220–4, OT 498–503, E. El. 399–400, Hel. 744–57 [as it stands], Phoen. 954–9).
C. W. Müller (ed.), Euripides. Philoktet: Testimonien und Fragmente, Berlin – New York 2000.
Spoken by the Scout. But the situation resembles Rhesus in that Tydeus, ‘raging and eager for battle’ (380), is discouraged by Amphiaraus from crossing the Ismenus on account of unfavourable sacrifices.
The eruption of Etna (Pyth. 1.15–28) was one of the best-known Pindaric episodes in antiquity. Cf. e.g. PV 351–72, Carc. II TrGF 70 F 5.6–7, [Longin.] De Subl. 35.4, Favorinus’ unfavourable comparison of Verg. Aen. 3.570–7 with Pi. Pyth. 1.21–6 (Gell. 17.10 ~ Macr. 5.17.7–14) and Strabo 5.4.9 = Posidon. fr. 39 Theiler.
The date and exact status of fr. tr. adesp. 90 are unknown. It may be comic paratragedy.
This rather than ‘anastrophic tmesis’, which is already rare in Homer (KG I 531, SD 425) and in genuine tragic verse restricted to Pers. 873/4 αἱ κατὰ χέρσον ἐληλαμέναι 
πύργον (
Heath: 
fere Ω) and, despite Barrett (on Hipp. 548–9), Phil. 343 
με 
ποικιλοστόλῳ μέτα.
But see Ag. 1306 (with Fraenkel), Cho. 885 (with Garvie) τί δ᾽ ἔστι χρῆμα;
Especially if the sequel, Il. 15.64–77 Πηλείδεω Ἀχιλῆος. ὃ δ᾽ …, is omitted with Zenodotus (ΣAT Il. 15.64 [IV 23.58–60, 62–3 Erbse] ~ Eust. 1006.2–4). West follows Hentze in deleting 64–71 (cf. his apparatus on 64–77).
On the text see 56–8n. (p. 141 n. 34).
. 105 (II 330.16–17 Schwartz = 85 Merro) already cites S. fr. 896 εἴθ᾽ ἦσθα 
ἔργα τοῖς λόγοις ἴσα, which may, however, be truncated or corrupt (Pearson and Radt nn.).
In Or. 717–18 
γυναικὸς 
/ 
᾽ οὐδέν, ὦ κάκιστε, 
ϕίλοις, cited by Matthiae (above), the first infinitive (without article) parallels the accusative of respect 
.
Diggle deletes Hel. 991–5 with Schenkl (991–2 already Hartung) and Reeve (GRBS 14 [1973], 154–5). The lines are defended by Dale (on 993), Burian (991–5) and Allan (991–5) and silently retained by Kannicht.
Unlike Hartung’s 
(17 [1852], 28, 122), which is not attested in tragedy and has only very late credentials for the required sense (LSJ s.v. 
I 1, 2).
Only at A. fr. 282 = Ar. Ran. 1291 
ἰταμαῖς κυσὶν ἀεροϕοίτοις. It becomes frequent in Middle and New Comedy (LSJ s.v. ἰταμός).
The same error runs through most witnesses at Rh. 106.
Paley’s assumption of an ellipse ἐλπίδι (on 105) is even harsher and not supported by passages like S. El. 1460–1 ὡς τις 
κεναῖς 
/ 
τοῦδε, Thuc. 1.81.6 
γὰρ ἐκείνῃ 
ἐλπίδι 
ὡς 
ὁ 
and [Lys.] 9.21 τίνι γὰρ 
δεῖ 
συμπολιτεύεσθαι. An equivalent to 
(with present infinitive) = ‘deem, suppose that …’ (LSJ s.v. 3) would be needed.
Halcyon or On Metamorphosis is a late-Hellenistic ‘Socratic’ dialogue, which Nicias of Nicaea (apud Ath. 11.506c) and Favorinus (apud D. L. 3.62) ascribed to Leon the Academic (otherwise unknown). Feickert also cites Hyp. fr. 205 δεῖ 
ἐκ 
ἐν 
καταστάσει εἶναι 
ἡλικίας, 
πυνθάνεσθαι, 
γυνή, ἀλλὰ 
μήτηρ, where 
adds suitable emphasis.
This is found in tragedy only at Tro. 981–2 
/ 
κοσμοῦσα, 
(suppl. Seidler). More commonly independent 
+ subjunctive of something that may happen (above): A. Suppl. 357 (with FJW on 375–8), 399–400, Ag. 131–4, 341–2 (with Fraenkel on 341), Alc. 315–16, HF 1399, Or. 776 (
<οὐ> Brunck).
Ba. 982–4 
is corrupt (Dodds on 982–4). For 
see Rh. 512–17n.
In his second edition (‘Addenda et Corrigenda’, liv) Porter comes to favour θράσει (OgV).
Cf. [Heraclit.] 22 B 136 DK = Orac. Chald. 159.2 des Places 
This piece has also been assigned to iambus, but see TrGF II, 297. Fr. tr. adesp. 733 fr. a 14 (= FGrHist 153 IV 7.10) does not preserve, or never has been, original tragic verse.
The phenomenon is far more frequent in comedy (West, GM 80). Passages like Sept. 369–652, where three strophic pairs near-symmetrically alternate with the Scout’s descriptions of the Seven and Eteocles’ replies, or Ag. 1401–47, where two choral remarks (1407–11 ~ 1426–30) punctuate Clytaemestra’s iambic speeches, are perhaps better regarded as epirrhematic compositions (Ritchie 330). But one may add to the cases above S. fr. 314.243–50 ~ 290–7 and 329–37 ~ 371–9 from the satyric Ichneutae.
Unless we read and divide with Willink (CQ n.s. 38 [1988], 96 = Collected Papers, 113–14): HF 1068/9–70 (Χο.) 
πρέσβυ. (Αμ.) σῖγα σῖγα, παλίντροπος ἐξεγειρόμενος 
/ 
(Willink: -ον L) 
Changing 
avoids creating by emendation split resolution in iambics or a dochmius Kaibelianus, depending on whether one takes 1070 ϕέρ᾽, … as 2ia mol or kδ δ. But the adjective seems better with δέμας, and it is in any case not clear that the usual colometry is inadmissible because of placing period-end after ‘hortatory’ 
(Wilamowitz) in 1069.
See L. P. E. Parker, CQ n.s. 18 (1968), 241–2, 253 and Diggle, apparatus Or. 329–30, 345–6.
I.e. with Willink’s colometry at Hipp. 877–8 
/ 
(CQ n.s. 49 [1999], 420 = Collected Papers, 289; cf. Diggle, Euripidea, 475–6 n. 158). The split would be virtually non-existent between 
and γάρ.
It falls short of Diggle’s category (c) (Euripidea, 377) in that the words repeated from the beginning do not conclude their respective dochmiacs.
Bothe’s two German translations of Euripides with textual notes seem to have been unknown to editors before D. Sansone (GGA 230 [1978], 237, QUCC n.s. 1 [1979], 157–9). Cf. Diggle, Euripidea, 518–23.
Porph. ad Il. 10.194 ff. (I 146.1–4 Schrader) = ΣB* Il. 10.198 (III.432.1–3 Dindorf). Cf. ΣT Il. 10.194 (III 38.57–8 Erbse) 
ἐκίνησαν, ΣA Il. 10.194 (III 38.64–6 Erbse) 
, 
Porph. ad Il. 10.194 ff. (I 146.26–7 Schrader).
R. Renehan’s objection (CPh 87 [1992], 373) that 
‘refers specifically to the surface of the earth’ and so cannot denote the soil in which one is buried, is not invalidated by E. Suppl. 829 
γᾶς ἕλοι and Rh. 962 οὐκ εἶσι 
πέδον. In both places the idea of ‘going below’ remains.
Cf. particularly Il. 15.494–9. On ‘Patriotism in the Homeric World’ see P. A. L. Greenhalgh, Historia 21 (1972), 528–37.
Cf. ΣT Il. 10.317 (III 67.85–6 Erbse) ὡς 
δειλὸς 
καὶ ῥιψοκίνδυνος (~ Eust. 808.49–50). The adjective ῥιψοκίνδυνος recalls Rh. 154–5 (n.) 
… τόνδε κίνδυνον θέλω / ῥίψας 
ναῦς 
Ἀργείων μολεῖν.
On the ‘communal’ aspect of Agamemnon’s and Achilles’ conflict see W. Allan and D. Cairns, in N. Fisher – H. van Wees (eds.), Competition in the Ancient World, Swansea 2011, 113–46 (especially 121–30).
Cf. M. Fantuzzi – D. Konstan, in E. Bakola et al. (eds.), Greek Comedy and the Discourse of Genres, Cambridge 2013, 256–74, who relate the device to the ‘series of false guesses or assumptions’ that feature in several Euripidean plays (e.g. Alc. 512–21, 803–21, Hec. 667–82) and detect a possible allusion to our scene in Men. Peric. 271–91 (Introduction, 44).
Alternatives for the end of 40 are Vollgraff’s 
(Mnemosyne n.s. 42 [1914], 83) or πάρ[εστι νῦν / 
(Mnemosyne n.s. 43 [1915], 72).
Or. 714–16 are interpolated and/or corrupt. Pylades is announced by Orestes (725–8) before he initiates the new dialogue at 729. Cf. Paris in Rh. 627–41 / 642 (below).
Following a suggestion by Di Benedetto and after H. Leidloff, De Euripidis Phoenis-sarum argumento atque compositione, Holzminden 1863, 24. Some of the supposed additions resemble Rhesus in their derivative technique: Phoen. 1310–11 ~ OC 1254–6, Phoen. 1315–16 ~ Ant. 1258 + 1297, Phoen. 1348–9 ~ Ant. 1281–2 + S. El. 1189 (Fraenkel 76–81, 82–3; cf. Introduction, 38–9). Mastronarde’s defence of the passage (on Phoen. 1308–1479, 1308–53) does not convince.
The explanations show that if there is some (etymological) link between κίνδυνος and dice / dicing (GEW, DELG s.v. 
; J. Knobloch, Glotta 53 [1975], 78–80), it had been forgotten by the second century AD.
At Med. 725, whether genuine or not, τοσόνδε (BOCDE) offers itself as an alternative to τόσον γε (fere AVP: τὸ σόν γε L). Cf. C. Austin – M. D. Reeve, Maia 22 (1970), 14, Collard on E. Suppl. [899–900].
Strictly speaking, Liapis (on 162–3) is right that διπλοῦς does not mean ‘mutual, reciprocal’. But the sense is implied at Ant. 14 μιᾷ θανόντοιν 
διπλῇ χερί (i.e. Eteocles and Polynices).
Followed by Chr. Pat. 1969 σύ τ’ εὖ παραινεῖς καὶ σὺ καιρίως λέγεις (~ Rh. 339). As often, the excerptor slightly altered the original for his text (Introduction, 53).
ΣV Rh. 165 (II 331.18–19 Schwartz = 85 Merro) comments 
τὸ οἴεσθαι ὅτι 
αἰτήσει.
Diggle follows Stevens in deleting Andr. 1279–82. The lines are defended by A. H. Sommerstein, CQ n.s. 38 (1988), 243–6.
In view of the standard Homeric Ὀϊλεύς, Proclus would hardly have written 
in prose, if it had not stood in his epic text. And cf. the inscription ΑΙΑΣ ΙΛΙΑΔΕΣ on an early-sixth-century Attic amphora picturing another scene from the Iliou Persis, Polyxena’s death (H. B. Walters, JHS 18 [1898], 282–8 ~ LIMC I.1/2 s.v. Aias II 8).
An. Gr. I 210.6 Bekker ἀποινᾶν· ἄποινα 
from a collection of rhetorical glosses probably refers to Dem. 23.28 (a law) τοὺς δ᾽ ἀνδροϕόνους ἐξεῖναι ἀποκτείνειν ἐν τῇ ἡμεδαπῇ … λυμαίνεσθαι δὲ μή, μηδὲ ἀποινᾶν and its explanation at 23.33 τὸ δὲ μηδ᾽ ἀποινᾶν 
χρήματα πράττεσθαι (‘to exact compensation’)· τὰ γὰρ ἄποινα χρήματα ὠνόμαζον οἱ παλαιοί (LSJ s.v. ἄποινα II 1).
Ποινάομαι (with an accusative object) is found only in IT 1433. But ποινή would have been familiar.
Cf. Rh. 122 (n.), where Achilles and Ajax share the epithet αἴθων.
This rendering also lies behind FJW on A. Suppl. 1 (II, p. 6), where our passage joins several Aeschylean examples (and one from Herodotus) of ἐϕοράω ‘with a connotation of protecting’.
Cf. Ba. 916, 956, 981. The words κατασκοπή and κατάσκοπος are rare in drama. Rhesus has eight cases of the latter for both the Trojan and the Achaean spies (125–6a n.).
LIMC III.1 s.v. Dolon B 2 (p. 661), III.2 (p. 525).
LIMC III.1 s.v. Dolon E 11 (p. 662), III.2 (p. 526).
See J. A. K. Thomson, CR 25 (1911), 238–9 (whose undue trust in a reconstruction of Dolon’s head with a helmet pervades the literature). The third vessel, an intact cup by the Dokimasia Painter of about 490–480 BC (St. Petersburg, Ermitage Б 1452), depicts the same scene without the gods (LIMC III.1 s.v. Dolon E 13 [p. 662], III.2 [p. 527]). Some scholars add a contemporary Attic black-figure oenochoe (Oxford, Ashm. Mus. G 251), but there the wolf-skin is merely tied round Dolon’s shoulders, and he seems to be wearing a skull-cap (LIMC III.1 s.v. Dolon 12 [p. 662], III.2 [p. 526]). In general, Wilamowitz, De Rhesi scholiis, 11 = KS I, 11 and F. Lissarague, RA n.s. 1 (1980), 3–30.
Too little of Eubulus’ comedy Dolon survives (frr. 29–31 PCG) to determine whether he appeared in it in disguise (Porter, xi).
Cf. also ΣbT Il. 10.23 (III 8.45–7 Erbse) Μενέλαον δὲ ὡς ἥττω παρδαλῆν ἐνδύει, Δόλωνα δὲ ὡς δειλὸν καὶ ἐπὶ λαθρίδιον πρᾶξιν ὁρμῶντα λυκέαν (with Hainsworth on Il. 10.334–5).
Ion 981 (L) is uncertain, since uniquely ‘the words’ would not ‘indicate an immediate intention to depart’ (Diggle, Studies, 101 n. 1). Various remedies have been proposed.
With Karsten’s ἐϕέστιος for ἐϕεστίους in 851. Od. 23.55 ἦλθε μὲν αὐτὸς ζωὸς ἐϕέστιος (compared by LSJ s.v. ἐϕέστιος I) is different in that the adjective alone connotes ‘home’.
Feickert (on 201) is wrong to speak of Trojan ‘tents’, which are nowhere mentioned in the play. Words like εὐναί, κοῖται and χάμευναι (9, 852) indicate the setting in an open camp (Introduction, 40–1).
One would indeed have to examine the Platonic MSS tradition to determine whether κλωπ- is a true varia lectio or a ‘conjecture’ perpetuated from the Aldine through the early editions and lexica.
Realism also lies behind the Homeric system. With the corslet on it would have been difficult to stoop and fasten the greaves, and the sword- and shield-straps might have got caught in the helmet’s plume (Kirk on Il. 3.330–8, BK on Il. 3.328–338, Hanson, The Western Way of War, 77). In Sept. 675–6 Eteocles asks for his greaves as the first item to be put on.
Diggle (PCPS n.s. 20 [1974], 6–8 + Studies, 48–9) wrote δεινοῦ for δεινῶ (L). R. Renehan (CPh 80 [1985], 155–6) defends the paradosis, with Wilamowitz’ interpretation of δεινῷ χάσματι θηρός as an explanatory apposition to πυρσῷ, ‘firebrand, torch’ (LSJ s.v. (A) I). But this metaphor for a ‘tawny mane’ is strained, and L(P) show further assimilative errors in this ode at 64, 372, 376, 377, 396, 398, 412 and 441 (cf. M. L. West, Textual Criticism and Editorial Technique, Stuttgart 1973, 23–4).
Hsch. χ 650 χορταιόβαμος· ὁ Σειληνός and Hsch. χ 651 Hansen–Cunningham χορταιοβάμων· χορταῖον τὸ ἔνδυμα τοῦ Σειληνοῦ probably derive from a single gloss on χορταιοβάμων (W. & L. Dindorf, in ThGL s.v. χορταῖος 8.1602 B), which has become fr. tr. adesp. 601.
E. El. 855–7 ἔρχεται δὲ σοί / κάρα ᾽πιδείξων οὐχὶ Γοργόνος ϕέρων / ἀλλ᾽ ὃν στυγεῖς Αἴγισθον, adduced by Feickert (on 222), is not parallel if, as it seems, κάρα … οὐχὶ Γοργόνος together are opposed to … Αἴγισθον (G. Gellie, BICS 28 [1981], 11 n. 12, D. Kovacs, CPh 82 [1987], 139–41, Cropp on E. El. 855–7).
Cf. Kranz, Stasimon, 264, Klyve on 224–63 (pp. 186–7) and Feickert on 224–63 (p. 153).
There is a weaker break between 
and D. See West, GM 134 n. 144.
Of the two plausible comic examples (Ar. Thesm. 1020, Eccl. 972 ~ 975b), the former stands in a parody of Euripidean monody, the latter in a sort of paraclausithyron (Parker, Songs, 442–3, 546–9).
In drama choriambic expansion of aeolic cola is ‘particularly Sophoclean’ (West, GM 118). The analysis of the last verse was in principle prefigured by Schroeder2 and Dale (below).
Likewise Schroeder2 (168), except that, with certain older editors, he leaves an intolerable mid-verse hiatus in the antistrophe by failing to divide after 248 … σαλεύῃ ~ 259 … ἐνέγκοι.
Unless Thgn. 5–10 is earlier. On the basis of historical allusions in the poems, West (Studies in Greek Elegy and Iambus, Berlin – New York 1974, 65–71) puts Theognis in the seventh to sixth century BC, while our late-antique and medieval sources give a floruit between the 59th and 57th Olympiads (552–41 BC). The date of the composite Hymn to Apollo is likewise disputed (Richardson, Three Homeric Hymns, 13–15).
T. R. Bryce (CJ 86 [1990], 144–9) claims that Hdt. 1.182.2 does not prove the existence of an Apollo cult in Lycia prior to the fourth century BC (whence Rhesus must be spurious) because ‘in referring to the oracle, Herodotos leaves the god in residence unnamed’ (146). But the opposite is likely to be true. Delphic Apollo could simply be ὁ θεός in the Histories (e.g. 1.19.2, 4.157.2, 7.148.2, 8.122), and at Alc. 112–18 (with Dale and Parker) the audience was expected to draw the right inference from a mention of Lycia alongside the oracle of Zeus Ammon in Thebes (~ Hdt. 1.182.2). See further A. W. Parke, The Oracles of Apollo in Asia Minor, London et al. 1985, 185–93.
The third (now largely discredited) ancient ‘etymology’ from lyk- (‘light’) appears to go back to the early Stoics (Cleanthes SVF I F 541, Antip. SVF III F 36), although Apollo had long been identified with Helios / the sun (Diggle on Phaeth. 225 [pp. 147–8]). See D. E. Gershenson, Apollo the Wolf-god (JIES Suppl. 8), McLean (VA) 1991, 17–19, 131–3.
Mantziou (101 n. 1) adduces Q. S. 8.290 δῖον Ἄρηα as a (very late) exception. For the stylistic rule see J. Wackernagel, Progr. Gött. (1912), 26–7 n. 2 = KS II, 993–4 n. 2 and West on Hes. Th. 991.
His intervention at Il. 10.515–22 is an emergency response and, significantly, comes too late (Hainsworth on Il.10.515–22).
On the athetesis of Il. 7.443–64 by Zenodotus, Aristophanes and Aristarchus (ΣΣAT Il. 7.443–64 [II 290.69–291.80 Erbse]) see Nickau, Zenodotos, 51–2, 178–80. ΣbT Il. 7.464 (II 293.28–30 Erbse) notes that it would be ‘strange’ (ἄτοπον) to have line 465 follow immediately upon 442.
It presumably goes back to Archilochus (frr. 105, 106 IEG) and Alcaeus (frr. 6, 208.1–9 Voigt) and is used with greatest effect by Aeschylus in Seven against Thebes (cf. Hutchinson on Sept. 62–4). See also Rh. 322–3 (321–3n.) ἡνίκ᾽ ἐξώστης Ἄρη ς / ἔθραυε λαίϕη τῆσδε γῆς μέγας πνέων.
Cf. App. Prov. II 85 (CPG I 411–12), which has the same source as part of our scholion (A. C. Pearson, CQ 11 [1917], 59). Also ‘Diogenian.’ II 47 (CPG II 25).
The relationship between Troy and her allies was not free of animosities (319–26, 762–9, 859a nn.). In our play these are personified by Rhesus and his Charioteer (Strohm 265).
Wilamowitz (Hermes 61 [1926], 282 n. 3 = KS IV, 409 n. 3) suggests that Hector tries to prevent ‘peasant talk’ of the kind we hear from the Goatherd and chorus of herdsmen in S. Poimenes (frr. 502 + 505). On the possible connection of this drama with Rhesus see below.
In E. Suppl. 638–9 λόγου δέ σε / μακροῦ ἀπολύσω Herwerden’s conjecture (Mnemosyne 5 [1877], 36) is preferable to L’s ἀποπαύσω (Diggle, Euripidea, 61 n. 11). The paradosis could only mean ‘I shall save you from much talking, i.e. from the need to ask many questions’ (Morwood on E. Suppl. 638–9), which would leave the expectation of brevity with the addressee.
That in 426–42 Rhesus complains ‘of the extreme difficulties he has had to face on his way to Troy’ (Liapis, ‘Notes’, 62) is a different story. There seem to have been no problems after crossing the Hellespont.
Sept. 356 
ἐκ τῶνδ᾽ εἰκάσαι 
πάρα is severely corrupt. See Hutchinson on 356, West, Studies, 112–13 and Sommerstein, Aeschylus I, 188–9 (apparatus and n. 46).
Cf. W. Burkert, Weisheit und Wissenschaft: Studien zu Pythagoras, Philolaos und Platon, Nuremberg 1962, 296–7 = Lore and Science in Ancient Pythagoreanism, Cambridge (Mass.) 1972, 317–18 and Collard–Cropp, Euripides VIII. Fragments, 527–9 n. 1.
He might have added PV 300–1 πετρηρεϕῆ / 
(i.e. that of Oceanus) and S. fr. 332 (= Hsch. α 8426 Latte) 
δόμους.
On bilingualism in Greek literature generally see J. Werner, in P. Händel – W. Meid (eds.), Festschrift für Robert Muth …, Innsbruck 1983, 583–95 and Faulkner on h.Ven. 113–16.
From the dubious iambic prologue (IA 49–114). E. Suppl. 660–2 
/ … / … 
and Phoen. 522 
(cited, among others, by Stockert on IA 83) are not equivalent. The first opposes cavalry and chariots, and in the second horses and chariots again appear in different clauses, not as a unity or hendiadys.
Apotympanismos (practised in Athens) ‘involved clamping the criminal by the neck, wrists and ankles to a large board and standing the board up … in such a way that the condemned man’s feet did not reach the ground; he would then be left degradingly and agonizingly exposed, with no one permitted to come near him, probably until sunset when, if still alive, he would be strangled by tightening the neck clamp’ (Sommerstein, Aeschylus I, 437 n. 17).
Phoen. 1104–40 has been excised by Morus. Whether (partly) genuine or not, the passage was inspired by Sept. 369–685.
Aeschylus’ description of the ‘Seven’ was well-known (305–6a n. with n. 123), and that of Tydeus stands first. Sept. 384–5 is parodied at Ar. Ach. 964–5 (cf. Olson).
Cf. K. Schumacher, Großherzogliche Vereinigte Sammlungen zu Karlsruhe: Beschreibung der Sammlung antiker Bronzen, Karlsruhe 1890, no. 782 with plts. XVI, XXII, Daremberg–Saglio s.v. ‘frontale’, E. Kunze, VIII. Bericht über die Ausgrabungen in Olympia, Berlin 1967, 191, 195 with plt. 73.
Achilles’ retelling of a ‘Libyan’ animal fable was famous. Ar. Av. 808 gives the moral (~ A. fr. 139.4–5), and the whole fragment is quoted by the scholia on the preceding line. Numerous other sources transmit part of the passage or allude to it (Radt, TrGF III, 252–6).
On this form see J. Wackernagel, Gött. Nachr. (1902), 742 n. 1 = KS I, 132 n. 1, Schwyzer 550 and West’s apparatus on Il. 5.656
His rarer use of literal 
(attributive and predicative) is discussed in 673b–4n.
Klyve (on 336–41 [pp. 225–6]) had already placed a lacuna before 335 ‘containing a speech … spoken by the messenger, pleading for the acceptance of Rh[esus] as an ally’. Yet as with Rosivach’s (58 n. 12) idea to keep the MSS order and assume a gap between 336–8 and 339–41, it is difficult to see how this could have avoided repeating much that has already been said. And the asyndeton is no argument against 335 being a ‘one-liner’ (or the beginning rather than the end of a longer utterance).
One would not wish to ascribe the incongruous MSS text to our poet. Where alternative versions coalesce (e.g. Ar. Ran. 1251–60, 1431a/b, 1437–53), the confusion usually arose in the early tradition.
At least not in the intended way. Rhesus’ ascension to heroic cult in Thrace is unforeseeable at this stage.
Not a ‘palimbacchiac’, which is a syncopated form of trochaic (
). It is impossible to tell whether the first long of our unit is anceps or not.
The first criterion at least also applies to his other instances of ‘interweaved’ ionics. Sept. 321–32 ~ 333–44 and Ba. 556–75 simply begin with that metre.
Ritchie tries to play down the import of the device. But the halting effect remains, even if both stanzas are syntactically intelligible on their own (as is also the case at Pyth. 12.17, Nem. 9.31 and Isthm. 6.35).
Musgrave (on 364) had already conjectured ἐπιδεξίαις. But the adjective is normally of two terminations, and the form in -αις would jar with οἰνοπλανήτοις.
She quotes V. Bers, Enallage and Greek Style, Leiden 1973, 3: ‘In the vast majority of examples one can make some sense of the adjective taken with the governing substantive. ’ This is the same as saying that one must ‘think of the substantives as coalescing into a single compound’ (Wilamowitz on HF 468, adapted by Fraenkel on Ag. 504).
The general sense is not obscured by the corruption. To the possible remedies in Diggle’s apparatus add 
ϕίλος 
Βακχεῖε (Willink, apud D. Kovacs, Euripidea, Leiden et al. 1994, 146–7).
At PV 792 West (Studies, 308) wrote 
for the transmitted 
(-ου) 
ϕλοῖσβον. But see S. R. West, Hermes 125 (1997), 377–8 in favour of Girard’s 
ἄϕλοισβον (i.e. the Russian steppe).
In Cho. 589–91 βλάπτουσι καὶ πεδαίχμιοι / λαμπάδες πεδάοροι / πτανά τε καὶ πεδοβάμονα the choice of πεδαίχμιοι was probably motivated by πεδάοροι (a metrical necessity) and πεδοβάμονα (Garvie on Cho. 588–91 [p. 206]). πεδάοροι (cf. Alc. fr. 315 Voigt) is Portus’ correction of M’s non-existent πεδάμαροι.
Eum. 405 (Wilamowitz, Einleitung, 154 n. 63, Taplin, Stagecraft, 47, 77, 388–90, Sommerstein on Eum. 404–5) and IA 590–634 (cf. Barrett on Hipp. 1102–50 [p. 368 n. 1], Taplin, Stagecraft, 77 n. 1). According to ΣMTAB Or. 57 (I 103.14–17 Schwartz), ‘some actors’ also perverted Orestes by giving Helen a procession from Nauplia, either before the prologue or, as Willink (on 57 ff.) reasonably contends, at 71.
Except that he (like his attendants) probably also wears the Thracian ζειρά (440–2n. [p. 286]).
On the textual status of Phoen. 1104–40 see 305–6a n. (p. 231 n. 123).
In Cho. 794–9 (where Zeus is asked to help the ‘colt’ Orestes in the ‘chariot-race’ to avenge his father) and Phoen. 947–8 it is probably meant to arouse sympathy (E. Petrounias, Funktion und Thematik der Bilder bei Aischylos, Göttingen 1976, 110–11, 170–1). For Menoeceus the requirement of ritual purity may also play a part (Mastronarde on Phoen. 947).
The same is to be supplied at h.Cer. 236a–8 Δημήτηρ / … / 
(i.e. on Demophon) and Ag. 105–6 … ἔτι γὰρ θεόθεν 
(or -πνεύει?) / πειθώ, μολπᾶν ἀλκάν, 
αἰών (for which see Fraenkel on Ag. 106 and M. L. West, Lexis 17 [1999/2000], 43–4).
As analysed by J. Duchemin, L’ΑΓΩΝ dans la tragédie grecque, Paris 21968, 81 and M. Lloyd, The Agon in Euripides, Oxford 1992, 7–8.
First by Valckenaer, Diatribe, 97, 103–5. See further Ritchie 96–7, Liapis, xlv with n. 133 (literature) and on 447–9. Lamachus in Ar. Acharnians parodies some of Rhesus’ prototypes.
Cf. Introduction, 33. From Hector’s comment in S. fr. 498 that ‘it is pleasant to tire oneself out and exercise one’s arms’ Sommerstein deduces a scene in which Cycnus, like Rhesus at 451–3 and 488, offered to engage the Greeks alone (Sophocles. Selected Fragmentary Plays II, 180, 206).
Liapis (‘Notes’, 70–1) adds that the addresses suspected to be spurious follow the pattern 
+ vocative’, while Rh. 388 has χαῖρ᾽ and no ὦ. Cf. M. L. West, Glotta 44 (1966), 142 n. and on Hes. Th. 964.
The story is alluded to in Pl. Lg. 685c2–d2 and Cephalio FGrHist 93 F 1 (pp. 441–2).
It is not quite clear from Diggle’s discussion that 
for ‘poetry’ (to which, after Page, he traces the restored κόσμον δ᾽ ὑμεναίων) essentially denotes ‘an ordered and metrically defined sequence of 
(Noussia-Fantuzzi on Sol. fr. 2.2 G.–P.2 = 1–3.2 IEG). But the meaning in Phaethon remains ‘songs in honour of our master’s marriage’.
So P. Oxy. 2163, supplementing Ar. Ran. 992 and Harpocr. I 259.10–260.3 Dindorf. Blomfield (Aeschyli Persae …, Cambridge 11814, XIV) had already conjectured οὓς σὺ προπίνεις, to which Taplin (HSPh 76 [1972], 66 n. 27) plausibly added θάσσων.
Phaeth. 246–7 Diggle = E. fr. 781.33–4 
should be deleted from LSJ. Read κατὰ σταθμά (Blass) or 
(Rau).
So ΣV Rh. 419 (II 337.25–6 Schwartz = 99 a2 Merro) = Hsch. α 3875 Latte 
συνεχῆ 
opposes the main scholiast’s preferred interpretation. Also ΣL Rh. 419 (II 337.27 Schwartz = 99 a3 Merro) ἄμυστις εἶδος 
and ΣV Rh. 438 (II 338.22 Schwartz = 101 Merro) τὰς ϕιάλας as against ΣP Rh. 419 (99 a4 Merro) 
(Merro: 
P).
As to the play title, ἐν 
is Hermann’s (Opuscula V, 189) palmary emendation of ἐν 
(V). But it is not clear whether the lines belong to Euripides’ tragedy of that name (in which case we would have to restore trimeters by accepting Hermann’s lacuna between 
and 
ἀμύστιδας) or, as iambic dimeters, to a homonymous comedy by Eubulus or Philyllius (Kassel–Austin, PCG V, 199). The surrounding text allows either view (cf. Merro 207–8, where add that the similarity to Cyc. 416–18 may be coincidence – or support Euripidean authorship of the fragment). All we can safely say is that the scholiast (or his source) was struck by the parallel occurrence of πυκνός.
Pindar and Bacchylides greatly elaborated this theme. Cf. e.g. Pi. Ol. 1.110 
ὁδὸν λόγων, Pyth. 11.38–9 
… κατ᾽ 
τρίοδον ἐδινάθην, / ὀρθὰν κέλευθον ἰὼν τὸ πρίν, Nem. 6.45–6, Isthm. 3/4.19–21, Bacch. 5.31–3, 9.47–8, (O. Becker, Das Bild des Weges …, Berlin 1937, 68–85, M. R. Lefkowitz, HSPh 67 [1963], 243 with n. 44 = First-Person Fictions. Pindar’s Poetic ‘I’, Oxford 1991, 27 with n. 44, Maehler [1982] on Bacch. 5.31).
Phoen. 1–2 are rightly deleted by Haslam (cf. 305–6a n.). A possible source for the first verse is [E.] Ep. 1.1 (478 EG) 
(M. W. Haslam, GRBS 16 [1975], 173 n. 83).
The likeliest remedies are Hartung’s … χλωρόν <τι> δεῖμα θράσσει (θράσσει iam Markland), with θάρσος in 609, and Diggle’s ὡς χλοερόν μοι 
δεῖμα θάσσει (GRBS 14 [1973], 50–2 = Euripidea, 68–70), taking up Murray’s θάσσει and a corrector of P who wrote 
χλοερόν. See also Collard on 598–9.
Hec. 962 σὺ δ᾽, 
could stand for Hector’s entire speech. Note also the verbal resemblance in Rh. 467–8 
/ 
.
Including at Andr. 1260–2 
/ … 
/ 
, where Achilles’ blessed existence contrasts with the ‘inhospitable’ sea. So also IT 438 
κατὰ was ‘corrected’ to εὔξ- only by Triclinius, and at Pi. Nem. 4.49–50 
/ νᾶσον West (apud Henry on 49) plausibly conjectured Ἀξένῳ. Cf. S. R. West, G&R 50 (2003), 157.
M. Vasmer, ‘Osteuropäische Ortsnamen: I. Das Schwarze Meer’, Acta … Universitatis Dorpatensis B. 1.3 (1921), 3–6 = Schriften zur slavischen Altertumskunde und Namen-kunde I, Berlin 1971, 103–5.
An English adaptation in W. Leschhorn et al. (eds.), Hellas und der Griechische Osten … Festschrift für Peter Robert Franke …, Saarbrücken 1996, 219–24 = Selected Onomastic Writings, New York 2000, 158–63. Cf. Encyclopaedia Iranica IV s.v. ‘Black Sea’, 310.
In Phil. 645–6 ἀλλ᾽ εἰ δοκεῖ, χωρῶμεν, ἔνδοθεν 
/ 
σε χρεία καὶ 
ἔχει the sequence of a plural verb and singular participle should probably be kept (Ll-J/W, Sophoclea, 193–4; cf. Sommerstein on Eum. 142, Dunbar on Ar. Av. 202–4). Page’s 
for 
(PCPS n.s. 6 [1960], 51) at any rate would destroy the reference of 
… χωρῶμεν to the departure from Lemnos discussed before.
See Wecklein, Appendix, 51 and Liapis, ‘Notes’, 72–3. The best proposal is Musgrave’s ἑλκύσας for τὰς ἐμάς (on 438), on the analogy of Cyc. 417 … 
ἑλκύσας and indeed the Auge fragment in ΣV Rh. 419 (418b–19n. with n. 154). But the corruption would be difficult to explain. With its contemptuous undertone (below), τὰς ἐμάς does not look like a gloss (for glossing with possessive pronouns see M. L. West, BICS 26 [1979], 107 and BICS 31 [1984], 186).
Or. 564–71 were deleted by Kovacs (Euripidea Tertia, 88–90). Yet δείν᾽ at least could refer to Tyndareus’ entire speech (Di Benedetto on 571).
In Ar. Ach. 136–9 Theoros excuses his long absence in Thrace with bad weather conditions. The whole country was covered in snow and the rivers had frozen over.
In 471 the MSS again disagree (
Λ: σὺν Δ). Cf. 59 (59–60a n.) 
(συν- L), 684, 763.
He adduces Vater (on 439), who did not in fact advocate this solution. Hermann (Opuscula III, 304) had already refuted the idea that 
could here mean ‘stay, keep back’ (LSJ s.v. (A) A II 9), on the ground that the Greeks were not attacking (‘Non enim cohibendi erant, qui non instabant’). Cf. Liapis, ‘Notes’, 74.
Cf. Rh. 428–9 ἀξένου δ᾽ 
/ 
ἀκτάς, 
στρατόν (L: 
Δ), unless 
is to be read there (428b–9n.).
At S. El. 1197 οὐδ᾽ 
οὐδ᾽ ὁ κωλύσων 
; the dative object of 
(i.e. Electra) has to be supplied from the preceding line.
E. Holzner, Studien zu Euripides, Prague 1895, 108–9, whose 
is an unidiomatic complement.
This relationship between the stanzas has only ever been questioned by Pace (QUCC n.s. 65 [2000], 127–29, Canti, 41–3, 56–9) and Jouan (LXX, 48 n. 245, 74 n. 246), who ‘with uncritical tolerance of [metrical and linguistic] anomalies’ (Willink, ‘Cantica’, 33 = Collected Papers, 572) keep the MSS text and colometry in both 454–66 and 820–32.
Diggle (in his respective apparatuses) offers to turn HF 1058 into an ithyphallic (
· οἴμοι) and Tro. 311 ~ 328 into 2ia (< ⋃ -> 
ὁ γαμέτας ~ τυχαῖς. ὁ χορὸς ὅσιος <ὅσιος>), but does not quote Eum. 158 ~ 165. At Sept. 782 ~ 789 the length appears in an iambo-dactylic context, and another analysis than as dochmiac may be desired (Hutchinson on 720–91 [pp. 162–3], though his scansion of 781–2 ~ 788–9 as a short form of the ‘archilochian’ does not appeal either).
Diggle (TrGFS) creates two further specimens at Hyps. 276 (fr. 64 ii.91 Bond = E. fr. 759a.1612) and Antiope V.54a 
(= E. fr. 223.83 
[suppl. Blass] with Kannicht’s apparatus). A form with contracted first biceps appears to be in evidence at Ai. 399 ~ 416 (Ritchie 310 with n. 2, Finglass on Ai. 348–429 [p. 240]).
The form does not occur in other tragedy (Ritchie 179). At IA 207 metre favours λαιψηροδρόμον 
(Hermann: -ιλῆα L). See Diggle, Euripidea, 470–1.
V is defective from Rh. 941, but the reading appears in its apograph Va.
The thought that the Trojans may have turned the tables recurs in Verg. Aen. 2.192–4 and 11.285–7, as an allusion to Rome’s later conquest of Greece (Klyve on 471–2, Liapis on 422–53).
Several other Homerisms cluster mainly in the first half of the scene (cf. Introduction, 30). Perhaps the subject-matter of Odysseus’ adventures was considered ‘epic’ enough.
See Klyve on 388–526 (p. 262).
In O the later accusative plural 
intruded (cf. Schwyzer 563, 575, Threatte II 247–8).
Ignoring this usage, Liapis (on 480) takes the verb absolutely, which makes no sense: ‘Yes, and we are not complaining …’ (i.e. because we are resisting well so far).
In Il. 19.422–3 (above) 
could in theory go with οὐ λήξω. But the word order tells against this, and no ancient scholar suggested it.
See West, ed. Iliad I, XVI (with literature). Aristarchus advocated 
in epic, according to a dubious rule (ΣAbT Il. 5.203 [II 32.28–30 Erbse]; cf. Σ Od. 5.290 [I 273.26–8 Dindorf]). This intermittently appears in Homeric MSS.
Whether or not Xenophon wrote the appendix on the decline of Persia after Cyrus’ death (Cyr. 8.8), it can hardly be later than 330 BC without referring to the victory of Alexander the Great.
Eum. 101 (3ia) ends in μηνίεται, and despite Ritchie (288), it is likely that the present tenses in OC 965 and 1274 have an ‘Attic’ long ι in anceps position (cf. W. Schulze, Quaestiones Epicae, Gütersloh 1892, 351 with n. 3, Wackernagel, Sprachliche Untersuchungen, 140).
In a positive sense already ‘Hes.’ fr. 198.3 M.–W. 
εἰδώς. The adjective is a variant on 
at Od. 1.1. See LfgrE s.v. 
B 2, D.
On the alternative influence of Od. 13.397–403 (~ 13.429–38) on Odysseus’ disguise in Rh. 710–21 see 710–21, 710–11, 712–13a, 715–16 (ψαϕαρόχρουν) nn.
ΣV Rh. 503 (II 339.25–6 Schwartz = 104 Merro) 
τροϕήν. ἐπαίτης need not be more than an inference from the line itself.
On the choice of reading and in favour perhaps of δόλιος see 893b–4n. with n. 309.
Mette ascribed the lines to Aeschylus’ Niobe (A. fr. **157a). But given the speaker’s identity and the very detailed parody of Helen in Ar. Thesm. 855–919, it is more likely to be a paratragic confection with a nod to Euripidean style.
Μέρμερος is attested as a proper name in Il. 14.513 and Od. 1.259. The adjective is employed directly of a person in Pl. (?) Hp. Ma. 290e4.
In fourth-century Athens convicts for such offences were executed, their bodies left unburied outside the borders of Attica and their property confiscated by the state.
West (apparatus) refers to Pers. 446 
μόρῳ δὲ τούσδε ϕῂς ὀλωλέναι; But the same sort of repetition is found in the preceding exchange between the Messenger (436–7) and the Queen (439–40). It seems better to regard πότμῳ as a gloss (Garvie on Pers. 441–6).
In E. Suppl. 1135 ποῦ λοχευμάτων χάρις Musgrave’s restoration of the nonsensical πολυχευμάτων (L) is preferable to Triclinius’ ποῦ 
… (‘sleepless nights’). See Collard on 1135–8.
Dated from the trimeter resolution rate to ‘within a few years of 420’ (Diggle, Phaethon, 47–9; cf. M. Cropp – G. Fick, Resolutions and Chronology in Euripides: the Fragmentary Tragedies, London 1985, 87).
Cf. M. Hose, Studien zum Chor bei Euripides I, Stuttgart 1990, 129–31 on the ode’s priamel structure.
Some knowledge of astronomy was recommended for soldiers, not least for the purpose of regulating night-watches (Xen. Mem. 4.7.4; cf. Plb. 9.14.5–15.15). But in practice more reliable results were obtained by the use of a water-clock: Aen. Tact. 22.24–5 (with Whitehead [pp. 159–60], P. Pattenden, RhM N.F. 130 [1987], 164–74), Veget. Epit. rei mil. 3.8.17.
See Diggle on Phaeth. 63–101 (pp. 96–7), J. G. Fitch (ed.), Seneca’s Hercules Furens. A Critical Text with Introduction and Commentary, Ithaca – New York 1987, on 125–204 (pp. 158–63) and M. Billerbeck (ed.), Seneca. Hercules Furens. Einleitung, Text, Übersetzung und Kommentar, Leiden et al. 1999, on 125–204 (pp. 241–5).
Ussher’s and Seaford’s interpretation of Cyc. 611 and 616 (lec = ‸ia ia) as trochaic is not supported by the word-ends.
I.e. 4da | lec (or 2‸ia) | 2‸ia. Diggle (Hyps. 275–7 TrGFS) takes the lines as D | ‘cyren.’ | ‸ia pe.
A dithyrambic poet, otherwise unknown. ‘Guesses’ at his date ‘have ranged from 490 to 480 B.C. … to the end of the century … If the lines are a true reflection of the style of Antigenes, it would seem fair comment that he was much more like Bacchylides than Timotheus, and that the first half of the century is the likelier’ (Page, FGE, 11).
Laetitia Parker also points out that 530–1 ~ 549–50 could be seen as a longer version of 527/8 ~ 546/7, if, instead of e | - - || with ‘link-anceps’ (527–8/546–7n.), the sequence 
| 
there passed for some sort of iambic compound. But any attempt to make this rhythmically equivalent to 
- (i.e. 
) remains speculative, and we simply do not know how to distinguish … 
+ ith from a possible unsyncopated ‘enoplion’ 
- (with - = 
throughout).
I.e. Crates fr. 89 Broggiato. There is a pun on the scholar’s name (11–12 Schwartz = 106.7–8 a1 Merro = 12–14 Broggiato): καὶ ταῦτα μὲν ὁ Κράτης. ἔοικε δὲ ὑπὸ τῆς ϕράσεως ἀμϕιβόλου <οὔσης> κεκρατῆσθαι.
So doubtfully M. Breithaupt, De Parmenisco grammatico, Leipzig 1915, 30 (on fr. 16), 34 (on fr. 17).
Euct. Parap. Pisc. 29 (23 March) 
(A. Rehm, Das Parapegma des Euktemon, Heidelberg 1913, 23 = W. K. Pritchett – B. L. van der Waerden, BCH 85 [1961], 17–52 [36]). For brief accounts of the fixed-star phases see D. R. Dicks, Early Greek Astronomy to Aristotle, London 1970, 10–16 and West, Works and Days, 379–80.
The same objection applies to Liapis’ identification of 
… σημεῖα as the constellation Bootes and, more specifically, its brightest star Arcturus (on 527–36 [pp. 219–20]).
E.g. suppl. Diggle: Πλειὰ[ς 
σκοτία] Wilamowitz prob. Lloyd-Jones (CR n.s. 21 [1971], 344 = Academic Papers I, 455–6).
Cf. (of the planets) h.Hom. 8.6–7 (Ares) 
ἑλίσσων / αἰθέρος 
ἐνὶ τείρεσιν. On the author of this late work (Proclus or some earlier Neoplatonist like Porphyry) see M. L. West, CQ n.s. 20 (1970), 300–4 (+ Homeric Hymns, 17, 189 n. 53) and Th. Gelzer, MH 44 (1987), 150–67.
‘Synecphonesis and Consonantalization of Iota in Greek Tragedy’ is treated by A. Kap-somenos, in E. M. Craik (ed.), Owls to Athens. Essays on Classical Subjects Presented to Sir Kenneth Dover, Oxford 1990, 321–30. But most of his examples are controversial, and the only two ‘parallels’ for the phenomenon occurring in final position do not bear examination. At Pers. 850 
there is no reason to make concessions for the variant παιδὶ ἐμῷ, when the possessive pronoun could be a gloss (West, Studies, 88–90; cf. his apparatus), and OT 865–7 ἔργων τε πάντων, ὧν νόμοι 
/ ὑψίποδες, 
/ δι᾽ 
(fere codd.: οὐρανίᾳ ᾽ν / αἰθέρι Enger) presents independent metrical and linguistic problems (L. P. E. Parker, CQ n.s. 18 [1968], 253; cf. L1-J/W, Sophoclea, 100).
Note the reverse error at A. fr. 388.2 and Phoen. 296.
Cf. ΣV Rh. 541 (II 341.17–18 Schwartz = 107 Merro).
Cf. ΣV Rh. 540 (II 341.14–15 Schwartz = 107 Merro).
For this notorious couplet, of which 253 was omitted by Zenodotus and athetised by Aristophanes and Aristarchus, see West’s apparatus (with literature), Hainsworth on Il.10.253 and Nickau, Zenodotos, 54–5.
There is nothing ‘odd’ about this lyric blend of imagery and thus no need to take 
ϕοινίας as the blood-stained river-bed of the Simois (Liapis on 546–50). This usage of κοίτη, moreover, is very late.
Thus ΣV Rh. 547 (II 341.19–22 Schwartz = 108 Merro). Whether the latter is due to sheer proximity or the fact that no other accusative was available, we cannot tell. There is in any case no hint that the scholiast(s) read 
or μερίμνας, as Paley (on 547) supposed.
Note the various adaptations from Homer, which account not only for the similarities with our passage, but also with Rh. 535–7 (n.).
Tro. 956 belongs under I 2 ‘lookout-man’, ‘watcher’.
Fr. com. adesp. 439 Kock (LSJ s.v. 
with Suppl. [1996]) is now Archil fr. 124 (a) IEG and the verb confined to the surrounding text (Ath. Epit. 7–8).
Willink’s objection (on Or. [1315–16]) that ‘… 
is an unnaturally violent v[er]b in this context’ is countered by e.g. Od. 22.468–9 
and Call. Aet. fr. 75.36–7 Harder τὸ δ᾽ 
(with Harder on 37), whether or not Or. 1315–16 are to be excised as an actors’ interpolation (Diggle). The aorist participle at any rate is impossible, and 
lurks in 1312.
For the error see FJW on A. Suppl. 458 (II, p. 363).
A fragmentary papyrus hypothesis assigned to A. Aitn(ai)ai (P. Oxy. 2257 fr. 1.5–14 = TrGF III, 126–7, A. fr. 451t 1) records no less than five scene-changes there, but whether and how they coincided with exits and re-entries of the chorus or choruses cannot be established (Taplin, Stagecraft, 416–18). S. Achilleos Erastai, a satyr play, and probably Troïlus (TrGF IV, 165–6, 453) are mentioned at the opening of the same note (after a possible reference to Eumenides, ingeniously restored by Lobel) and may thus also have involved a shift of location.
J. S. Scullion (Three Studies in Athenian Dramaturgy, Stuttgart – Leipzig 1994, 109–28) rejects a scene change in the latter half of the play. But see Finglass, Ajax, 14–20 and on 815–65 (p. 379).
Parker (on Alc. 746) offers some reasons why the chorus might stay, receding into the background of the acting area.
The view advocated by Ritchie (118–19) and others, that in Phaethon the chorus followed Clymene into the palace at 226 Diggle = E. fr. 781.13 and re-entered at 270 Diggle = E. fr. 781.61, cannot be upheld (Diggle, Phaethon, 150, Taplin, Stagecraft, 376).
Less likely, our poet over-interpreted Odysseus’ prayer to Athena at Il. 10.281–2 δὸς δὲ 
as did Aristarchus, according to ΣA Il. 10.282 (III 61.37 Erbse) 
τὸ 
(Liapis on 575–6). In theory the Alexandrian could even have got the idea from Rhesus.
IT 991–3 and IA 1026, added by Ritchie, can be dismissed. The former is very probably corrupt and the latter falls with the rejection of Murray’s strange punctuation.
At Sept. 631–2 τὸν ἕβδομον δὴ τόνδ’ (fere codd.: τόν τ’ Blomfield) ἐϕ᾽ ἑβδόμαις πύλαις / λέξω, 
the paradosis can be saved as preparing for the apposition (Hutchinson on 631; cf. SD 209).
Dionysus’ off-stage revelation of his divinity in Ba. 576–603 with the following narrative to the chorus of maenads (604–41) also resembles an epiphany (cf. Bond on HF 815 ff.).
See W. S. Barrett, in R. Carden, The Papyrus Fragments of Sophocles, Berlin 1974, 171–235 (particularly 184–5).
D. Fitzpatrick, in Shards from Kolonos, 251–2, 258 and hesitantly already M. W. Haslam, in P. J. Parsons et al. (eds.), The Oxyrhynchus Papyri XLIV, London 1976, 2.
So U. v. Wilamowitz-Moellendorff, Aischylos. Interpretationen, Berlin 1914, 58–9, 246, followed by W. Nestle, Die Struktur des Eingangs in der attischen Tragödie, Stuttgart 1930, 36, H. J. Mette, Der verlorene Aischylos, Berlin 1963, 112, M. L. West, CQ n.s. 50 (2000), 345 = Hellenica II, 238 and Fantuzzi, Entretiens Hardt LII, 170.
His alternative suggestion that she comes through a concealed side-door can be ruled out.
See e.g. D. Seale, Vision and Stagecraft in Sophocles, Chicago 1982, 176 n. 3, D. J. Mastronarde, Cl. Ant. 9 (1990), 278.
Cf. Wilamowitz, Hermes 61 (1926), 287 = KS IV, 414, Strohm 261, Burnett, ‘Smiles’, 39 and R. S. Bond, AJPh 117 (1996), 270. Vater’s idea (Vindiciae, lv), revived by Taplin (Stagecraft, 366 n. 1) and Burlando (Reso, 81, 83, 84 n. 52), that Athena was represented as a disembodied voice has rightly been rejected by Ritchie (128–9) and others. There would be no parallel for such a procedure, and our poet is most unlikely to have missed the opportunity for a spectacular epiphany in mid-play. Moreover, this would not solve the problem of the fourth actor (595–641n.) because someone would have had to deliver Athena’s lines from off-stage (L. Battezzato, CQ n.s. 50 [2000], 371).
The problem had already been noted by K. Schneider (RE Suppl. VIII s.v. Ὑποκριτής, coll. 191–2), who rejects the three-actor rule, and Bond (AJPh 117 [1996], 270 n. 28), who offered no further comment.
The Nurse enters the palace at 946, while the Heracles procession becomes visible at 964. But the Old Man need not have been among the first, and song offers more opportunities for delay than spoken verse.
For the traditional opinion that Pylades and the Servant were played by the same actor see e.g. Garvie, Choephori, xlix. Poll. 4.109–10 is too corrupt and probably unreliable to permit further conclusions with regard to Memnon (cf. TrGF III, 236–7).
J. P. Poe, Philologus 148 (2004), 28–9, who does, however, follow Battezzato (369) in sending Odysseus off at 626.
Bothe’s ᾽λλιπόντε (3 [1824], 365–6), ‘having failed / fallen short (of your duty)’, for which cf. Ar. Pl. 859 … ἤνπερ μὴ ᾽λλίπωσιν αἱ δίκαι, would come closest to the paradosis. Less convincing proposals are listed by Wecklein, Appendix, 52.
Xen. An. 4.7.3 τῇ γὰρ στρατιᾷ οὐκ ἔστι τὰ ἐπιτήδεια, εἰ μὴ ληψόμεθα τὸ χωρίον is different in that the main clause refers to the present and future alike (KG I 138).
A papyrus fragment (P. Oxy. 2078) ascribed to Pirithous, of which the authorship has since antiquity been disputed between Euripides and Critias. See Introduction, 24.
The Antiope passage quoted there (now E. fr. 185.2) belongs under διαπρέπω.
Feickert (on 633), as more hesitantly Vater (617) and Ammendola (633), takes ὑπάρχειν … κατθανόντα as a mere periphrasis for κατθανεῖν (LSJ s.v. ὑπάρχω B I 5, KG I 38–9 n. 3, Schwyzer 811–12). But this would be rather weak in this context and hardly to the point.
Of the Homeric parallels quoted there cf. especially also Il. 23.782–3 
/ μήτηρ 
The notion is touched on in E. El. 979 
θεῷ; (Apollo) and Or. 1666–9 
Λοξία μαντεῖε, σῶν 
/ οὐ 
ἄρ᾽ ἀλλ᾽ ἐτήτυμος. / 
ἐσῄει δεῖμα, 
τινος κλύων / ἀλαστόρων δόξαιμι 
ὄπα.
Il. 2.23 is punctuated as a question in a number of MSS, and the Rhesus passage perhaps favours this interpretation.
I.e. TrGF V.2 (60) ΡΗΣΟΣ 642.13, the second spurious prologue to Rhesus.
In the former passage καλεῖ with appositional 
at verse-end may express a certain pompousness on the part of Dicaeopolis. In the latter Cobet’s 
καλοῦμεν (Mnemosyne 5 [1856], 246) is more difficult for the change of word-order involved. But his comparison with Hel. 1643–4 is striking.
Hec. 540 is corrupt from 538. The adjective has further been conjectured at Phaeth. 269 Diggle = E. fr. 781.60, though all recent editors read Bekker’s εὐμενεῖς.
Liapis (on 681/678–9) does not succeed in explaining away the plurals. Battezzato (Lexis 22 [2004], 279) adds that Diomedes’ failure to reappear would give the impression that he never left Rhesus’ quarters.
A scattered entry would probably be unexampled in fifth-century tragedy (Hutchinson on Sept. 78–181 [pp. 56–7]). Yet if reports like Vit. Aesch. 9 (TrGF III, TA1) τινὲς δέ 
ἐν 
ἐπιδείξει 
σποράδην εἰσαγαγόντα τὸν χορὸν τοσοῦτον ἐκπλῆξαι τὸν 
ὡς τὰ 
ἐκψῦξαι, 
δὲ 
refer to later revivals (cf. Introduction, 42 with n. 81), this may not have been unusual in the fourth century.
Despite repeated allusions to searching in the fragments of E. Telephus (E. fr. 727a fr. 1.8, 11, fr. 14.1), it is unlikely that, as Ritchie (129–30) wants to believe, this was acted out in a scene which may have inspired Acharnians or Thesm. 597–614 + 655–87 (E. W. Handley – J. Rea, The Telephus of Euripides, London 1957, 35–6, C. Preiser, Euripides: Telephos. Einleitung, Text und Kommentar, Hildesheim 2000, 433).
Cf. Wilamowitz, GV 265 and Th. Drew-Bear, AJPh 89 (1968), 397. The only other scene where a tragic chorus comes to grips with the actors on stage seems to be OC 813–86 (835, 856–7). In A. Suppl. 825–910 the Egyptians probably do not lay hands on the Danaids before 808, and the coryphaei of Agamemnon (1650–3) and Heracles (252–74) merely threaten. At Hel. 1627–41 the identity of Theoclymenus’ opponent is disputed (M. Kaimio, Physical Contact in Greek Tragedy. A Study of Stage Conventions, Helsinki 1988, 65–78). In favour of the chorus-leader (L) see 149–94n. (p. 174).
PV 114 is ambiguous (West, ed. Aeschylus, 505, Griffith on PV 114–19).
Sept. 109 is defended by Hutchinson. On Hel. 694–5 see Diggle, Euripidea, 184–6.
F. Lourenço (JHS 120 [2000], 136 n. 24) may be right to reject Phoen. 655b ~ 674b (cf. Mastronarde on 656 ‘Metre’ [p. 334], 655bis, 674bis–75), but his analysis of Ba. 578 and 584 as (iambic) lecythia is unlikely. Hel. 352 and IA 1288 are metrically and/or textually uncertain, and Hel. 358 depends on Diggle’s conjecture.
Liapis, who supports the excision, takes 685 as a botched imitation of Ar. Ach. 282, which could as well have been in our poet’s mind (675–91, 685nn.). He fails to account for 686 (cf. ‘Notes’, 86–8).
The actual distribution was anticipated by Musgrave (on 688), who prints a conjecture in the first half of the verse, and Nauck (II3 [1871], 332). Cf. L. Battezzato, Lexis 22 (2004), 280–1 with n. 12.
The fact that the sentries do not react appropriately and later know nothing of Rhesus’ death (Porter, xiv; cf. Mastronarde, Contact and Discontinuity, 81) could theoretically be attributed to their over-excitement and go unnoticed in the course of the play.
Feickert, with elaborate stage-directions (89; cf. on 675–91, 675–82, 683–91, 683, 685, 686, 689), leaves this role to Diomedes, an idea Battezzato (Lexis 22 [2004], 279–80, 281–2) would endorse, if the armour were shown on stage. Battezzato’s main contention that Diomedes speaks 683(a) and 685a, only to be superseded by Odysseus in 686b–91, founders on the difficulty of having one character answer a question addressed to another. Phil. 974 and 1293 are different in that Odysseus interrupts from outside, and so is PV 589, since Zeus, provided that he and not Prometheus is the recipient of the preceding line (Mastronarde, Contact and Discontinuity, 115–16; cf. Griffith on 588), would not be expected to reply to Io.
Note also Alc. 249 ≠ 177–8, 911–25 (with Dale and Parker on 248–9), Or. 1075–7 ≠ 765 (Willink on 1075), E. El. 164 ≠ 160, 279, 1160 and perhaps Hipp. 1183 ≠ 1212, 1229 (Barrett on 1183).
Dawe is probably right to delete 1402–3 and retain 1407–8 with a lacuna and some further corruption (STS III, 136–7; cf. West, GM 91). But Dindorf’s Binneninterpolation, accepted by most modern editors, does not affect the speaker distribution either.
From Hellenistic times on αἱ 
is attested as a general term to cover the Aegean islands other than the Cyclades. Ascriptions varied in detail, and Virgil (Aen. 3.126–7) could still write sparsasque per aequor/Cycladas. Its use for the Dodecanese alone is modern (O. Maull – L. Bürchner, RE III A 2 s.v. Sporaden, coll. 1857, 1871).
Zanetto and Pace (Canti, 51, 53) retain Δ’s 
as a hypodochmiac. However, the only example of this colon with a resolved first long (or perhaps rather a standard dochmiac of the form 
- corresponding with a hypodochmiac) is IA 246 
τε παῖς (~ 235 καὶ κέρας 
ἦν), which besides involving a proper name, comes from a passage of doubtful Euripidean authorship.
On 
(sic) ~ S. frr. 236, 312. See Radt’s apparatus and Pearson on 236.
Mastronarde, Contact and Discontinuity, 77 n. 6. For the following examples cf. 21 n. 7, 22 with n. 14 and 25 n. 28. The Phrygian Slave in Orestes, while sharing certain traits with the Charioteer (804–81, 866–7nn.) and being unable to give the entire story (Or. 1498–9), does recognise the presence of the chorus (1375) and replies properly to most of their questions (1393, 1425, 1453, 1473).
The scene parodies the final appearance of pain-stricken characters (and θρῆνοι) in general, but Hippolytus is the obvious Euripidean example. See also Rau, Paratragodia, 142–4.
The lines are most likely distorted by interpolation (Fraenkel, Zu den Phoenissen, 71–86). Cf. 149–94n. (pp. 175–6) with n. 74.
On the subjectivity of Euripidean messenger-speeches see de Jong, Narrative in Drama, 63–116. Cf. also 756–61n.
In view of the preceding two instances of extra-metric ἰὼ ἰώ (and especially the unambiguous 731), there is no need to see in 728b a first anapaestic monometer, as Diggle (Euripidea, 119) seems to suggest.
Liapis takes 728b as non-lyric (with Kirchhoff’s τύχη for τύχα), a trochaic tetrameter catalectic cut short by ϕεῦ ϕεῦ, as ‘the charioteer, for some time now in the throes of pain, is unable to fill out’ the line. But such ‘realism’ is alien to Greek tragedy. The sequence of lyric, recitative anapaests and iambic trimeters adequately represents the Charioteer’s regaining mastery of himself.
On all the preceding compositions and the division of lyric and non-lyric (iambic) metres between the characters see Barrett, Collected Papers, 386–419 (especially 389, 392–3, 398, 406–8, 409–10, 411).
The ascription of the fragment is not above suspicion (cf. TrGF III, 218), and West (CQ n.s. 50 [2000], 347–50 = Hellenica II, 241–6) went so far as to attribute the entire play to Aeschylus’ son, Euphorion.
The metrical corruption, for which see L. P. E. Parker, CQ n.s. 16 (1966), 23 and Diggle’s apparatus on 1179–80, does not affect the substance of the text.
The traditional order of the two Hippolyti, established from a short note in the hypothesis by Aristophanes of Byzantium, has been questioned (J. C. Gibert, CQ n.s. 47 [1997], 85–97) or even reversed on the basis of metrical evidence and new papyrus finds of a hypothesis to 
(κατα)καλυπτόμενος (G. O. Hutchinson, ZPE 149 [2004], 15–28, O. Zwierlein, Lucubrationes Philologae 1. Seneca, Berlin 2004, 57–90, especially 84–5). Against all three, however, Luppe (Philologus 142 [1998], 174 n. 3, ZPE 151 [2005], 11–14 + ZPE 156 [2006], 38) argues cogently for keeping the relative position, if not the date, of the extant play, whereas Cropp and Fick (ZPE 154 [2005], 43–5) doubt the evidential value of slightly lower resolution rates (compared to Alcestis, Medea and Heraclidae) in this case.
Hec. 214–15 ἀλλὰ θανεῖν μοι / συντυχία κρείσσων ἐκύρησεν belongs to an actors’ interpolation (211–15 del. Wilamowitz). For the metrical difficulty in 215 see Diggle, Euripidea, 315.
E.g. Ag. 1072–1330 (with Fraenkel, Ag. III, p. 623), OT 1307–1415, Alc. 244–392, Med. 96–266, Hipp. 198–430. Cf. Schadewaldt, Monolog und Selbstgespräch, 143–4, Zanetto, Ciclope, Reso, 155–6 n. 91, Finglass on S. El. 254–309.
I do not, however, agree with their respective ironic interpretations of the play.
St. Byz. α 570 Billerbeck νῆσος 
EM 451.51 βασίλεια χείρ and 461.44–5 Πολυδεύκεια χείρ, ᾿Αγαμεμνόνεια give doubtful evidence, especially since Hdn. II 454.15 Lentz has Πολυδευκεία χείρ καὶ Ἀγαμεμνονεία ναῦς in the same context. A. Meschini (AFLPad 1 [1976], 180) further cites A. D. Coni. 233.7–9 Schneider ἦν οὖν καὶ τὸ εὐήνωρ, <ἀϕ᾽ οὖ> κτητικὴ παραγωγὴ 
εὐηνόρεια ὡς Ἑκτόρεια κ . . ρεια καὶ πρακτόρεια, εὐηνορέα (Schneider: -όρεια Ab) καὶ ἔτι 
εὐηνορ<έη>, ὡς Ἑκτορέη.
The lines have caused much confusion among scholars ancient and modern. A succinct account of their problems and likeliest interpretation (with reference to earlier works) is given by A. H. M. Kessels, Studies in the Dream in Greek Literature, diss. Utrecht 1973, 30–3.
Much emphasis has been put on the observation that, while Iliadic dreamers are always male, structurally important dreams in the Odyssey and tragedy (see also Eum. 94–139 and PV 645–57) usually occur to women (W. S. Messer, The Dream in Homer and Greek Tragedy, New York 1918, 8, 27–8, 51, 65; cf. J. Hundt, Der Traumglaube bei Homer, Greifswald 1935, 42 n. 7). However, as Kessels (n. 279), 113–15 points out, the Homeric discrepancies can reasonably be ascribed to the different roles the sexes play in each poem, whereas hardly enough of Greek tragedy survives to judge whether our epic-based passage constitutes a remarkable deviation from the ‘norm’.
At Ba. 1061 ὄχθων δ᾽ ἔπ᾽ ἀμβὰς ἐς ἐλάτην ὑψαύχενα 
ἀμβὰς Bruhn: ἐπ᾽ ἐμβὰς P, ἐς Heath: εἰς P) the Aldine offers ὄχθον δ᾽ ἐπέμβας, which could be fitted into the sentence by combination with Musgrave’s and Tyrwhitt’s ἢ ᾽λάτην. But Dodds (on 1061) rightly remarks that this is farther from the paradosis than Bruhn’s proposal, and also accords less well with the topographical details given in 1048 and 1051–2.
E.g. Pi. Pyth. 4.178, 6.37, Ai. 134 (recitative anapaests), Ant. 950, Ba. 154, 568. More Pindaric examples are listed in P. Maas, Epidaurische Hymnen (SKGG 9.5), Halle/Saale 1933, 10 with n. 5 and H. Maehler (ed.), Pindari Carmina cum Fragmentis, Pars II. Fragmenta. Indices, Leipzig 1989, 188.
The words are rightly distinguished by A. K. Orlandos – I. N. Travlos, Λεξικὸν Ἀρχαίων Ἀρχιτεκτονικῶν Ὅρων, Athens 1986, since whereas ἀντηρίς is related to ἀντερείδω (hence properly ‘stay, support’), 
must be a substantival feminine of ἀντήρης, expanding on the sense ‘opposite, facing’ (i.e. the street).
Wellauer’s νείρᾳ (after Portus’ νείρῃ) at Ag. 1479 ἐκ τοῦ γὰρ ἔρως αἱματολοιχὸς 
τρέϕεται(·) is very uncertain (Denniston–Page on 1477–80, West, Studies, 222), and Hsch. ν 245 Latte νειραί· κατωτάται. οἳ δὲ κοιλίας τὰ κατώτατα belongs to a secondary adjective (cf. Hsch. ν 246 Latte νειρὴ κοιλίη· κοιλία ἐσχάτη) attested also in the masculine (Lyc. 896) and neuter (Hsch. ν 247 Latte νειρόν· σϕοδρόν. ἔσχατον).