^{page 139 note 1} Or between two only in the earlier tragedies of Aeschylus.

^{page 139 note 2} The word ‘regular’ is used to signify an actor of the same standing as the three actors granted by the State, as distinct from a supernumerary, mute, etc.

^{page 139 note 3} The validity of the three-actor rule is here accepted, although the arguments of Rees, , The So-Called Rule of Three Actors in the Classical Greek Drama, Chicago, 1908Google Scholar, are considered below.

^{page 139 note 4} Aesch. Eumeniden, 1833, p. 172, n. 9.

^{page 139 note 5} Geschichte der griechischen Litteratur, 1st ed. 1841, ii, p. 55Google Scholar; 4th ed. 1882, i, p. 510; English trans., 1840, p. 305; 1858, i, p. 403.

^{page 139 note 6} Schultze, K. F. R., De chori Graecorum tragici habitu externo, 1856, p. 60Google Scholar; Müller, A., Lehrbuck der griechischen Bühnenalterthümer, 1886, p. 175Google Scholar; Das attische Bühnenwesen, 1902, p. 68; wood, G. Nor, Greek Tragedy, 1920, p. 167, n. 4Google Scholar; Völpel, R., ‘Zum Schauspielerproblem’, Rheinisches Museum für Philologie, lxxvi, 1927, p. 16Google Scholar; Willem, A., Melpomène, 1932, p. 149Google Scholar; Lesky, A., Die griechische Tragōdie, 1938, p. 125Google Scholar.

^{page 139 note 7} Die dramalische Technik des Sophokles, by von Wilamowitz, Tycho (Philologische Unter suchungen, 22, Berlin, 1917), chapter viiGoogle Scholar, ‘Oed. Col.’ by von Wil, Ulrich., p. 333 textGoogle Scholar and n. 1; Griechische Tragoedien übersetzt von Ulrich von Wil., xiv, Die griechische Tragoedie und ihre drei Dichter, 1923, p. 128 nGoogle Scholar.

^{page 139 note 8} Oedipus Coloneus, 3rd ed. 1900, p. 7.

^{page 140 note 1} The Messenger's part may equally well be here allotted to the first actor.

^{page 140 note 2} Histoire de la Littérature grecque, iii (3rd ed. 1913), p. 258, n. 1Google Scholar. An almost identical distribution was hinted at by Bergk, Th., Griechische Literaturgeschichte, iii (1884), p. 84Google Scholar.

^{page 140 note 3} Die Vertheilung der Rollen unter die Schauspieler der griechischen Tragoedie, Berlin, 1842, pp. 51–2Google Scholar.

^{page 140 note 4} ‘Ueber die Rollenvertheilung im Oedipus auf Kolonos des Sophokles’, Philologus, xii, 1857 (pp. 750–4), p. 752Google Scholar. Richter in his distribution gave the parts of the Stranger, Theseus, and the Messenger to a παραχορήϒημα or supernumerary; but a παρ, (discussed more fully below) could undertake only short extra parts, and Richter was really postulating a fourth regular actor; cf. Ascherson, (p. 752), ‘Wenn Richter den Polyneikes dem παραχορἠγημα … gegeben hätte, so hätte er mit vier Personen auskommen können’Google Scholar.

^{page 140 note} De Mensura Tragoediarum, Berlin, 1822, p. 45Google Scholar. Like Richter (cf. n. 4), Lachmann incorrectly assigned to a παραχορἠγημα roles of such length (in this case those of the Stranger, Ismene, and Creon) that they could be played only by a fourth regular actor (although both scholars had in fact intended to give merely three-actor distributions). Ascherson (pp. 751–2) notes that in Lachmann's arrangement either the Stranger or the Messenger or both may belong to the third actor: if both, the cast becomes identical with Jebb's.

^{page 141 note 1} The Rhesus, far from needing four actors, can without difficulty be played by three and was possibly intended for only two (cf. Paley, , Euripides, i, p. 9Google Scholar). Mūller, K. O., Aesch. Eum., p. 172, n. 9Google Scholar, ‘Es ist merkwürdig, dass, wie Aeschylos erst in seiner letzten Trilogie, der Orestea, drei Schauspieler zuliess, so wiederum Sophokles erst am Ende seiner Laufbahn, in dem Oedipus auf Kolonos, einen vierten hinzugenommen hat’; cf. ProfRose, H. J., A Handbook of Greek Literature, 1934, p. 171Google Scholar, ‘the fact that it needs a fourth actor fits very well with the tradition that it is the work of Sophokles' old age’.

^{page 141 note 3} Hermann, J. G., Sophocles, vol. ii, 1827, pp. cclxxxii–cclxxxviGoogle Scholar.

^{page 141 note 4} Cf. Siess, H., ‘Chronologische Untersu chungen zu den Tragödien des Sophokles’, Wiener Studien, xxxvi (1914), pp. 244–94Google Scholar, xxxvii (1915), pp. 27–62: after a thorough examination of many aspects of Sophocles' style she concludes (p. 62), ‘der O.C. hingegen etwas älter als Phil.’: Jebb, , Soph. El., p. lvii, n. 2Google Scholar, ‘The extraordinarily high proportion in the Philoctetes (409 B.C.) must be considered as indicative of the poet's latest period’.

^{page 142 note 1} Cf. Guglielmino, F., Arte e Artifizio nel Dramma Greco, Catania, 1912, p. 87, n. 1Google Scholar. The technical names for the three actors have been avoided in this paper. See, however, Todd, O. J., ‘TPITATΩNICTHC: a Reconsideration’, C.Q. xxxii (1938), pp. 30–8Google Scholar, where Rees's conclusions are criticized.

^{page 142 note 2} Kaffenberger, H., Das Dreischauspielergesetz in der griechischen Tragödie, Diss. Darmstadt, 1911Google Scholar; a criticism of this work by Fensterbusch, C. is to be found in Bursians Jahresbericht der Altertumswissenschaft, vol. ccliii (1936), pp. 38–41Google Scholar.

^{page 142 note 3} Flickinger, R. C., The Greek Theater and its Drama, Chicago, 4th ed. 1936, chap. iiiGoogle Scholar.

^{page 142 note 4} Allen, J. T., Stage Antiquities of the Greeks and Romans and their Influence, 1927, p. 136Google Scholar: ‘The occasional awkward silence of a fourth person who though addressed does not reply, as Pylades in the Orestes of Euripides (vss. 1591 ff.), who says not a word in spite of the ardent appeal of Menelaus, is difficult to explain on any other supposition than that for some ardent appeal of Menelaus, is difficult to explain on any other supposition than that for some reason the poet had only three actors at his disposal’: cf. Haigh, , Attic Theatre (3rd ed. 1907), p. 236Google Scholar.

^{page 142 note 5} Schlesinger, A. C., ‘Silence in Tragedy and the Three-Actor Rule’, Proceedings of Am. Phil. Assoc. lx (1929), p. xxviGoogle Scholar, Classical Philology, xxv (1930), pp. 230–5Google Scholar; ‘The Ins and Outs of the Three-Actor Rule’, Classical Philology, xxviii (1933), pp. 176–81Google Scholar.

^{page 142 note 6} This aesthetic law had already been assumed by J. Richter, op. cit., p. 3.

^{page 143 note 1} A strange estimate of Sophocles' dramatic skill.

^{page 143 note 2} p. 187: cf. Kaff. p. 22. To this point Schlesinger, p. 231, had no reply.

^{page 143 note 3} Attic Theatre ³, 1907, p. 224 (cf. p. 236). Among other scholars advocating the assumption of only three actors in the O.C. are: Schmid-Stählin, , Christs Geschichte der griechischen Litleratur, 1908, i, p. 301, n. 6Google Scholar; Inama, V., Il Teatro Antico greco e romano, Milan, 1910, p. 125Google Scholar; Gugliel-mino, p. 87, n. 1 and p. 101; Webster, T. B. L., An Introduction to Sophocles, 1936, p. 121Google Scholar.

^{page 143 note 4} Disputatio de distributione personarum inter histriones in tragoediis Graecis, Marburg, 1840, pp. 42–4Google Scholar.

^{page 143 note 5} Croiset, op. cit. iii³, p. 258, n. 1; Navarre, O., Dionysos, 1895, p. 216Google Scholar; Flickinger, pp. 180–1. The cast given by Schneidewin, , Oedipus auf Kolonos, 2nd ed. 1854Google Scholar, Einleitung, p. 36, is the same, except that he attributes the Messenger to the third actor: but this clashes with Ismene's part from 1670 to the end.

^{page 144 note 1} It may be remarked in passing that it seems strange that Hermann here believes the three actors capable of playing the same role when elsewhere (pp. 25–31) he lays particular stress on differences in the various actors' abilities.

^{page 144 note 2} pp. 32–3, ‘ut etiamsi spectatores, id quod vix evitari poterat, prioris actorem in posteriori recognoscerent, ipsa recordatione voluptas … non imminueretur …’ etc.

^{page 144 note 3} Schneidewin, Einl., p. 36, ‘Theseus … den Töchtern gegenüber jetzt gleichsam Vaterstelle einnimmt’Google Scholar.

^{page 144 note 4} Neue Jahrbücher für Philologie und Paedagogik (Jahns Jahrbücher), xxxi. i (1841), pp. 456–7Google Scholar.

^{page 144 note 5} Wecklein, N, Oed. Col., 1880, Einleitung, , p. 8Google Scholar.

^{page 144 note 6} Flickinger hinted (p. 181) at a modification of Wecklein's scheme; Ismene's part 1670–end could be acted by the mute who plays her role from 1099 to 1555, and her lyrics 1724–36 sung, off-stage by the third actor who was about to enter as Theseus at 1751: this is the suggestion of Kaffenberger (q.v.). The splitting of Ismene's part is thus avoided, but Theseus remains divided between two actors. Other possible distributions are not here mentioned, as all involve considerably greater difficulties.

^{page 144 note 7} Teuffel, W., ‘Die Rollenvertheilung im sophokleischen Oedipus auf Kolonos’, Rheinisches Museum für Philohgie, N.F. ix (1854), pp. 136–8Google Scholar.

^{page 144 note 8} Teuffel assigns the Messenger to the third actor, but the part might be equally well played by the first.

^{page 144 note 9} The words ὡς ⋯ν Ἀϒαμ⋯μνονι Αἰσχύλου, condemned as spurious by Dindorf, may refer to Cho. 900–2, since ‘plays in a trilogy were sometimes quoted under the name of the first of the series, and ancient authorities are very casual in their references’ (Sikes, and Willson, , Prometheus Vinctus, 1898, p. 143Google Scholar). Pollux apparently considered that these lines were spoken by a supernumerary. The schol. ad loc. (quoted above), however, believed the lines to have been delivered by the third actor, who had changed into Pylades' mask from that of the Messenger (or οỉκεής). Cf. p. 147, n. 1 below, and Haigh, , Attic Theatre ³, 1907, p. 234, n. 2Google Scholar.

^{page 145 note 1} ‘The Meaning of Parachoregema’, Classical Philology, ii (1907), pp. 387–400Google Scholar.

^{page 145 note 2} Navarre, O., Le Théâ;tre grec, Paris, 1925, p. 184Google Scholar; cf. Wilamowitz, , Die griechische Literatur des Altertums (in Die gr. und lat. Lit. und Sprache, 1912), p. 84Google Scholar, ‘und warum sollte nicht, selbst wenn kein Extrahonorar bewilligt war, ein Sprecher für ein paar Worte aufzutreiben gewesen sein?’

^{page 145 note 3} Richter, G., De mutis personis quae in tragoedia atque comoedia Attica in scaenam producuntur, Halis Saxonum, 1934, p. 77Google Scholar.

^{page 145 note 4} Both L. Campbell and Jebb appear to have accepted it.

^{page 145 note 5} The words of Oedipus addressed to Theseus when the latter enters the scene (891 ὦ ϕ⋯γτατ', ἔγνων γ⋯ρ τ⋯ προσϕὠνημ⋯ σου) seem to render this argument incontrovertible. If Theseus was, in fact, here represented by a different actor from the one who had previously taken the part, how could Sophocles have ventured gratuitously to draw the audience's close attention to the new actor's voice by putting into Oedipus' mouth a statement which might so easily be refuted?

^{page 146 note 1} The Theatre of the Greeks, 8th ed., p. 307, n. 6.

^{page 146 note 2} Die dramatische Technik des S., p. 333, n. 1.

^{page 146 note 3} Patin, H., Études sur les Tragiques grecs, Sophocle, 6th ed., 1881, p. 227Google Scholar.

^{page 147 note 1} For changing mask and costume the actor would be able to make use of the short gap between the end of the ode and the beginning of the dialogue as well as the eight lines of conversation 720–7. Compare the very similar instance in Aesch, . Cho. 891–9Google Scholar, on which Flickinger (p. 179) comments, ‘This would be a “lightning” change indeed. … Yet the ancient scholiast accepts it and I do not believe we are warranted in pronouncing it impossible’.

^{page 147 note 2} Jebb and most edd. assign 1688–92 and 1715–19 to Ismene, against the MSS., which continue them to Antigone. L. Campbell (1st ed. 1871, only) followed the MSS. (see his notes on 1687 and 1716). It would make little difference, however, if Ismene sang 1688–92 and 1715–19 as well.