^{page 28 note 5} The upper date depends on the abnormal proportion of satyr-plays in the corpus of Pratinas known to Suidas, and the acknowledged mastery of Aeschylus in this genre (Diog. Laert. II, 133; Paus, III, 13, 5) which may well have ensured his confirming, if not inaugurating, the convention. The lower date must precede 341 (I.G. II², 2320), and if the change to having a single satyr-play before the entire festival went hand in hand with the introduction of a single old tragedy (both represent a ‘flavouring’ of the traditional) it might be 386 (I.G. II², 2318 viii).

^{page 29 note 1} The Inachus is to be argued to, not from: it may not be satyric. See below.

^{page 29 note 2} Hermes, LXXXIV (1956), 1 ffGoogle Scholar. (He regards Silenus as ‘Chorführer’, 4 and 11.)

^{page 29 note 3} That Medea needs only two actors altogether is irrelevant. Even if we reject Rivier's notion of Medea's discomfiting or bending to her will a series of personalities (finally her own, too) by an unseen inner power (Essai sur le tragique d'Euripide, ch. 11), it is blear that the play is built on an economical structure with Medea (practically) omnipresent as pivot and one other major character ‘challenging’ her in turn, with a chiastic arrangement of scenes (so now, Buttrey, T. V., A.J.P. (1958), 1 ff., esp. 6Google Scholar). A. M. Dale speculates (edn. of Alcestis, p. xix) that ‘a limitation normal for satyr-plays at this period could also extend to the occasional tragedy’; but, apart from the question of satyr-play participants (which needs, as here, separate argument), it is hard to understand the extension of a limitation (except for economic reasons and permanently). But this editor rightly rebukes Schmid–Stählin for regarding Silenus in Cyclops as κορυφαῖος (with whom he in fact converses, vv. 82 ff.).

^{page 30 note 1} Pickard-Cambridge, , The Dramatic Festivals of Athens (hereafter D.F.A.), p. 137 Google Scholar (speaking of actors' parts). Cf. his further remark ‘nor is it likely that in a contest to which great importance was attached the State would have provided three actors for one competitor and four for another’. The connexion is different, but the principle is relevant here.

^{page 30 note 2} Kühnert ( Roscher, , Lexicon, IV, col. 456Google Scholar) thinks that Papposilenus as such may have been a product of the theatre, and Attic vases do not contradict this. Although the ‘old philosopher’ aspect of Silenus (cf. Plato, , Sympos. 215 B Google Scholar) was known to Hellenistic art, so was that of the ‘old reprobate’ outdoing the satyrs in drink and despicable behaviour (and, in Cyclops, unfortunate fate), and the satyr-plays stressed this (cf. Pollux' note, IV, 142, that Silenus, as compared with the satyrs, is τὴν ἰδέαν θηριωδέστερος). It may not be fanciful to see here one formative element of the έξάρχων theory of tragedy's development.

^{page 30 note 3} For twelve, see now Fraenkel, E. ad A. Ag. 1344 Google Scholar; for fifty, Fitton-Brown, A. D., C.R. (1957), pp. 1ffGoogle Scholar. See below.

^{page 30 note 4} Naples 3240; Furtw.–Reich. III, p11. 143–5; Beazley, , A.R.V. p. 894 Google Scholar (Pronomos painter, no. 1). Bulle, , Eine Skenographie, pp. 27–9Google Scholar, shows it to be part of a votive offering to Dionysus by Pronomos himself. (See, for convenience, Pickard-Cambridge, D.F.A. fig. 28.)

^{page 30 note 5} So Pickard-Cambridge, , D.F.A. p. 180 Google Scholar, n. 1—yet he himself accepts the painting as evidence for chorus-number (p. 242). Kühnert went so far as to make twelve by suggesting adding to the satyrs not Silenus, but either Demetrius (Arnold) or Charinus (Prott). (I do not know why A. M. Dale, op. cit. p. XIX, n. 2, says there are two actors besides Silenus: Hesione is a woman and has bare feet, and therefore is no actor; but these facts cast doubt on the actor-status of Laomedon and Heracles—and note Laomedon's (own) beard.)

^{page 31 note 1} Demetrius is a puzzle, too. This cannot be an Old Comedy, as the story will not do, but we know Demetrius as a poet of such (Diog. Laert. V, 85). The vase is too early, 420 being the upper limit (D.F.A. p. 179, n. 4; Brommer, , Satyrspiele, 11 Google Scholar; Beazley, loc. cit.), for us to take into account the (Cilician?) school of tragedians (Ταρσικοί) which perhaps included another Demetrius (Diog. Laert. IV, 58 and V, 85).

^{page 31 note 2} D.F.A. p. 183, n. 1 and fig. 40; A.R.V. p. 965.

^{page 31 note 3} Beazley, , Hesperia, XXIV (1955), 316 fGoogle Scholar. and pl. 88 a (London, E 467). (My colleague R. T. Williams kindly called my attention to this reference.) Perhaps the Pronomos vase is also anachronistic in another sense: at any rate all Aeschylean satyr-scenes on vases are fifty years late (so Brommer, loc. cit. p. 49).

^{page 31 note 4} Forthe heavy cost to the choregos, see D.F.A. p. 89. For the existence of professional choreutae, see D.F.A. p. 91 which refers to the fourth century but argues a tradition; note, too, that the tribal restriction on chorus members did not apply to drama: D.F.A. p. 76, n. 3).

^{page 31 note 5} Then Alcestis, being heavily satyric in theme (Stith-Thompson, R 185, connects Alcestis- and Sisyphus-stories of overcoming Death), therefore conforms in this respect too, needing two actors with no Silenus (and not having three ‘free’ actors, as the enforced silence of Alcestis at the end shows: the emotive effect is probably secondary, despite A Winter's Tale). Murray's perception that the chorus sometimes splits into five speakers in this play (ad Arg. II, 6) seems likely for 872 ff., 888 ff. at least, but it is hard to know what conclusion to draw: is a multiple of five necessary as the total number? (The tendency to hemichoric arrangement is no argument against a fifteen-chorus: cf. Ajax 866–78 and Jebb ad loc.)

^{page 32 note 1} See D.F.A. p. 242, n. 1. For the matter of comic actors, see Tzetzes, , Prol. de Com. 16 Google Scholar; this Pickard-Cambridge finds ‘not difficult to accept’ (D.F.A. p. 148). Nothing, I fancy, can be deduced from all this as to the successive totals of the tragic chorus itself. It could have passed from twelve to fifteen as that of the satyr-play, from fifty to fifteen while that of the satyr-play rose from twelve to fifteen, or even from fifty to twelve while the satyr-play had twelve (then both to fifteen). That both should pass from fifty to twelve to fifteen is least likely: a wild scene of fifty boisterous satyrs would be unmanageable, and there was no mythological compulsion to give the satyr-chorus this number (as with the Danaids, Nereids, etc.).

^{page 32 note 2} Admittedly ξένον νοῶ τίς ᾗν (i, 23)—but why no names? As Lobel says, either there is a negative hereabouts, or τίς = ποῖος.

^{page 33 note 1} That Zeus should appear in the prologue or an early episode, as supposed by Lobel, seems to be excluded by fr. 275 P. (=schol. Aristoph, . Plut. 807 Google Scholar) τοῦ Διὸς εἰσελθόντος πάντα μεστὰ ἀγαθῶν ἐγένετο (ἐγίνοντο cod.; the phraseology varies)—which equates his role with an ex machina disentanglement or the like. Hermes, however, may have been on the scene before our fragment begins.

^{page 33 note 2} Aeschylus uses κάρβανος at Suppl. 118 = 130 and 914 in connexion with Egyptians (and Io's descendants); Egyptians are black and κάρβανος would suit them (see Lobel) ; finally Epaphus was black (Prom. 851) because he was born in Egypt.

^{page 34 note 1} ii, 6–9 are all uttered by the chorus, as Pfeiffer saw (so correct Hunt–Smyly). πόνους ἐλᾶν is an unusual collocation.

^{page 34 note 2} It may be that σύριγγο[ in i, 7 is his music here (so Ovid, , Met. 1, 677 Google Scholar). Argus can hardly be playing the pipe and singing (as we know he was from fr. 281 P.). The weapon cannot be referred to the second attack, as the chorus would not see it then.

^{page 34 note 3} Pfeiffer and Page were not wrong in their reluctance to believe that the metamorphosis occurred during the action ( Siz. Bay. Akad. (1938) 2, 55 Google Scholar; Select Papyri, III, 23 Google Scholar), arguing on the basis of the Tebtunis fragments alone. The evidence of the Oxyrhynchus papyrus makes room for it (and for that which must precede it) by showing the omission of the wanderings. It is hard to think that the two papyri refer to different plays: for P. Tebt. 692 we know of no other Inachus–Argus play, and Chaeremon's Io is excluded on stylistic grounds. Iophon's Αὐλῳδοὶ σάτυροι is attractive (counter-music to Hermes?) but its only fragment (1N.) speaks of a ‘procession of sophists’.

^{page 34 note 4} Pearson, frs. vol. 1, 199. Brommer, op. cit. figs. 13 and 14.

^{page 35 note 1} As for the vase-paintings, it should be remembered that the connexion thereon of satyrs with Perseus cannot be securely related to any known play (cf. Brommer, op. cit. p. 74). The Inachus cannot be ‘prosatyric’ as Alcestis: there is too much of the tragic in the normal form of the story for it to be satyric without an actual infusion of Silenus and satyrs.