¹ See my ‘P.Oxy. 2078. Vat.Gr. 2228, and Vergil's Charon’. CQ 50 (2000) 192–6CrossRefGoogle Scholar. Aeacus says:

(Vat. gr. 2228 f. 482^r)

What is this? I see a figure hastening hither apace – bold is his spirit indeed! Stranger, you must tell me who you are that come near these regions, and what matter brings you.

For the text with translation and Heracles' reply, see Page, D. L., Select Papyri III (Cambridge, MA 1942) 122f.Google Scholar, vss. 16–19 and 20–31 between frs. 1 and 2 of P.Oxy. 2078. Both speeches follow immediately after a hypothesis of the play all preserved by the otherwise unknown writer Johannes Diaconus and Logothetes as F 1 in Bruno Snell's second ed. of the frs. collected in his Tragicorum Graecorum Fragmenta (Göttingen 1986) 1.171–8Google Scholar under 43 Critias as F 1–14, with R. Kannicht's important addendum of P.Oxy. 3531, first edited by H. M. Cockle (London 1983) 29–36. as F 4a on pp. 349–51: two further papyri are doubtfully ascribed to the play as F 658 among the Adespota in Snell and Kannicht II (Göttingen 1981)240–2.

Authorship of Pirithous is ascribed to Euripides by Page, to ‘Euripides (or Critias)’ by Cockle, and to Critias by Snell and Kannicht, as it is by Diggle, J., who makes the main fragments readily available in his Tragicorum Graecorum Fragmenta Selecta (Oxford 1998) 172–6Google Scholar under ΚΡΙΤΙΑΣΠΕΙΡΙΘΟΥΣ. The question of authorship will arise later, but the task of establishing where Aeacus is located in the Underworld, on which the first section of this paper focuses, ultimately does not hinge on it.

² As indicated in n. 20 below.

³ The uniqueness of the descent-version in Pirithous is documented more fully in my article (n. 1) 193f., to which I add here, exempli gratia, comparison with a scene painted on the Nekyia Krater now in New York by a contemporary of Euripides and Critias. This shows Hades and Persephone inside their house both ruling from their thrones over their subjects, which include Theseus and Pirithous held fast on chairs under the terrible gaze of Persephone. Several other versions portray the pair growing to a rock. By contrast, in the hypothesis of Pirithous preserved by Johannes Diaconus (n. 1). Pirithous is said to sit, alone, upon a rock. For the Krater scene, see Jacobsthal, P., ‘The Nekyia Krater in New York’, Metropolitan Museum Studies 5 (1934–1936) 117–18CrossRefGoogle Scholar; P. Friedländer, ‘Zur New Yorker Nekyia’. AA (1935) 20–1. Heracles' pro-Athenian sympathies towards Theseus in Pirithous are noteworthy: cf. n. 20 below. They are in marked contrast to Charon's towards Theseus, Pirithous, and Heracles in Verg., Aen. 6.392ffGoogle Scholar.

⁴ A quite different approach is taken here from that pursued by Feldherr, Andrew, ‘Putting Dido on the Map: Genre and Geography in Vergil's Underworld’, Arethusa 32 (1999) 85–122CrossRefGoogle Scholar. who is concerned to show how Vergil's netherworld beyond the Styx is arranged in territorial and administrative entities reflecting a Roman conceptualization of space and power under the principate and is, in short, a mirror image of Rome's own imperial sway.

⁵ If, as is mentioned in my article (n. 1) 194, Vergil derived ideas for Aeneas' catabasis from either Naevius' nekyomanteion depicting an evocatio by Aeneas of his father on the shore of Avernus (conjectured by Corssen., P. ‘Die Sibylle im sechsten Buch der Aeneis’, Sokrates 1 (1913) 12Google Scholar) or Fabius Pictor's dream-version of Aeneas' descent ap. Cic., De Div. 1.21.43Google Scholar (on which see Perret., J.Virgile: connaissance des lettres (Paris 1965) 115–16Google Scholar), it is unlikely that he found an extensive topography of the Underworld in either source.

⁶ Clark (n. 1) 195–6.

⁷ References to Hades' gates as entrances to this god's realm below earth's thin crust (Hom., Il. 22.61 ff.Google Scholar) are collected by Vermeule., E.Aspects of Death in Early Greek Art and Poetry (Berkeley, Los Angeles, London 1979) 35–6 with 219Google Scholar n. 62. See also at n. 32 below.

⁸ U. von Wilamowitz so designates Aeacus when discussing Heracles' fight with Cerberus, in Euripides Herakles (Berlin 1889) II. 138Google Scholarad 612 thus: ‘Die etappen der entwickelung sind (1)… (2) kampf um den hund mit dem Tode in gestalt des titanen Menoitios oder des ‘ianitor orci’ (sic), der seit Kritias Aiakos heisst.’ Wilamowitz here refers to Critias as author of Pirithous.

⁹ Cockle (n. 1) 30. Snell (n. 1) 172. too, cites Lucian, De Luctu 4, and adds. ‘Wil. ms.’. Diggle (n. 1) 173 gives Aeacus' speech, printed as fr. 1, the heading ‘tabulae exordium?’.

¹⁰ Page (n. 1) 121–2.

¹¹ Acheron is named by Euripides in Alc. 444 as a ‘lake’ upon which Charon sails, but the lake is at first unnamed in 253, as in the Frogs. Dover, K. J., Aristophanes Frogs (Oxford 1993) 254Google Scholar linking Frogs 470–3 to Od. 10.513–4 suggests that Acheron may have been a lake for Homer also.

¹² These colourful renderings of infernal names are David Barrett's from his translation Aristophanes: The Wasps, The Poet and the Women, The Frogs (Harmondsworth 1964. with reprints) 161.

¹³ Dover (n. 11) 227–8 ad 273 discusses at length an apparent reversal of Heracles' sequence of events, as Dionysus and Xanthus seem to see the sinners in the mud before the monsters. The sinners, in truth, are barely touched upon, being quickly disposed of with a comic glance at the audience, perhaps to suppress conflict with the location of the main dead beyond the door of Pluto's house, from which they exit with Aeschylus and Euripides onto the stage at 830ff. Comparable with the scene in the Frogs is Polygnotus' mid-fifth-century wall-painting ap. Paus. 10.28.4, 5, and 7 depicting sinners and monsters on Acheron's bank (the far bank, no doubt, since Odysseus is said in section 1 to be already in Hell) with other figures spreading out from there. Not only does each version include among its sinners one who has maltreated a father, but Polygnotus' flesh-eating demon Eurynomus corresponds to the malignant creature Empusa in Frogs 293.

¹⁴ See at n. 8 above.

¹⁵ Aeacus is both Hades' gate-keeper invoked as Άḯδεω πυλαουϱέ in a 2nd-c. BC epitaph from Smyrna, (GVI I. 1179.7Google Scholar) and Hades' key-holder, who τὰς ϗλεῖς τοῦ ῞Αΐδου φυλάττει at Apollod., Bibl. 3.12.6Google Scholar and is called ϗλῃδοῦχος in a later epitaph from a tomb at Rome, (IG XIV. 1746.4Google Scholar). The Table of Colotes at Olympus described by Paus. 5.20.3 depicts Hades holding his own key in the role of ϗλῃδοῦχος usually assigned to Aeacus.

¹⁶ Evag. 15.

¹⁷ Apol. 41a, Gorg. 523e ff., esp. 524a and 526c.

¹⁸ Only in Plato is Aeacus' court placed at the Dividing of the Ways and in this meadow. Dodds, E. R., Plato Gorgias (Oxford 1959) 375Google Scholar suggests that Plato's meadow and all other infernal meadows originate from Homer, 's ‘asphodel meadow’ (Od. 11.538–9Google Scholar). If Dodds' suggestion is correct, the meadow in which Aeacus holds court can be imagined as lying beyond infernal waters (cf. Od. 11.157–9), though Plato himself does not say so. Other infernal topographies support this conjecture: beyond infernal waters we find not only Aristophanes' Meadow of Initiates in Frogs 326. but also Vergil's Fields of Mourning (Lugentes campi). Division of the Ways (partes… se via findit in ambas) and Fields of Joy (locos laetos… compos) in Aen. 6.441. 540, and 638–40 respectively. Another link between Aeacus and Plato's meadow is Achilles, a descendant of Aeacus (Hom. loc. cit.). who strode through Homer's asphodel meadow after death (Od. 11.539).

¹⁹ Boardman, J. et al. , (eds.). Lexicon Iconographicum Mythologiae Classicae (Zürich 1981–Google Scholar), s.v. Aiakos with pl. 1. 3. It would seem that this stick is not the staff, ῥάβδος, put in his hand by Plato, Gorg. 526c, as his judge's badge of office, but see below at n. 29.

²⁰ To my mind Aristophanes' exploitation of Pirithous does not refute E. Norden's hypothesis supported by H. Lloyd-Jones that an early epic account of the descent of Heracles lay behind the description of Dionysus' descent in the Frogs. See, respectively, P. Vergilius Maro: Aeneis Buck VI ³ (Leipzig and Berlin 1926) 5Google Scholar with n. 2. and ‘Heracles at Eleusis: P.Oxy. 2622 and PSI 1391’. Maia 19 (1967) 206–29Google Scholar = Greek Epic, Lyric and Tragedy (Oxford 1990) 167–87Google Scholar. According to Norden. Apollodorus 2.5.12 summarized an epic version of Heracles' descent known to Vergil, most probably, he thought, through a mythological handbook, after the epic had directly influenced Bacchylides and Aristophanes. Lloyd-Jones adds to these authors Euripides (e.g., at Her. 610–13) and Pindar (P. Oxy. 2622) and infers from Heracles' pro-Athenian sympathies and connection with Eleusis that the epic was composed c.550 BC by an Athenian or a person belonging to the orbit of Athenian culture. Robertson., N.Hermes 108 (1989) 274–300Google Scholar thinks it may have formed part of the Hesiodic Aegmius frs. 294–301 M-W attributed to Cecrops of Miletus. In my article (n. 1) 195 I express the view that the epic could have survived to Vergil's day as many epics did, but Vergil, as Aristophanes, would have taken every chance to read also whatever separate literary treatments existed in the articulate and well-developed tradition about Heracles' descent, including Pirithous.

²¹ Dover (n. 11) 51–2 provides a conspectus of sigla and scholia. The fact that R, the mid-10th-c and oldest MS of the Frogs, names Aeacus at 464–5. 630ff. and 738. but not thereafter in the last scene, nor at all in the list of dramatis personae that refers only to θεϱάπων Πλούτωνος. does indeed betray confusion over the sigla naming Aeacus. The question to be faced, though, is not whether such sigla go back to Aristophanes' text, but whether those who inserted sigla were correct to identify Aeacus as the speaker of certain verses in the Frogs. Indeed the sigla probably do not derive from Aristophanes if Lowe, J. C. B., ‘The Manuscript Evidence for Changes of Speaker in Aristophanes’, BICS 9 (1962) 27–42Google Scholar is right to infer that all changes of speaker in Greek dramatic texts were marked originally by diacritical marks only, or possibly by mere spaces with or without an accompanying paragraphos.

²² Here I give no more than one or two instances cited by Dover for each term. For a fuller list, see Dover's valuable collection of references (n. 11) 52–3. To the divine door-keepers there listed I add (n. 39) Hecate in her role in Aen. 6.

²³ The significance of this appellation for Aeacus' slave-status was recognized by Fritzsche, F. V., Aristophanis Ramie (Zürich 1845) 233Google Scholar. reaffirmed by Dover (n. 11) 50. but denied (on the ground of the tragic tone of 464ff.) by Radermacher., L. Aristophanes' ‘Frösche’: Einleitung, Text und Kommentar ³ (reissued Graz-Vienna-Cologne 1967) 211Google Scholar.

²⁴ Dover (n. 11) 55.

²⁵ Aeacus' indulgence to a fellow-slave probably also has political overtones. Slaves who took part in the Athenian victory at Arginusae in 406 (Xen., HG 1.6.24Google Scholar) were given freedom and the rights of citizenship (Hellanicus, FGrHist 323a F25). the point underlying Frogs 33–4, 190–2. and 693–9.

²⁶ This power was observed by Ste. Croix., G. E. M. deThe Class Struggle in the Ancient Greek World (London 1981) 505–6Google Scholar.

²⁷ Sommerstein, A. H., Frogs (Warminster 1996) 10Google Scholar with references to Critias' political activity. The same argument applies to three other plays of disputed authorship, namely Tennes. Rhadamanthys, and Sisyphus. von Wilamowitz., U.Kleine Schriften (Berlin 1931–1937) I. 449–50Google Scholar, likewise put the performance of Pirithous in a period before Critias went into exile from Athens, i.e. before the summer of 406 BC (Xen., HG 2.3.15 and 36Google Scholar).

²⁸ Dover (n. 11) 55.

²⁹ See Dover loc. cit, Sommerstein (n. 27) 198f. ad 464, and for the vase-paintings n. 19 above.

³⁰ Cerberus, , the Tartareum … custodem of Aen. 6.395Google Scholar. is called ianitor at Aen. 6.400 and ianitor Orci at Aen. 8.296.

³¹ See n. 3 above.

³² Sourvinou-Inwood., C. ‘Reading’ Greek Death to the End of the Classical Period (Oxford 1995) 306–9Google Scholar discounts Hermes Chthonios in Hom. Od. 24 as being the work of the ‘Continuator’, and relying on Eur. Alc. 357–62. where Admetus assures Alcestis that he would go to Hades if he had Orpheus' charms so that neither Cerberus nor Charon would stop him from restoring her to life, argues that Charon controls the movement not only into but also (like Cerberus) out of Hades.

³³ For reconstruction of the fragmentary Pindaric similes and their location, see respectively Lloyd-Jones (n. 20) 215f. = 175f. and my ‘Two Virgilian Similes and the Ήϱαϗλέονς ϗατάβασις’. Phoenix 24 (1970) 244–55CrossRefGoogle Scholar. Lavecchiu., S. ‘P.Oxy. 2622 e il “Secondo Ditirambo” di Pindaro’, ZPE 110 (1996) 1–26Google Scholar, leaves their location unmentioned in his re-examination of the fragment. Lobel, E., the first editor of P.Oxy 2622 (London 1967) 63–5Google Scholar, inferred the inclusion of Meleager in the ghost-scene from the mention of Με]λέαγϱον a few lines below the similes.

³⁴ In response to my proposed ancestry of the Vergilian similes in my article (n. 33) 250–3, Thomas, R. F., Reading Virgil and his Texts: Studies in Intertextuality (Ann Arbor 1999) 267–87, esp. 278–81Google Scholar, thinks that Homer's leaf and bird similes at Il. 6.146–9 and 3.2–7, and Apollonius' wave, leaf and bird similes at Argon. 4.214–19 and 238–40 were sufficient models for Vergil's inventiveness without the help of Bacchylides or Pindar or Norden's hypothetical epic catabasis mentioned in n. 20 above. However, Vergil did not learn their location beside Cocytus from Homer, whose similes are comparisons to men-in-this-life and not ghosts-in-the-next, or from Apollonius, who applies his similes to the Colchians pouring along the river bank or over the sea in the faintly catabatic framework of Jason's voyage. Consideration of Vergil's sources for the similes, it seems to me, should not be isolated from the other main motif to be discussed below in the same ghost-scene. Since motifs now appearing as separate episodes in Aeneas' infernal journey once appeared together in the same scene, it is highly unlikely that Vergil re-invented one motif, namely the similes comparing ghosts seen by the descending hero beside Cocytus, from sources unconnected with Heracles' descent but borrowed the other from the same scene in Heracles' infernal journey. For reasons too lengthy to detail here, I suspect with Norden and Lloyd-Jones that Vergil knew the common account.

³⁵ In determining the relative location of these Homeric ghosts, observe the following references: Od. 10. 508–12 refer to Odysseus about to cross Ocean and enter Hades' house; 513–14 to Cocytus (called a branch of Styx) and Pyriphlegethon as rivers flowing into Acheron, itself perhaps a lake formed at their confluence marked by a rock (see n. 11 above); 527–9 to sacrifices beside the trench as Odysseus looks backwards at a streaming river; and 530ff. to ghosts rising from the trench. After Odysseus makes the Ocean-crossing in Od. 11.21–2, his mother points out in 155–9 that he has also passed over the infernal rivers (cf. 10.513–14) before seeing her and the other ghosts that have been rising from the trench continuously since 36.

³⁶ Mynors, R. A. B., Virgil: Georgics (Oxford 1990) 316–7Google Scholar on Georg. 4.471–4.

³⁷ Marshall., C. W. ‘The So-called False Digamma in Bacchylides Ode 5.75’, Mnemosyne 47 (1994) 373–5CrossRefGoogle Scholar rightly, it seems to me, understands ἰόν with a digamma here as meaning ‘poison’, put as metonymy for an arrow Heracles has dipped in the poison of the Hydra slain in an earlier labour. On Heracles' bow as a dubious emblem, see F. M. Dunn, ‘Ends and Means in Euripides’, in Roberts, D. H. et al. (eds.). Classical Closure (Princeton 1997) 96–8Google Scholar.

³⁸ The correspondence between both sets of females was first observed, later accepted by Dover (n. 11) 263, in my article (n. 33) 252 n. 22, where at the time it escaped my notice that they also both occupy the same residence. This correspondence extends the comparisons with the Gorgon-motif made by Brown, C. G. in ‘Empousa, Dionysus and the Mysteries: Aristophanes, Frogs 285 ff.’, CQ 41 (1991) 41–50CrossRefGoogle Scholar.

³⁹ Since Hecate holds no keys she is not quite ϗλῃδοῦχος for which title see at nn. 15 and 22 above. Accordingly in my ‘Vergil, Aeneid 6: The Bough by Hades' Gate’ in Wilhelm, R. M. and Jones, H. (eds.), The Two Worlds of the Poet: New Perspectives on Vergil (Detroit 1992) 167Google Scholar I have referred to Hecate in this passage, after Eitrem, as ῥηζιπύλη (PGM. 4.2751) and προθυϱαία (Procl., Hymn 6.2.14Google Scholar. cf. PGM. 4.2719. ).

⁴⁰ For more on this interpretation, see my article (n. 39) 167–78.

⁴¹ With this vertical fissure cf. also Georg. 4.467 and Aen. 7.570. On the various Vergilian passages as synonyms and on the layout of the house, see, respectively my article (n. 39) 174 and Wistrand, E., ‘Virgil's Palaces in the Aeneid’, Klio 38 (1960) 146–54CrossRefGoogle Scholar.

⁴² For instances see those mentioned in or at nn. 13 and 38 above.

⁴³ E.g. the Harpies, coupled by Vergil with Gorgons, Gorgones Harpyiaeque, at Aen. 6.289. The comment by Austin., R. G.P. Vergili Maronis Aeneidos Liber Sextus (Oxford 1977) 122Google Scholar that they are death-spirits in Hom., Od. 20.77–8Google Scholar explains their nature but not their location in Pluto's infernal palace. Stubbs, H. W., ‘Vergil's Harpies: A Study in Aeneid III’, Vergilius 44 (1998) 3–12Google Scholar more helpfully notes that they have been relocated by Vergil's inventiveness from one underground cavernous location to another. No longer inhabitants of the cave beneath Mount Dicte in Crete, whither according to Apollonius they had fled (Argon. 2.424) from the Strophades Islands (cf. Aen. 3.209–11) after being chased to these islands (Argon. 2.223–34), they were transported by Vergil from beneath that cave to beneath the cave leading to the netherworld at lacus Avernus in Italy.

⁴⁴ I am grateful to both of the journal's referees for their helpful observations, in particular to Colin Austin for his detailed comments. For support of my research I am grateful to the Social Sciences and Humanities Research Council of Canada.