¹ Earlier forms of this paper were delivered to audiences at Queen's University, Belfast (1977); the Edinburgh Branch of the Classical Association (1980); Stanford University (1981); and the Oxford Philological Society (1981): I am grateful to many scholars who contributed to the discussion on these occasions, and to Professor R. G. M. Nisbet, Dr R. S. Padel, Ms P. K. M. Kinchin and Mr Robert Wells for helpful criticisms of a written draft.

² Unless the names of these men or of their prospective hosts were meaningful to an ancient reader, his first hint that the setting was in Cos would be the toponym Ἅλεντα at the end of the first line (cf. n. 45 below), then the references to Clytia and Chalcon in 5 and the description of Burina in 6–9. Further confirmation comes from the tomb of Brasilas (10–11), perhaps from the reference to Philetas without an ethnic (40), and from Lycidas' departure for Pyxa (130–1).

³ Cf. Arist. Rhet. 3.1418b23 ff.

⁴ Note also Idyll 20, not by Theocritus but by another author of uncertain date (cf. Gow, A. S. F., Theocritus 2nd edn, Cambridge, 1952, ii. 364 ffGoogle Scholar.)

⁵ Reference of the first person to the poet in early iambus has of course been questioned, e.g. by West, M. L., Studies in Greek Elegy and Iambus (Berlin, 1974), 28 ffCrossRefGoogle Scholar.

⁶ See Gow (n. 4) ii. 133–4 and scholia ad loc.

⁷ Ibid. 141 (on 40).

⁸ Ibid. 155 (on 93).

⁹ Ibid. 118–20.

¹⁰ βουκολιασδώμεαθα (36) can perhaps be explained away as a ‘bucolic metaphor’ where any poet can be seen as a βουκόλος or αἰπόλος (cf. van Groningen, B., ‘Quelques problèmes de la poésie bucolique grecque’, Mnem. 11 (1958), 293–317, at 310 ffCrossRefGoogle Scholar.). ⋯ν' ⋯ρεα βουκολέοντα (92) is less susceptible of such treatment.

¹¹ See Gow (n. 4) ii. 127–9. I am more reluctant than Gow to credit the scholiast's Simichidas of Orchomenos. I also doubt that the name Simichidas was as familiar to Theocritus' contemporaries as the name Sicelidas, and do not see how this is demonstrated by the way it is introduced at line 21 (ibid. 129).

¹² Dover, K. J., Theocritus. Select Poems (Basingstoke, London, 1971), 148–50Google Scholar.

¹³ Williams, F., ‘A Theophany in Theocritus’, CQ 21 (1971), 137–45CrossRefGoogle Scholar.

¹⁴ In particular, no Cydonia has yet been identified on Cos in a literary or epigraphic text.

¹⁵ 150, citing Kühn, J.-H., ‘Die Thalysien Theokrits’, Hermes 86 (1958), 40–79Google Scholar.

¹⁶ The Muses' gift to Hesiod at Theog. 29 ff. is certainly recalled by Lycidas' gift to Simichidas and emphasised by the thematic echo at line 44 of Theog. 27–8. See below and n. 51.

¹⁷ Hence there are attractions in the suggestion of Brown, E. L., ‘The Lycidas of Theocritus' Idyll 7’, HStCP 86 (1981), 59–100Google Scholar, that Lycidas is Pan. But this identification, like that of Williams, fails to meet criteria (1) and (3).

¹⁸ Giangrande, G., ‘Theocrite, Simichidas et les Thalysies’, Ant. Class. 37 (1968), 491–533CrossRefGoogle Scholar.

¹⁹ Williams (n. 13) 138.

²⁰ E.g. Herodotus 9.5; Demosthenes 20.131; Ditt. Syll. ³ 84.5; IG vii 1178 (Tanagra).

²¹ Williams (n. 13) 139 with nn. 5–7. For some problems in linking Cretan Cydonia with Apollo cf. Brown (n. 17) 71 (the father of the Cydon who founded it is in fact stated by the scholion on Ap. Rhod. 4.1492 to have been Hermes, not Apollo, as in the possibly confused entry of Stephanus of Byzantium, our only witness alleging that Cydonia in Crete was once called Apollonia). Brown's own hypothesis, that the ethnic Κυδωνικόν leads us, inter alia, to the god Pan, is open to the objections to a divine identification for Lycidas offered in the text. Nor am I persuaded by his suggestions that Κυδωνικόν might suggest κύων and hence ɸύλαξ (as Pan is called by Pindar, fr. 95.2); nor that it might suggest a quince and hence a carved quincewood figure of Pan (84–5, but apparently abandoned by 87): on my reading of Brown's proposals we are to understand Lycidas throughout as a carved wayside figure, so it is mysterious how he can walk off at 130 ff. (Proponents of Cretan Cydonia seem incidentally to have missed the possible support of a mountain there called Tityros, Strabo 10.12 (479C).)

²² Brown (n. 17) 86–7 takes ἄνδρα as bringing out ‘that the god is a good “man” with the Nymphs (Muses), a ladies' man’. But the ‘Theocritean’ usage of 8 (not usually ascribed to Theocritus) which he offers as a parallel ὦ τράγε, τ⋯ν λευκ⋯ν αἰγ⋯ν ἄνερ (8.49) is different because the genitive αἰγ⋯ν dependent on ἄνερ points the reader to this meaning: I do not see how the Greek ⋯σθλ⋯ν σὺν Μοίσαισι Κυδωνικ⋯ν εὕρομες ἄνδρα could be taken as ‘a good man with the Muses’ in a sexual sense. Brown (99) also notes that 7.86 αἴθ' ⋯π' ⋯με⋯ ζωοῖς ⋯ναρίθμιος ⋯ɸελες ἦμεν might be taken as evidence that the singer, Lycidas, is not immortal (since a god would have been a contemporary of Comatas as well as of Simichidas). This point seems to me to have some weight, and I am not convinced by his suggestion that 83–9 are all intended to be part of Tityrus' song: that seems to be excluded by the way in which the line closing Lycidas' (and on Brown's view Tityrus' inset) song is immediately followed by (90) χὢ μ⋯ν τόσσ' εἰπὼν ⋯πεπαύσατο. This clearly refers to Lycidas.

²³ Williams (n. 13) 140, 142–3. Brown (n. 17) 74–5 notes that only the scholiast so far attests an Ἀπόλλων Πύξιος, and that the stone's epithet Φύξιος is not certainly limited to Apollo. Brown's own proposal that ⋯π⋯ Πύξας should be understood as ‘to the tune or accompaniment of boxwoods’ requires Theocritus to have coined a form πύξη alongside the regular ⋯ πύξος; to have expected his readers to see that ‘boxwoods’ ‘ meant ‘Pan-pipes’ (Latin buxus often means ‘flute’, cf. OLD s.v., but neither it nor πύξος seems ever to mean ‘Pan-pipes’, and πύξος nowhere else means even ‘flute’ or aulos); and to have further confused his readers by the use of τ⋯ν (sc. ⋯δόν?). But τ⋯ν ⋯π⋯ Πύξας εἶρɸ' ⋯δόν can mean only ‘he went on the road to Pyxa’. The matter would be unambiguous (if it is not so already) were Theocritus to have written Φύξας; and perhaps he did – the form Φύξιος appears in the scholion on 130/131 c (p. 109 Wendel). It might, of course, stem from ancient scholarship and not from the text of 7, but there is some chance that Theocritus wrote Φύξας. On the location of Pyxa cf. n. 45.

²⁴ Williams (n. 13) 141; Ovid, Met. 2.680–2. Tibullus 2.3.11 ff. also elaborates a picture of a dishevelled Apollo tending Admetus' cattle, but mentions neither goats nor pastoral garb.

²⁵ I owe this point to Prof. R. G. M. Nisbet.

²⁶ Gow (n. 4) ii. 139.

²⁷ The MSS of Longus offer the spelling and accentuation Φιλητ⋯ς. The MSS of Theocritus 7.40, of Stobaeus and of most other Greek writers likewise write Φιλητ⋯ς or Φιλήτας. The spelling Φιλίτας, which appears in MS A of Athenaeus and in some other writers, is preferred by modern scholars, doubtless correctly, for the Coan scholar-poet, because it is that of a Coan text of c. 200 b.c. (Paton, and Hicks, , The Inscriptions of Cos, Oxford, 1891, 10 b 37 and 54)Google Scholar; note also Φίλιτις ibid. 47 15 and Ann. scuol. arch. At. n.s. 25–6 (1963), 169 no. ixa 56 (c. 200 b.c.). It seems quite likely that the etacistic form Φιλητ⋯ς (on the analogy of Philemon?) had become established by the second century a.d., since it seems to be that known to Augustan poets: it is also that of an inscription of the Roman imperial period, Paton and Hicks 310. If so, Longus could have written Φιλητ⋯ς and thought this to be the correct form of the Coan poet's name. It is also possible that he wrote Φιλίτας and that his MS tradition altered that form in the same way as Φιλίτας is assumed to have been corrupted in the paradosis of Theocritus and other authors.

²⁸ The identification was made at least as early as Reitzenstein, , Epigramm und Skolion (Giessen, 1893), 260 n. 1Google Scholar. For other material cf. Cairns, F., Tibullus (Cambridge, 1979), 25 ff. and nn. 110 and 112Google Scholar.

²⁹ Hunter, R. L., A Study of Daphnis and Chloe (Cambridge, 1983), 77CrossRefGoogle Scholar compares Tibullus 1.8.

³⁰ Longus 2.4–6. For his picture of Eros here Longus also draws on Bion, fr. 13; cf. Hunter (n. 29) 77.

³¹ Longus 2.32. I owe the point about Longus' small cast to M. D. Reeve.

³² On the names Amaryllis and Tityrus see further below Section (ii).

³³ Some scholars, e.g. Gow (n. 4) ii. 135 on 7.12, had noticed this Cydonia but dismissed it.

³⁴ It is usually assumed that Lycidas' celebratory party takes place wherever Lycidas imagines himself as being (Cos?) and without the presence of Ageanax. There are singular oddities in this scenario. Lycidas celebrates either the departure of Ageanax (which by the normal conventions of eros we might expect to make him miserable or anxious) or his safe arrival at Mytilene – but how he is to know that Ageanax has completed a voyage of 150 or more miles is left obscure. τ⋯νο κατ' ἆμαρ (63) seems to refer to his safe arrival, hardly the same day as his departure, and difficult to refer also to the day on which Ageanax might be implied to have granted Lycidas long-withheld favours, and hence little support for the view that Lycidas, is rejoicing ‘as much for Ageanax's complaisance as for his safety’ (Gow (n. 4) ii. 145 on 52–P, cf. ibid. 148 on 62. I assume that Gow's 50 miles' (ibid. 148 on 63) for the length of Ageanax's voyage is a printing error for 150, though even that would be simply the distance as the crow flies, and a boat would have to travel much further). Matters would be slightly eased if we are intended to imagine the location of Lycidas' song as his native Cydonia, on the above view near enough to Mytilene for the voyage to be completed in a day. But the other oddities remain disturbing. I am tempted to wonder whether the answer lies in the meaning of μεμναμένος (69): the cognate verb μνάομαι is used both in the sense ‘remember (cf. LSJ s.v. I) and ‘woo’ (ibid. II, citing Od. 1.39; 11.287; 16.77; 19.529). It would not be untypical of Hellenistic Gelehrsamkeit to argue that μεμναμένος should also have this range of meaning and for a learned poet to incorporate an illustrative use in his work. If this speculation is correct, then Ageanax is present and being wooed at the party, and it is a party that takes place after he has arrived and at his destination, i.e. Mytilene.

³⁵ Gow notes names of the form Ageanax/Archeanax from Cos, Smyrna, Miletus and the Troad. For Lesbos cf. Archeanaktidai in Alcaeus (112.24 L–P) and Damoanaktidas (ibid. 296(b); cf. below n. 74); Kleanaktidai in Alcaeus (112.3 L–P) and Sappho (98 L–P) — all these pre-sumably Mytilenean; and Polyanaktidai in Sappho (99 L–P). Later at least two Mytileneans bore the name Lesbonax, the philosopher of the first century b.c. (cf. PIR ² L 160) and a sophist of the second(?) century a.d. cf. Kleine Pauly iii. 584–5 s.n. Lesbonax 2).

³⁶ Cf. Cichorius, C., Römische Studien (Berlin, Leipzig, 1922), 321–3CrossRefGoogle Scholar, noting the Mytileneans Cn. Pompeius Longus (IG xii.2.88, late first century b.c.?) and A. Pompeius Longus Dionysodonis (IG xii.2.249, first century a.d.?). But Longus is not a rare name (21 Longi are noted by Petersen, L. in PIR ² v. 1 (Berlin, 1970), 92)Google Scholar; we still lack evidence of the name Longus persisting in Lesbos down to the latter part of the second century a.d.; and there is no Longus in the group associated with a Dionysiac cult in Rome whose Lesbian connections are indicated by Pompeia Agrippinilla, cf. Vogliano, A. and Cumont, F., AJA 37 (1933), 215–70CrossRefGoogle Scholar; Merkelbach, R., Roman und Mysterium in der Antike (München, 1962), 193Google Scholar. That Dionysodorus is an ⋯ρχιερεύς does little to strengthen links with a novel in which Dionysus has a significant role, since the office was a mark of social eminence rather than religious devotion, and was concerned with the cult of the emperor in particular.

³⁷ ⋯ν Λέσβῳ σηρ⋯ν ⋯ν ἄλσει Νυμɸ⋯ν θέαμα εἶδον κάλλιστον ὧν εἶδον· εἰκόνα γραπτήν, ἱστορίαν ἔρωτος… (pref. 1) … κα⋯ ⋯ναζητησάμενος ⋯ξηγητ⋯ν τ⋯ς εἰκόνος τέτταρας βίβλους ⋯ξεπονησάμην… (pref. 3). It is more likely that a writer who says ‘When hunting in Lesbos…’ means the reader to think of him as a visitor: a native might be expected to explain that it was his place of birth or residence. It is also more likely that a visitor should require instruction about a picture and local legend than a resident. But it is only a matter of likelihood, and it is always possible to argue that the topoi of the preface have more to do with putting the author in a comparable position to his expected audience than with reporting biographic fact. For details which seem to come from literary sources cf. Scarcella, A. M., La Lesbo di Longo Sofista (Roma, 1968)Google Scholar. One possible motive for a visit to Lesbos would be to receive or give rhetorical instruction as a sophist: we know from Philostratus, V.S. 1.22 (526) that Dionysius of Miletus taught there early in his career (presumably in the first quarter of the second century a.d).

³⁸ That Longus requires the Philetan cowherd Daphnis to tend goats (balancing Chloe's tending of sheep) explains why his ‘Philetas’ must, unlike the Philetan Lycidas, be a cowherd.

³⁹ E.g. Cairns (n. 28) 25. A significant role for Philetas in the development of bucolic poetry has been argued for by Puelma, M., ‘Die Dichterbegegnung in Theokrits “Thalysien”’ Mus. Helv. 17 (1960), 144–64, at 150Google Scholar. Against e.g. Lohse, G., ‘Die Kunstauffassung im VII Idyll Theokrits und das Programm des Kallimachos’, Hermes 94 (1966), 413–25, at 420Google Scholar.

⁴⁰ Antigonus, , Hist. Mirab. 19Google Scholar (23) Keller (= Philetas, fr. 22 Powell) has been taken as evidence for Philetas' treatment of the Aristaeus legend, but this can only be conjectural. Supplementum Hellenisticum Euphorion, fr. 429 was argued by Scheibner, G. in Zucker, F. (ed.), Menanders Dyskolos als Zeugnis seiner Epoche (Berlin, 1965), 103 ffGoogle Scholar. to be by Philetas, and goats and sheep appear in vv. 25–6. But Carden, R. J. D., BICS 16 (1969), 29 ffGoogle Scholar. showed that Euphorion, fr. 130 Powell almost certainly fits 48 and that Euphorion is most probably the author. It seems to me that Philetas could still be the author (v. 6 Ὠρομέδον[το]ς, whether the mountain or an alternative name for the giant Eurymedon, would be easier to explain) but in any case mythological material seems to predominate and we are very far from bucolic.

⁴¹ Hermesianax 7.75–8.

Οἶσθα δ⋯ κα⋯ τ⋯ν ⋯οιδόν, ὃν Εὐρυπύλου πολι⋯ται

Κῷοι χάλκειον στ⋯σαν ὑπ⋯ πλατάνῳ

Βιττίδα μολπάζοντα θοήν, περ⋯ πάντα Φιλίταν

ῥήματα κα⋯ π⋯σαν τρυόμενον λαλιήν.

⁴² I owe this observation to Mr A. S. Hollis.

⁴³ Dover (n. 2) was deterred from accepting his possibility (iii) (which would involve Theocritus ‘acknowledging his debt to a particular older poet who developed the idea of bucolic poetry’) (a) by the apparent lack of pre-Theocritean bucolic poetry (a dangerous argument from silence – all we can say is that we have no surviving poetry clearly datable before Theocritus that bears the marks of the genre as practised by him) and (b) by the fact that ‘the Roman period, which possessed what we do not possess, regarded Theokritos as the inventor of bucolic poetry’. It is not obvious that the Roman period did so regard Theocritus. The life and scholia have no such explicit statement, but simply term Theocritus ⋯ τ⋯ν βουκολικ⋯ν ποιητής or ⋯ τ⋯ν τ⋯ βουκολικά συγγραψάντων ἄριστος (Scholia in Theocritum vetera, ed. Wendel, C., Leipzig, 1914, p. 1Google Scholar (Prolegomena) 4; p. 9.6–7). That the writer of the εὕρεσις τ⋯ν βουκολικ⋯ν (ibid. p. 2.4 ff.) ‘looks not to poets earlier than Theokritos but exclusively to cult’ (Dover, p. lix) does not seem to me to count for or against the possibility that a poet or poets earlier than Theocritus did write something like bucolic poetry: the writer is trying to offer an aition of where bucolic poetry began, not where Theocritus (specifically) looked for his models. (For a brief survey of known pre-Theocritean approaches to bucolic poetry see Dover lx–lxv; I only saw Halperin, D., Pastoral before Theocritus (New Haven, 1983)CrossRefGoogle Scholar after this article had been drafted.) That Virgil chose Theocritus as his model confirms that he was judged the best bucolic poet, but does not show that he was regarded as the ‘inventor’: indeed we shall see that there are some details in the Eclogues which imply the existence of more Hellenistic bucolic poetry than we have.

⁴⁴ Cf. Gow (n. 4) i. 130 n. 7.

⁴⁵ Gow (n. 4) ii. 31 thinks that the use of the article indicates that Theocritus does not intend τ⋯ν Ἅλεντα to refer to the deme Haleis but to a place or river from which the deme took its name. That Haleis in 5.123 is the name of a river in S. Italy and is otherwise attested as a river name is not a good reason for conjecturing that it is a river name in Cos, and it would be odd that the goal of the journey should be stated as a river which plays no later part in the poem. If the article does preclude reading τ⋯ν Ἅλεντα as the name of the deme, then most probably it is the name of the village which gave the deme its name. Although the main centre of population in the deme Haleis has not yet been identified, it is likely to have been in the area of Pyli ; cf. Sherwin-White, S. M., Ancient Cos (Hypomnemata 51, Göttingen, 1978), 59CrossRefGoogle Scholar. If Theocritus expects any knowledge of Coan geography in his readers, he can rely on them knowing that Haleis is a deme whose territory is reached about 10 km from Cos town, and on placing Phrasidamus' farm near the deme centre, some 3 to 4 km into the deme territory and on the foothills of Oromedon (rather than in the flat plain). They may also know that the only deme between that of Cos itself and Haleis is the deme Phyxa, whose centre seems to be near Asphendiou, even further up the slopes of Oromedon; they will therefore appreciate that someone bound for Phyxa would have to part from travellers to Haleis by turning left before the territory of the deme Haleis is reached (cf. Sherwin-White 59 and her map, p. 10; for a map with conjectural deme boundaries, Modona, A. N., ‘L'isola di Coo nell' Antichità classica’, Mem. ist. storico-arch. di Rodi 1 [1933])Google Scholar. Lycidas' departure at 130–1 will therefore make good sense, and his making for Pyxa will corroborate the other indications (e.g. ⋯ν ⋯ρει 51) that his natural habitat is hill country.

⁴⁶ Cf. Gow (1952) i. xx. A reference to a ferry to (?) Calymnos at 1.57 may attest Coan knowledge for the goatherd of 1, as may the reference to the runner Philinus at 2.115 for the lover of 2 (and, less convincingly, the ‘Coan oath’ of 2.160). But neither poem seems to be intended to be set in Cos, and if the setting of 1 is specific, the West is a stronger candidate. 5 seems clearly set near Sybaris, so if the Haleis of 5.123 is Coan it must be taken as a deliberate fusion of South Italian and Coan background, but it need not be (cf. Gow ad loc.). I see no good reason for understanding the στομάλιμνον of 4.23 as a reference to the Coan κώμη (attested by Strabo 14.657, but again cf. Gow ad loc.). (Of course, as Robert Wells pointed out to me, Cos is twice mentioned by name in the encomium of Ptolemy, 17.58 and 64.)

⁴⁷ I am persuaded by Zanker, G., ‘Simichidas' walk and the locality of Bourina’ CQ 30 (1980), 373–7CrossRefGoogle Scholar, that ancient Burina is correctly identified with the spring which now bears that name, quite high on the slopes of Oromedon 5 km south-west of the city of Cos and above the Asclepieion, and that it cannot therefore be the same as the spring by which the harvest festival takes place at 131 ff. (as argued by M. Puelma, op. cit. (n. 38) and by Arnott, W. G., ‘The Mound of Brasilas in Theocritus' Seventh Idyll’, QUCC n.s. 3 (1979), 99 ff.)Google Scholar. Puelma's argument that if Burina is not on Phrasidamus’ farm its mention at 6 ff. lacks point (partly met by Zanker n. 17) is of course fully countered by its role as an allusion to Philetas on my hypothesis. But it may also be relevant that the hillside above the Asclepieion where Burina nestles is clearly visible to travellers to Haleis for the first part of their journey. The repetition of αἴγειροι πτελέαι τε (8 and 136) can also be allowed a function in the poem – to remind the reader of Burina when the harvest festival begins – without insisting that the springs are identical. As indicated in n. 45 I doubt the location of the harvest festival on the plain near Linopóti, as suggested by Paton and adopted by (e.g.) Zanker (see his map p. 377) and Arnott.

⁴⁸ Professor Nisbet reminds me of the Hellenistic admiration for ultra-realism in art, of which Theoc. 1.27 ff. offers an example: note especially (41) ⋯ πρέσβυς, κάμνοντι τ⋯ κάρτερον ⋯νδρ⋯ ⋯οικώς with the following remark that ‘you would say he was fishing with all his strength’. Brown's hypothesis that Lycidas is a cult-statue of Pan would also meet this point, though he does not so exploit it.

⁴⁹ Cf. Cairns (n. 28) 25–6 and n. 112.

⁵⁰ Gow, and Page, , Hellenistic Epigrams (Cambridge, 1965), ii. 149Google Scholar reject the ascription, apparently because the alternative MS ascriptions are to authors not included in Meleager's garland. Dover (149) seems to accept the attribution to Asclepiades. Stadtmüller, endorsed by Beckby, accepted the alternative ascription to Archias.

Apart from this there are no obvious allusions to Asclepiades in Theoc. 7, but in the absence of all but some of Asclepiades' epigrams there may well be allusions to his poetry that we cannot identify.

⁵¹ Theog. 30–3.

⁵² On Tityrus see further below, Section (i). Comatas, of course, is the name of a character in Idyll 5, cf. below, Section (ii).

⁵³ †Τω ου μοι πολέων γαίης ὕπερ ἠδ⋯ θαλάσσης

⋯κ Δι⋯ς ὡραίων ⋯ρχομένων ⋯τέων.

Οὐδ'⋯π⋯ Μοῖρα κακ⋯ν ɸέρει,⋯λλ⋯ μένουσιν

ἔμπεδ' ⋯εί, κα⋯ τοῖς ἄλλα προσαυξάνεται.

It may also be significant that the only appearance of an αἴγειρος and one of the two appearances of πτελέαι in Callimachus are in his Demeter hymn (6.26 and 37), where they (and a πίτυς, ⋯χναι and γλυκύμαλα) are in the desecrated ἄλσος. Another Philetan echo?

⁵⁴ E.g. the threatened violence of Idyll 5.

⁵⁵ Cf. n. 47.

⁵⁶ E.g. Lasserre, F., ‘Aux origines de l'Anthologie: II. Les Thalysies de Theocrite’, Rh.M. 102 (1959), 307–30, at 325Google Scholar, followed by Puelma (n. 38), 156, Lawall, G., Theocritus' Coan Pastorals (Cambridge, Mass., 1967), 102 ff.Google Scholar, and Segal, C., ‘Theocritus' Seventh Idyll and Lycidas’, Wien. Stud. 8 (1974), 20–76Google Scholar, reprinted in id., Poetry and Myth in Ancient Pastoral (Princeton, 1981), 110–166, at 148 ff.

⁵⁷ It may be significant that Theocritus makes no attempt to bridge the gap between arable and pastoral activity by introducing a sacrificial sheep, as does Adaeus, Garland of Philip II = A.P. 6.258, influenced by Theocritus 7.155 ff. In deciding to elaborate his festival of Demeter Theocritus may also be influenced by the popularity of her cult on Cos, and in particular by the existence of an important cult centre in the deme Haleis; cf. Sherwin-White (n. 45), 305–12. As she points out, 312, this sanctuary, belonging to the deme, cannot be the private cult of Phrasidamus' family; but it could of course be a reason for Theocritus seeing especial appropriateness in mentioning — or inventing? — that cult. It also seems unlikely that the Haleis cult of Demeter will not have been mentioned in Philetas' Demeter.

⁵⁸ Aetia fr. 1.9 ff. Do 155–7 indicate that Theocritus intends to turn his hand next to an epyllion like the Demeter? Callimachus' own Hymn to Demeter, arguably intended for recitation on Cos, should owe something to Philetas' Demeter too; cf. n. 53.

⁵⁹ Leo, F., ‘Virgils erste und neunte Ecloge’, Hermes 38 (1903), 3 n. 1Google Scholar, repr. in Ausg. Kl. Schriften (Rome, 1960), ii. 13 n. 1Google Scholar: ‘die Übereinstimmung in der Sache ist so auffallend dass man auf eine direkte Beziehung schliessen muss’. My attention was drawn to this observation by Dr R. G. Mayer.

The ancestry of Eclogue 1 and the influence of the model of Longus 2.3 ff. are of course more complicated than this brief account can set out. For connections with Theocritus, Callimachus and Gallus and the conclusion that the common model must be Philetas cf. DuQuesnay, I. M. LeM., ‘Vergil's First Eclogue’, Papers of the Liverpool Latin Seminar 3, ed. Cairns, F. (Liverpool, 1981), 29–182 at 38–51, esp. 39–40Google Scholar; and Hunter (n. 29), 79–81. As DuQuesnay and Hunter show (and as had already been drawn to my attention by Ms. P. K. M. Kinchin), there are also themes from this nexus in Propertius 1.18.19–32. Thus the echo-motif of Longus 2.7 and Ecl. 1. appears as ‘sed qualiscumque es resonent mihi “Cynthia” silvae, nec deserta tuo nomine saxa vacent ’ (Prop. 1.18.31–2); Philetas, in love with Amaryllis, piped πρ⋯ς ταῖς ɸηγαῖς, cf. ‘sub tegmine fagi’ (Ecl. 1.1) and ‘vos eritis testes…fagus et Arcadio pinus amica deo’ (Prop. 1.18.19–20); and the plights of Philetas in Longus and of Propertius in 1.18 are similar. Prop. 1.18 in turn exploits Callimachus' Acontius and Cydippe (cf. Cairns, F., ‘Propertius 1.18 and Callimachus’, CR 19 (1969), 131–4)Google Scholar and Hunter (81) suggests that Callimachus may there have drawn on Philetas.

I add here a further possible link between Eclogue 1 and Longus which should also go back to a common source: Longus 2.11.3 the progress of Daphnis and Chloe's experience of love might have continued εἰ μ⋯ θόρυβος τοιόσδε π⋯σαν τ⋯ν ⋯γροικίαν ⋯κείνην κατέλαβε. Ecl. 1.11–12: Tityrus' enjoyment of otia is astonishing, because ‘undique totis | usque adeo turbatur agris’. If these two passages do have a common model we shall have to be cautious about treating confiscations and the plight of Meliboeus as an alien element brought into the pastoral by Virgil.

⁶⁰ For a brief review of the knowledge of Latin literature shown by Greek authors of the first two centuries A.D. see Hunter (n. 29), 76–7. Williams, G. W., Change and Decline (Berkeley, Los Angeles, London, 1978), 125–34Google Scholar argues for some cases of anthology poets showing the influence of Latin writing. Fisher, Elizabeth, ‘Greek translations of Latin literature in the fourth century A.D.’, YCIS 27 (1982), 173–215Google Scholar (with bibliography of this question on p. 174 nn. 3 and 4) questions the general view that Greeks were indifferent to Latin language and literature, but only substantiates a different view for the later third and the fourth centuries.

⁶¹ DuQuesnay (n. 59) 60 with nn. 192–200. Cf. Cairns (n. 28) 27 with n. 118, comparing also Tibullus 2.5.31–2.

⁶² Hunter (n. 29) 81–2 (comparing also Bion, 's Lament for Adonis 7–11)Google Scholar.

⁶³ Σ on 78/9, p. 99 Wendel, apparently attributing the story to Lycus of Rhegium (though Λύκος is a conjecture: MSS have Λύκιος). Cf. Gow (n. 4) ii. 152 on 7.78.

⁶⁴ Erucius, Garland of Philip I = A.P. 6.96. Gow and Page allow that Virgil may have known the epigram of Erucius (whom they date tentatively to 50–25 b.c); but this seems to entail the view that Virgil's relocation of Corydon and the pastoral in Arcadia depended on a jeu d'esprit by Wilamowitz, Erucius., Textgeschichte der griechischen Bukoliker (Berlin, 1906), 111 nGoogle Scholar. i judged Erucius to be dependent on Virgil; so too G. W. Williams (n. 60) 126. Hunter (n. 29) 76–7 thinks this ‘not improbable’, but he recognises that Greek knowledge of Latin literature at this period is exiguous (cf. above n. 72). For Reitzenstein's view that similarity indicates a common model, see Reitzenstein (n. 28) 131 n. 2 (on p. 132).

⁶⁵ I would also take the mysterious appearance of Comatas in Idyll 7 within Lycidas' song (78–89) as a prima facie indication that Comatas figured in Philetas — presumably, in view of Idyll 5 and the scholion on 7.78/9 (cf. n. 63), in a South Italian setting. If this, and my other speculative suggestions, were correct, then Philetas' pastoral poetry would be associated with three areas of the Greek world (Lesbos, Arcadia and South Italy), but given the range of geographical interest shown by other Hellenistic poets this is no objection. That Philetas should depict a peaceful pastoral landscape in South Italy c. 290 b.c. is less surprising than Theocritus' location of Idylls 4 and 5 in an area that had been devasted by war since Rome's confrontation with Tarentum in 282 b.c. and the consequent invasion of Pyrrhus. I would see the precedent of South Italian pastoral by Philetas as making Theocritus' choice of setting more intelligible. For an additional reason which may have led Philetas to Comatas and South Italy see below n. 75.

⁶⁶ It the above is correct, it follows that Virgil's influential choice of Arcadia as a chief constituent of his pastoral landscape, apparent first in Ecl. 7 and further elaborated in Ecl. 10, was prompted not merely by Arcadia's associations with Pan and the primitive life but by its pedigree as a pastoral landscape in earlier poetry, that of Philetas. The location would thus be less innovative than on most accounts (e.g. Rosenmeyer, T. G., The Green Cabinet (Berkeley, Los Angeles, London, 1969), 232–46Google Scholar; Coleman, R. G. G., Vergil, Eclogues (Cambridge, 1977), 22Google Scholar and on Ecl. 7.4). A case has also been made for Gallus' part in the location of pastoral in Arcadia by D. F. Kennedy in a paper read to the Gallus colloquium at Liverpool on 23rd April 1983. That Gallus may have so exploited Arcadia counts neither for nor against my suggestion that Philetas did.

⁶⁷ Longus 3.15.1 ff.

⁶⁸ Lyca as the name of a hetaera, Athenaeus 13.567e–f (citing the comedians Timocles and Amphis); Lycaena, Lucian, dial. mer. 12.1. Lycinna is elusive as a real name in the Greek world: it does not appear in Fisk or Pape-Benseler, nor in the indices of CIL vi.

⁶⁹ The pan-pipes are a common element to the divine-encounter nexus (associated with a cave of the Nymphs and precepts about poetry) and the symbolic-gift nexus argued for in (ii) above: on my hypothesis, two Philetan scenes and not just one. It is of course possible that only one Philetan scene is involved, and that I am wrong in tracing to Philetas some of the constituents which require us to suppose two. Note that Propertius 3.3.1 (visus eram molli recubans Heliconis in umbra) has been suggested (by DuQuesnay (n. 59) 40) to belong with Eclogue 1.4 (lentus in umbra) and Propertius 1.18.21 (sub umbras) in alluding to Gallus and Philetas (cf. above n. 59). Another possible verbal link (but perhaps too trivial to be significant) is Eclogue 1.23 sic parvis componere magna solebam; cf. Propertius 3.3.5 parvaque tam magnis admoram fontibus ora; Longus 2.33.2 (a Philetan context, cf. (ii) above) ⋯ δ⋯ [sc. σύριγξ] ἦν μικρ⋯ πρ⋯ς μεγάλην τέχνην οἶα ⋯ν στόματι παιδ⋯ς ⋯μπνεομένη.

⁷⁰ See Gow (n. 4) ii on 7.148.

⁷¹ Pref. 1: …⋯ν ἄλσει Νυμɸ⋯ν…καλ⋯ν μ⋯ν κα⋯ τ⋯ ἄλσος, πολύδενδρον, ⋯νθηρόν, κατάρρυτον· μία πηγ⋯ πάντα ἔτρεɸε, κα⋯ τ⋯ ἄνθη κα⋯ τά δένδρα…

⁷² Hubaux, J., ‘La dieu Amor chez Properce et chez Longus’, Acad. Roy. de Belgie (Bull. Cl. des lettres et des sc. mor. et poi.) 5.39 (1953), 263–70Google Scholar has also argued that similarities between Prop. 3.16.11–20 and Longus 3.5.4 point to a common source in Philetas. I am persuaded by Hunter (n. 2) that they are rather to be explained by use of topoi (more widely attested) than by a common ancestor.

⁷³ Cf. Page, D. L., Sappho and Alcaeus (Oxford, 1955), 294–9Google Scholar.

⁷⁴ 286 (a) Lobel and Page; cf. Page, op. cit. (n. 84) 289–90, and Nisbet, R. G. M. and Hubbard, M., Horace Odes Book 1 (Oxford, 1970), 58Google Scholar on 1.4. The theme of spring's arrival seems also to have been treated in an erotic context by Alcaeus in 296(b) Lobel and Page, too scrappy to make it clear whether the Damoanaktidas addressed is an ⋯ραστής or ⋯ρώμενος. If the latter, then Alcaeus' Damoanaktidas may have some part in the ancestry of Lycidas' ⋯ρώμενος Ageanax. The reputation of Alcaeus as a poet of pederastic love in Theocritus' generation is demonstrated, as Robert Wells has pointed out to me, by Theoc. 29 and 30.

⁷⁵ In considering why Philetas chose Lycidas as the name of a pastoral character one further factor may have been relevant, viz, his debt to Lycus of Rhegium for the story of Comatas (if it is correctly argued above that Philetas told that story, and correctly asserted by the scholiast that it was to be found in Lycus; cf. above n. 63). We may note that this Comatas story was in some way associated with a cave of the nymphs (Σ Theocr. 7.78/9b) and that Lycus had an interest in unusual springs and rivers (cf. FGrHist 570, frr. 7–11, 14) which was not confined to the Western Mediterranean (fr. 14 cites one in India = Pliny, N.H. 13.17). It is possible that Pliny derived his notice about the spring in the island Cydonea that flowed only in spring (cf. Appendix II) from Lycus (though he is not cited among the sources for books 2 or 5).

If this were so, then it might also be that Philetas was indebted to Lycus for local colour about Lycidas' Cydonian haunts as well as for the Comatas story. To choose the name Lycidas would thus be a compliment to a contemporary currently influential in Alexandria (cf. Spoerri, W., Kleine Pauly 3 (1975), 818Google Scholar s.n. Lykos n. 12) as well as an evocation of Alcaeus.

⁷⁶ Especially 130.17 ff. L–P. For the probable location of the sanctuary at which Alcaeus watched beauty contests cf. Stella, L. A., Parola del Passato 11 (1956), 322–3Google Scholar, followed by Robert, L., ‘Recherches Epigraphiques’, Rev. des Ét. Anciennes (1960), 300 ff.Google Scholar, repr. in Op. Min. Sel (Amsterdam, 1969), ii. 816 ffGoogle Scholar. They place it at Messon/Mesa, some 5 km north of the site of Pyrrha at the N.E. end of the gulf of Kalloni. A further (but, I should guess, minor) factor drawing the attention of a Coan poet to Lesbos could have been the legend according to which Cos was colonised from Lesbos by Macareus' son Neandros after Deucalion's flood (Diodorus 5.81.8).

⁷⁷ For the MSS designated by F and V and their relationship see Longus, , Daphnis et Chloe, ed. Reeve, M. D. (Leipzig, 1982), v–xivGoogle Scholar.

⁷⁸ For a recent account of Pyrrha see Paraskevaidis, M., RE 24 (1963), 1403–20Google Scholar. Paraskevaidis thinks the city's acme to have been in the Hellenistic period, arguing from the nearby temple at Messa/Messon, and notes that a basilica at Achladeri and the fact that Pyrrha had one of Lesbos' five bishoprics point to its being a place of some importance in the early Christian period (loc. cit. 1411). Kondis, J. K., Λέσβος κα⋯ ⋯ Μικρασιατική της Περιοχή (Athens, Athens Center of Ekistics, 1978), 344–6Google Scholar, gives greater weight to the notices of Strabo and Pliny (N.H. 5.139 ex his Pyrrha hausta est mari), and concludes that Pyrrhan territory was administered by Mytilene (346).

⁷⁹ Cf. Bowie, E. L., ‘The novels and the real world’, Erotica Antigua (Acta of the International Conference on the Ancient Novel 1976)(Bangor,1977), 4Google Scholar. It is perhaps worth noting that mistakes over the most basic details can be made even by modern novelists who appear to take great pains over a realistic setting. From a novelist educated at Oxford, Rachel Billington, I note: ‘Gordon and I were walking back from dinner at the Mitre…we walked down the Broad. Just as we reached Queen's College…’ (A Woman's Age, Penguin edition, 1981, 135)Google Scholar.

⁸⁰ 3.21.3 ⋯πε⋯ δ⋯ ἄκρᾳ τιν⋯ ὑποδραμόντες εἰς κόλπον μηνοειδ⋯ κα⋯ κοῖλον εἰσήλασαν, μείζων μ⋯ν ἠκούετο βοή, σαɸ⋯ δ⋯ ⋯ξέπιπτεν εἰς τ⋯ν γ⋯ν τ⋯ τ⋯ν κελευστ⋯ν ᾄσματα.

⁸¹ Kondis (n. 78) marks Kydona island/Cydonea as the site of a mineral bath from c. 480 b.c. to the fourth century a.d., (figs. 22–8), but I can find no evidence for this in his text nor have I discovered any elsewhere. I am tempted to wonder whether this classification is based on Pliny's notice about the hot spring and nothing more.

⁸² Cf. above n. 75.

⁸³ See Kondis (n. 78) 73.