previous next

HYMN TO DEMETER

    BIBLIOGRAPHY

    A. FICK, in Bezzenberger Beiträge xvi. (1890) p. 26 f. R. Y. TYRRELL, Hermathena ix. 20, p. 33-40, 1894. V. PUNTONI, L'Inno Omerico a Demetra, 1896. T. W. ALLEN, J. H. S. xvii. p. 49 f. MAASSE. , Ἶρις Indogerm. Forschungen i. 157 f. PRELLER-ROBERT, Griechische Mythologie i.^{2} p. 747-806. L. BLOCH, art. “Kora und Demeter” in Roscher F. LENORMANT, “The Eleusinian Mysteries” in Contemp. Rev., 1880. W. M. RAMSAY, art. “Mysteries” in Encycl. Brit. ninth ed. 1884. L. DYER, Gods in Greece, ch. 2, 1891. P. GARDNER, New Chapters in Greek Hist. ch. 13, 1892. O. RUBENSOHN, die Mysterienheiligtümer, 1892. ROHDEE. , Psyche p. 256 f., 1894. M. P. FOUCART, Recherches sur l'origine etc. des Mystères, 1895. W. PATER, Greek Studies, 1895. D. PHILIOS, Éleusis, ses Mystères etc., 1896. L. R. FARNELL, Cults of the Greek States, ii. ch. 16 (for Hecate), 1896. L. CAMPBELL, Religion in Greek Lit. p. 245 f., 1898. A. LANG, The Homeric Hymns (Translation) p. 53 f., 1899. G. FRAZERJ. , The Golden Bough, second ed. ii. p. 168 f., 1900. M. P. FOUCART, Les Grand Mystères d'Eleusis, 1900. O. KERN, art. “Demeter” in Pauly-Wissowa, Real-Encycl., 1901. N. SVORONOSJ. , Journal Internat. d'Arch. Numism. iv. p. 169 f., 1901. G. F. SCHÖMANN, Griechische Alterthümer (ed. by Lipsius), p. 387 f., 1902. HARRISONE. , Prolegomena p. 150 f., 1903.
Subject.—Persephone, while gathering flowers on the Nysian plain, is carried off by Hades, with the connivance of Zeus. Her cry reaches the ears of Hecate and Helios; Demeter, too, hears her voice, but does not see the rape, or know the name of the ravisher. Distracted with grief, the mother wanders for days seeking news of her daughter. She meets Hecate, who does not know that Hades has done the deed; but the two goddesses go together in quest of Helios, from whom they learn the truth. Then Demeter, angry with Zeus, leaves Olympus and visits the earth in the guise of an old woman. Reaching Eleusis, she meets the daughters of King Celeus, and is engaged to nurse their brother Demophon. She would make the child immortal, but is thwarted by the curiosity of his mother Metanira. She reveals herself to the Eleusinians, commands them to build her a temple, and departs from Eleusis. But she is still wrathful with the gods, and causes a great dearth, so that mankind is in danger of perishing from famine. So Zeus sends Hermes to bring back Persephone from the underworld. Hades, however, has given the maiden a pomegranate seed to eat, which binds her to him; and Demeter, after a joyful meeting with her daughter, tells her that she must now stay with Hades for a third part of every year. The wrath of Demeter is now appeased; she makes the fruits of the earth to grow again, and instructs the chiefs of Eleusis in the performance of her rites, the knowledge of which is necessary for the happiness of men in the nether world.

The Rape and Return of Persephone is a favourite theme in classical poetry. The version of Pamphos is several times mentioned by Pausanias (see on 8, 99, 101); it seems to have been essentially similar to the Homeric hymn, though differing in details, perhaps owing to Athenian influence. Pindar devoted an ode to the subject (Paus.ix. 23. 2), and Euripides tells the story in a choral song ( Hel.1301-1368). There are references to it in Alexandrian literature (Callim. h. Dem. vi., Nicand. Ther. 483-487), and in Nonnus (Dion. vi. 1-168) and the Orphic Argonautica (1197-1201). It was especially popular with the Roman poets: Ovid has two accounts in full ( Fast.iv. 419-616, Met. 385-661); Statius alludes to the myth (Achill. ii. 149-151), and Claudian composed a whole epic de raptu Proserpinae.1

The distinctive features of various ancient poems concerned with Demeter generally, and the rape of Persephone in particular, have been analysed by Pater in his Greek Studies. He pays a warm tribute to the merits of the hymn to Demeter, noting especially its pathetic expression and descriptive beauty. Many readers of the hymn will agree with Prof. Mahaffy (Greek Class. Lit. i. p. 151) in calling it “far the noblest” of the collection. Foreign critics, as a rule, are less favourable; some of the German commentators, and recently Puntoni, among the Italians, have been so much occupied in dissecting the hymn into parts that they appear to have had no time to appreciate its excellence as a whole.2

Relation of the hymn to the Mysteries.—Great as is the poetical value of the hymn, perhaps its chief interest lies in the fact that it is the most ancient and the most complete document bearing on the Eleusinian mysteries. There is nothing esoteric or official in its tone; the writer was not a priest, but a poet, whose primary object was to describe, in fitting language, the pathetic and beautiful story of Demeter and Persephone. But he was an orthodox believer, who had undoubtedly been initiated; and he was at pains to prove that the rites observed at Eleusis were derived from the actual experiences of the divine founders of the mysteries. We can thus reconstruct from his narrative a picture, more or less complete, of the early Eleusinian ritual at a period anterior to the intrusion of Bacchic and Orphic elements. Thanks to the work of Mannhardt and Frazer, much light has now been thrown on the primitive meaning of this ritual—a meaning which had become obscured, if not altogether lost, by the time of the hymn itself.

It seems probable that the early Eleusinian ceremonies were purely agrarian3: the corn was worshipped under two forms—the ripe ear or Corn-Mother (Deo, Demeter), and the new blade or Corn-Maiden (Core).4 When the time of sowing was past and the Maiden was underground, it was thought necessary to propitiate the Mother, or rather, perhaps, to influence her by sympathetic magic, in order to secure the reappearance of the Maiden. Hence the Eleusinians prepared themselves by various acts of ritual to hold communion with the Corn-goddess. During the period of preparation (“κάθαρσις”), the adults fasted (cf. 49), and perhaps abstained from bathing (50). To prevent a failure of the crops, complete purification was required, for their fields, their children, and themselves. They cleansed and fertilised the land by running over it with lighted torches (48). So also they purified their children by making them pass over the fire (239). The women, who in the earliest times seem to have been mainly, if not exclusively, concerned with these rites, held a “παννυχίς” or holy vigil (292). In order, probably, to unite themselves more closely with the goddess, her worshippers pelted one another with stones, until the blood flowed, an offering acceptable to Demeter, as to the gods of many peoples (265). Finally they broke their fast by partaking sacramentally of the body of the Corn-goddess, in the form of a “κυκεών”, or mixture of wheat and water (208).

The development of this primitive Eleusinian religion is a matter of speculation. The simple agrarian ritual may have remained unaltered for centuries; but it is plain that the ideas underlying the ceremonies must have been greatly changed before the age of the hymn. As has been already remarked, an elaborate myth had obscured the meaning of the ceremonies which it purported to explain. The mimetic ritual (to secure the renewal of the crops) had come to be thought a commemoration of the story of Persephone, whose loss and recovery was represented by a sacred play.5 The old agricultural magic had been transformed into a Mystery, and the Maiden had become a great goddess of the underworld, with power to reward or punish mankind after death (see 480-482).

Date of the hymn.—These ideas of future happiness for the souls of the initiated are, of course, quite foreign to Homeric eschatology, and furnish a terminus a quo for the date of the hymn. And there are landmarks in the later history of the Eleusinian cult which supply us with a terminus ante quem. The hymn makes no mention of Iacchus, who played so important a part in the ritual of Eleusis, as known to us from the Frogsof Aristophanes.6 It is true that arguments ex silentio are dangerous, and we cannot be sure that Iacchus was altogether absent from the mysteries when the hymn was composed. There may have been a “δαίμων”, perhaps also known as Plutus (489), connected with the great goddesses from very early times (Lenormant p. 856; Dyer p. 174).7 But we may safely conclude that Iacchus, who was either the brother of Persephone, or her son by Hades, was of little importance until a period subsequent to the age of the hymn (Gardner p. 385, after Lenormant). It follows that the hymn certainly preceded the introduction of Dionysiac rites at Eleusis, when Iacchus was identified with Dionysus (Bacchus). The procession of Iacchus from Athens to Eleusis was established by the time of the Persian war ( Herod. viii. 65); Lenormant is therefore probably correct in assigning the commencement of Dionysiac influence to the first half of the sixth century B.C. The insignificance of Triptolemus and Eumolpus, who are merely two of the Eleusinian chiefs, is also a sign of antiquity (see 153). On these grounds the hymn appears to belong to a date at least not later than the beginning of that century; Lenormant himself (p. 852) assigns it to the end of the eighth or the beginning of the seventh century. Most scholars are substantially in agreement with the view that the hymn is the work of the seventh century; e.g. Förster (p. 39), who suggests the first half, and Duncker (Griech. Gesch. iii. ch. 14), who favours the middle of the century. So Francke (de hymn. in Cer. compositione etc., 1881), following Voss (between Hesiod and Solon).

We may therefore reject the theory of a later date, held by Baumeister (the period of the Pisistratids) and Fick (B. B. xvi. p. 27), who places the hymn between 540 and 504 B.C.

Linguistic evidence is inconclusive, but does not negative the theory of a seventh-century date. Gemoll (p. 279) quotes a number of forms (e.g. “ὄχοισιν, θυσίαισιν, κόρη”) and words (e.g. “ἀδικεῖν, τηρεῖν”) which are not Homeric, and which seem to him to belong to later 8 But we cannot arbitrarily fix a time for their first appearance; we can at most call them post-Homeric. For the evidence of the digamma see p. lxix f.

Place of composition.—Many critics, since Voss, have attributed the hymn to an Attic writer. If the word “Attic” is taken to imply “Athenian,” there is little to be said for the view. The Athenians are nowhere mentioned (the emendation introducing the name in 268 is now abandoned), and there is no hint of the famous procession from Athens to Eleusis. The mysteries appear to be still purely parochial. This silence about any Athenian interest seems to refute the conjecture of Preller (adopted by Baumeister) that the hymn was composed for recitation at the Panathenaea. It is highly probable, in fact, as has often been suggested, that at the time of composition Eleusis was still independent of Athens. Unfortunately the date of the political fusion of Eleusis with Athens is uncertain, although it was undoubtedly not later that Solon,9 and probably took place at least a generation earlier. If this argument is sound, we have also a confirmation from history to support the theory of considerable antiquity for the hymn.

Although the claim for an Athenian origin seems to fail, there is reason to believe that the hymn is “Attic” in the broadest sense of the word, i.e. Eleusinian (Grote Hist. Greece, part ii. ch. 10, Förster, p. 24). The author was clearly familiar with the mythology and topography of Eleusis, and must have been initiated into the mysteries. In no early Greek document, perhaps, is “local colour” so clearly marked. The Eleusinian origin of the hymn has nevertheless been denied by various scholars, whose arguments, however, are not very cogent.10 The principal objection is perhaps the fact that, in the hymn, the descent of Persephone to the underworld takes place at Nysa, whereas local tradition laid the scene at Eleusis itself.11 But this tradition is mentioned by no authors earlier than Phanodemus and Pausanias (see on 17), and we need not suppose that it was primitive. When the Athenians became interested in the mysteries, they localised the scene in Attica itself (Schol. on O. C. 1590; see Preller-Robert i.^{2} p. 759 n. 1); and this implies that there was no rigid and orthodox belief in a “κατάβασις” at Eleusis.

Influence of the hymn.—Extant literature shews little or no trace of any imitation of the hymn. Callimachus may have known it, but there is practically no evidence to be extracted from his poem (see on 49 f.), and he differs from the Homeric version in some particulars (cf. on 200); see Gutsche op. cit. p. 28 f. Apollonius Rhodius may have adapted the episode of Demophon (237 f.) to his account of the childhood of Achilles; but there is nothing in the passage (4.869 f.) which may not be independent. Apollodorus, however, must have been acquainted with the hymn, as his own account of the myth (i. 5) is identical in its main outlines. He disagrees in some details: e.g. Demeter discovers the name of the ravisher from the men of Hermione, not from Helios; Demophon is consumed by the fire; the mission of Triptolemus is narrated. Apollodorus mentions Panyasis and Pherecydes as authorities for the genealogy of Triptolemus; he must therefore have collated their accounts, at least, with the Homeric hymn, and have adopted a composite version of the myth. Actual citations of the hymn appear in Philodemus (see on 440) and Pausanias, who mentions it in three places (i. 38. 2 f., ii. 14. 2, iv. 30. 3).

Diction.—In language, the poem is more closely connected with the hymn to Aphrodite than with any other in the collection (see h. Aphr. Introd. p. 198). The writer was evidently a close student of Hesiod; Francke (p. 11 f.) collects a large number of words and forms in the hymn, which are wanting in Homer, but occur in Hesiod. A passage containing the names of Ocean nymphs is borrowed from the Theogony (see on 417).

Integrity of the hymn.—There is no reason to suspect the presence of any interpolated passages; there is indeed no single line which may not have been original. The story moves in a simple and straightforward way from beginning to end, and all the episodes fall into their proper places. A summary of the various attempts to disintegrate the hymn (by Matthiae, Preller, Hermann, Wegener, and Bücheler) is given by Gemoll (p. 278), and need not be repeated here. The latest editor, Puntoni, while criticising the previous efforts of the “higher critics,” has added a theory, no less unconvincing, to the number. He believes that the hymn as it stands is a fusion of two distinct poems, one of which narrated the rape of Persephone without alluding to Eleusis and the mysteries, while the other treated the mourning of Demeter and the institution of the Eleusinian cult (p. 2, 111). Puntoni apportions the lines of the hymn between these two earlier poems and the additions of a later editor. The grounds for this elaborate and minute dissection are quite illusory; they consist mainly in the supposed unsatisfactory position held by Hecate, and in a number of grammatical and logical incongruities in the text. The most tangible of these are in 53 and 58. It appears unnecessary to refute Puntoni's long argument in detail; his method is inapplicable to early poetry, and perhaps to imaginative literature in any age; some of his objections betray a want of familiarity with epic usage, and even with Greek as a language.12 The conclusion of Baumeister and Gemoll, that the hymn is practically untouched and uncontaminated, is adopted in the present edition.

That no inference can be drawn from the plural “ὕμνοι” in the title (a misapprehension of Bücheler's, ed. p. 3) is plain from its appearance before the other hymns. It is to be read “τοῦ αὐτοῦ ὕμνοι. εἰς τὴν δήμητραν”.

Δήμητραν is the form of the accusative in the title of h. xiii. in all MSS. except It J. is a variant in Theog. 454 and Paus.ii. 14. 3, and is required by the metre in an epigram quoted by Paus.i. 37. 2(Preger Inscr. gr. metr. 203. 2); so orac.ap. Euseb. P. E. v. 34 “εἰς πάτρην φυγάδας κατάγων Δήμητραν ἀμήσεις”.

θεόν: “θεάν” (M) in one syllable is perhaps not impossible; “θεῶν” and “θεᾶς” are common in synizesi in Hesiod and Tragedy; Rzach Dialekt des H. 375. Smyth (Ionic § 28) quotes synizesis in “σάκεα, στήθεα, βέλεα” etc. But the metre practically requires “θεόν”, and Voss's correction is confirmed by h. xiii.1, where M again has “θεάν”, while the other MSS. give “θεόν”.

[2] The rape of Persephone by Hades points to an original “ἱερὸς γάμος”, or annual holy marriage between a god and goddess of vegetation, instances of which are frequent in Greece and elsewhere; see Frazer G. B. i. p. 227 f., ii. p. 186 f., Harrison Proleg. p. 549 f. Here, as often, the marriage is by capture (ib. ii. p. 195 f.) The presence of Hades in the myth suggests an early chthonian triad, Demeter, Core and Zeus Chthonius (Hades, Pluto); see references in Pauly-Wissowa 2754. But the relation of the male God to the two goddesses at Eleusis is uncertain. It may be noted that the “ἱερὸς γάμος” was obscured before the period of the hymn; as Ramsay remarks (p. 127), the annual Theogamia had become a mere disagreeable episode in the life of the two goddesses.

Cf. Theog. 913ἣν Ἀϊδωνεὺς ἥρπασεν ἧς παρὰ μητρός: ἔδωκε δὲ μητιέτα Ζεύς”. For the influence of Hesiod on the hymn see Introd. p. 13.

[4] χρυσαόρου: Hermann thought that the epithet could only have been chosen by an interpolator. But Demeter is “ξιφηφόρος” in Lycophr. 153, where the schol. notes “ἐν τῇ Βοιωτίᾳ ἵδρυται Δημήτηρ ξίφος ἔχουσα”. Possibly the title may suggest that the goddess has won her land by the sword, and protects her agricultural worshippers (so Kern in Pauly-Wissowa 2749, comparing Callim. h. Dem. 137φέρβε καὶ εἰράναν, ἵν᾽ δς ἄροσε τῆνος ἀμάσῃ”, and the name of the hero Triptolemus); but in any case there is little or no fixity of divine attributes in early literature; the golden sword is an epithet of Artemis in orac.ap. Herod.viii. 77.See further on h. Apoll. 395; for the nominative form “χρυσάορος”, h. Apoll. 123.

[5] βαθυκόλποις: see on h. Aphr. 257.

[6] The “ἀνθολογία” of Persephone is a feature in most of the accounts of the rape. It may have been introduced as a natural girlish act, and so have no mythological importance; see parallels in Preller-Robert i.^{2} p. 758 n. 2. On the other hand, flowers play a considerable part in ritual connected with deities of vegetation, so that the “ἀνθολογία” may be paralleled by festivals such as the “ἠροσάνθεια” (Hesych.), at which Peloponnesian women gathered flowers. There was an actual “ἀνθολογία” in the mysteries at Agra; see Svoronos p. 235.

ἴα: see on 8.

[7] ἀΓαλλίδας: Hesychius explains by “ὑάκινθος θρυαλλὶς ἀναγαλλίς”. According to Murr die Pflanzenwelt in d. griech. Myth. p. 246 it is an iris.

ὑάκινθον: for the hyacinth (hyacinthus orientalis, Murr) in connexion with Demeter (Chthonia) see Paus.ii. 35. 5.Hyacinths are frequently mentioned among the flowers gathered by Proserpine; cf. Ov. Fast.iv. 437 f., Ov. Met.v. 392.Here, however, it is perhaps introduced simply as a common spring flower, as in Il. 14.348 κρόκον ἠδ᾽ ὑάκινθον”, and often in later poetry, e.g. Mosch.i. 65(a similar list of flowers in the rape of Europa), h. Pan 25.

[8] Νάρκισσόν: see on 12 and 428. The narcissus was the peculiar flower of the Great Goddesses; cf. O. C. 683, Hesych. “Δαμάτριον: ἄνθος ὅμοιον ναρκίσσῳ”. The origin of the connexion is perhaps uncertain; at all events we may doubt whether it was due to etymology (“νάρκη” the numbness of death), as some suppose (Preller-Robert i.^{2} p. 760, Pater Greek Studies p. 103, 152). There may have been a later mystic explanation. The flower was certainly chthonian, being also sacred to the Eumenides (schol. l.c. from Euphor. fr. 43, Düntzer). It was planted on graves (Anth. Plan. App. 120). The narcissus was specially mentioned by Pamphos in his version of the rape: Paus.ix. 31. 9κόρην τὴν Δήμητρός φησιν ἁρπασθῆναι παίζουσαν καὶ ἄνθη συλλέγουσαν, ἁρπασθῆναι δὲ οὐκ ἴοις ἀπατηθεῖσαν ἀλλὰ ναρκίσσοις”. Pausanias' allusion to “ἴα” refers to the common tradition; Aus. Mir. 82, Diod.v. 3(the Sicilian version), Förster p. 31. On the violet see Cook in J. H. S. xx. p. 1 f.; he compares Bacchyl.iii. 2, for its connexion with Persephone, which, however, is not very clearly marked, although in later times it was distinctly funereal. In the hymn, attention is drawn to the narcissus, not to the violet, which is only one among a number of flowers. Later poets generally include it in their list of flowers in the “ἀνθολογία”; cf. Nicand. Georg. fr. 74. 60ὑάκινθον ἰωνιάδας τε χαμηλὰς

ὀρφνοτέρας, ἃς στύξε μετ᾽ ἄνθεσι Περσεφόνεια”; Ov. Met. v. 392aut violas aut candida lilia carpit”; Shakespeare Winter's Tale iv. 4. 116 f. violets dim.
ὃν φῦσε δόλον: cf. Il. 8.494 ὅν ποτ᾽ ἐς ἀκρόπολιν δόλον ἤγαγε”. καλυκώπιδι: this beautiful epithet is not found in Homer; cf. h. Aphr. 284, h. Dem. 420, and Orph. h. lxxxix. 2.

[9] Πολυδέκτͅη: so 404, 430 “Πολυδέγμων”. The idea of Hades as the “host of many” is especially Aeschylean: cf. Suppl. 157 “τὸν πολυξενώτατον Ζῆνα τῶν κεκμηκότων”: P. V. 152 “Ἅιδου τοῦ νεκροδέγμονος”: Theb.860πανδόκον εἰς χέρσον”. See Preller-Robert i.^{2} p. 804. On the euphemistic names of Hades and Persephone see Rohde Psyche p. 192.

[10] τό Γε: this correction depends upon Homeric usage, and gives good sense: the confusion of TE and TE is of course common; cf. Il. 5.853, P. V. 42, 248 etc, and 280 “αὐτῆς” for “αὐγῆς. τότε” can hardly be justified.

[12] τοῦ καί=Il. 1.249τοῦ καὶ ἀπὸ γλώσσης”, where Leaf notes that the “καί” introducing a merely epexegetic sentence is very unusual.

ἑκατὸν κάρα: as the flower is miraculo<*>sly created, the exaggeration of its “hundred heads” need not be pressed; but the writer is doubtless thinking of the Narcissus tazetta, the “polyanthus” or “bunch” species (see on 428), Murr p. 248.

[13] Tyrrell's correction of “κῶδις τ᾽ ὀδμῆ” is recommended by the fact that it only posits the omission of a syllable (“ζη”); for such omissions cf. p. xviii, and h. Apoll. 407 (“πρῶτα” for “πρώτιστα” in all MSS. except M). For the construction cf. Od. 5.59, ι” 210, and for the crasis of “καί” cf. Il. 2.238, Ζ” 260 with Leaf's note, Il. 13.734, γ 255, ζ 282, κοὔ 227, κἀγώh. Herm. 173, “κἀκ πολλῶνTheog. 447, “κοὔ” Parmenides 51, “καὐτοί” Xenophanes vi. 5. See Kühner-Blass Griech. Gramm. i. p. 225, Smyth Ionic § 308 for exx. in other poets, H. G. § 377, La Roche H. U. i. p. 283 f., van Leeuwen Ench.p. 50 f.

[14] ἐΓέλασσε: see on h. Apoll. 118.

[15] ἄμφω: here indeclinable; a use not found in early epic. Cf. Apoll. Arg. 1.165 (gen.), 1169 (dat.), Theocr. xvii. 26.

[16] χάνε δὲ χθών: this explanation was natural when the scene of the ascent or descent of Pluto was localised on a plain; so, according to the actual Eleusinian tradition, the chariot disappeared through the opening ground (fragment of a vase from Eleusis, Ath. Mitth. xxi. pl. 12; J. H. S. xxii. p. 3). In some traditions Pluto disappeared in a cave ( Ausc. Mir. 82). At Enna he ascends through a cave, and descends into the open ground, Diod.v. 3. 4.

εὐρυάγυια: in Homer of cities only. The epithet is less suitable to “χθών”. Gemoll compares “δίκα εὐρυάγυια” (Terpand. fr. 6), for a more general use.

[17] Νύσιον ἂμ Πεδίον: on the various places called Nysa see i. 8. Whether the Nysian plain is here purely mythical, or whether the poet was thinking of a particular place, it is impossible to say. Förster (p. 268 f.) argues for the Carian Nysa; Preller-Robert (i.^{2} p. 758 n. 3) for the Thracian. The poets generally speak of Nysa as a mountain (e.g. Soph. Ant.1130, i. 8), but the locality is so vague that “πεδίον” may well stand; cf. Apoll. Arg. 2.1214οὔρεα καὶ πεδίον Νυσήϊον”. Hesiod does not localise the myth, but the schol. on Theog. 913 lays the scene by the Ocean. Various other places are mentioned: e.g. Crete (Bacchyl. fr. 64), Eleusis itself (Phanodemus fr. 20, Paus.i. 38. 5, Orph. h. xviii. 15); see Introd. p. 12, and Preller-Robert i.^{2} p. 759, Roscher ii. 1313, Förster l.c. In later times the Sicilian tradition prevailed (first in Carcinus ap. Diodor. v. 5; cf. Mosch.iii. 128, Opp. Hal. iii. 489, and often in Latin poetry; Ov. Fast.iv. 353, Ov. Met.v. 385, Lucan vi. 740, Stat. Ach.ii. 150, Claud. de rapt. Pros. ii. 71). Modern poets have chiefly followed the Romans: That fair field Of Enna where Proserpine, gathering flowers, Herself a fairer flower, by gloomy Dis Was gathered (Milton).

ὄρουσεν ἄναξ: the trochaic caesura in the fourth foot is very rare, except when the caesura is preceded by an enclitic or other monosyllable; see on 248. Tyrrell (Hermath. ix. 20 p. 34) suggests “ὄρουσ᾽ ἅναξ”, to avoid breaking a “law universal in Greek poetry from Homer to Nonnus.” But the exceptions to the rule in Homer are amply sufficient to justify the text; see H. G. § 367. 2, Hermann Orphica p. 693, van Leeuwen Mnemosyne, 1890, p. 265 and Ench.p. 18-22, Eberhard Metr. Beob. i. p. 23 f. The last word is usually of four syllables as in Il. 6.2, Ω 60, ρ” 399= Od. 20.344, and here; or five, as in Od. 18.140 and h. Apoll. 36 (where however see note); very rarely of three, as in Od. 12.47. The law is more rarely broken in post - Homeric verse; examples are Theog. 23Ἑλικῶνος ὑπὸ ξαθέοιο”, Theog. 319πνέουσαν ἀμαιμάκετον πῦρ”, Scut. 222ὥστε νόημ᾽ ἐποτᾶτο”. In Theog. 435 Köchly transposes “ἀγῶνι ἀεθλεύωσιν” and in Op.693 for “φορτἴ ἀμαυρωθείη” one MS. has “φορτία μαυρωθείη”. Sometimes, as in Od. 5.272 ὀψὲ δύοντα”, Theognis 881, Tyrtaeus fr. 7. 1θεοῖσι φίλος”, the two words are rhythmically one; but Theognis 931 “οὐδὲ θανόντ᾽ ἀποκλαίει”, id. 981 “λόγοισιν ἐμὴν φρένα θέλγοις” are real exceptions; cf. id. 923. In later literature the following exceptions may be noted: verse ap. Phaedr. 252B, Hom. Epigr.vii. 1, Orph. h. liii. 3, lxxxv. 5; Evenus fr. i. 5 (Gaisf.); Pythag. χρυς. ἐπ. 6, 37, and 70; ep.ap. Paus.iv. 1. 8(line 3); and often in Oppian (Ven. i. 190, ii. 60, 120, 202, 536, iii. 237, 244, iv. 232, 431). There are several exceptions in Diog. Laert. (Anth. Pal. vii. 96. 3, vii. 104. 1, vii. 126. 1); so Schol. ibid. vii. 568. 1.

[18] Πολυώνυμος: first in Hesiod and h. Apoll. 82. Preller thinks the epithet specially appropriate to Pluto, whose titles were numerous; see Preller-Robert i.^{2} p. 804, Rohde Psyche 192 f. For the “ἐπωνυμίαι” of Pluto cf. Paus.ix. 23. 4(on a hymn to Persephone by Pindar) “ἐν τούτῳ τῷ ᾁσματι ἄλλαι τε ἐς τὸν Ἅιδην εἰσὶν ἐπικλήσεις καὶ χρυσήνιος, δῆλα ὡς ἐπὶ τῆς Κόρης τῇ ἁρπαγῇ”. So in h. Apoll. 82 Apollo is “πολυώνυμος”, i.e. has many titles in different lands. On such accumulation of titles see Lobeck Agl. i. p. 401, who quotes e.g. Ov. Met.iv. 11 f., Gruppe Culte u. Mythen i. p. 555 n. 44, Adami p. 222 f. (where many references are collected), viii. Introd. The primary meaning of the word may therefore stand, in the case of gods; but, as applied to inanimate objects, “πολυώνυμος” is simply “famous”; cf. Theog. 785 (“ὕδωρ”), Pind. Pyth.i. 17(“ἄντρον”).

[19] χρυσέοισιν: cf. Pindar's epithet “χρυσήνιος” from Paus. quoted above. ὄχοισιν: in 375 “ὄχεσφιν”, which Voss and others needlessly read here.

[20] ἰάχησε: so xxvii. 11; forms from “ἰαχέω” do not occur in early epic; but cf. xxvii. 7 “ἰαχεῖ”, Callim. h. Del. 146ἰαχεῦσα”.

[21] ὕπατον καὶ ἄριστον=Il. 19.258 (nom.).

[23] ἐλαῖαι: this is usually held to be corrupt, but no emendation is at all satisfactory; the conjectures, apart from their graphical eccentricity, err in endeavouring to introduce a person or persons (Demeter or the nymphs). But the categories “ἀθάνατοι” and “θνητοὶ ἄνθρωποι” are exhaustive, with the exception specified in 24. Any title of Demeter is peculiarly out of place: she heard the second and louder cry 38, 39, which sets her in motion. The reading of M “ἐλαῖαι” runs counter to the usual notions of Greek poetical taste. This, however, is no reason for suspecting the text. In late, especially Latin, poetry inanimate nature is often personified (e.g. Verg. Ecl.i. 38Verg. Ecl., x. 13, and many instances given by Forbiger). We have to learn that the idea was earlier than has been supposed. The sense here would be: “neither gods nor men heard her; and the trees were deaf” (J. H. S. xvii. p. 50). The nearest analogies in Greek poetry are Bioni. 31τὰν Κύπριν αἰαῖ

ὤρεα πάντα λέγοντι καὶ αἱ δρύες αἰαῖ” “Αδωνιν
καὶ ποταμοὶ κλαίουσι τὰ πένθεα τᾶς Ἀφροδίτας” and Theocr. vi. 74. So even in prose, Lycurgus 150 “νομίζοντες οὖν Ἀθηναῖοι ἱκετεύειν ὑμῶν τὴν χώραν καὶ τὰ δένδρα, δεῖσθαι τοὺς λιμένας”. If this view is thought untenable, we are thrown back on Ilgen's “Ἕλειαι” or “Marsh-nymphs” (= “νύμφαι ἑλειονόμοιApoll. Arg. 2.821, 3. 1219). In favour of this, it may be noted that the Nymphs form a class apart from gods and men; cf. h. Aphr. 259. But, as Tyrrell notes, “νύμφαι” seems absolutely required; cf. Theocr. v. 17τὰς λιμνάδας Νύμφας”.

[24] Περσαίου: Hecate is daughter of the Titan Perses (=Persaeus here) and Asterie, according to Theog. 411, Apollod.i. 2. 4.Other poets give other genealogies; see Farnell Cults ii. p. 502, Preller-Robert i.^{1} p. 322, Roscher 1899.

ἀταλὰ φρονέουσα: “ἀταλός” (the der. is doubtful) seems properly to refer to youthful merriment; cf. Il. 18.567, λ” 39, Theog. 989 (others translate “tender”; so Rouse in K. Z. 1899, xxxv. p. 462, connecting the words with a priv. and “τάλ-ας”, i.e. “not capable of endurance,” cf. E. M. 161. 47). The E. M. explains Il. 6.400 παῖδ᾽ ἀταλόφρονα” by “ἁπαλὸν φρόνημα ἔχοντα, τουτέστι νήπιον, ἀνόητον”. The sense “merry” does not seem particularly suitable to Hecate in this connexion. Baumeister, followed by Gemoll, understands “kindly,” i.e. to Demeter; but there is no authority for this meaning, nor is it easy to see how “ἀταλὰ φρονεῖν” could be appropriate to a “κουροτρόφος” (a title of Hecate), as others assume; “κουροτρόφος” is not the same as “κοῦρος”. Possibly the author thought of Hecate as a young goddess “with youthful thoughts.” See also L. Meyer Griech. Et. i. s.v. “ἀταλός” “kindlich”; Prellwitz Et. Wört. p. 37 “jugendlich.”

[25] ἐΞ ἄντρου: cf. Apoll. Arg. 3.1213κευθμῶν ἐξ ὑπάτων” (of Hecate). No particular cave is meant. Whether Hecate was originally a moon-goddess, or, as Farnell supposes, an earth-goddess, a cave would be appropriate for her home. In this hymn, at all events, she is certainly a moon-goddess, as is shewn by the mention of Helios in 26. So Sophocles (fr. 480) associates Helios and Hecate as sun and moon. Hecate heard the cry, but did not see the rape, as it was daytime, and she was therefore in her cave; Helios heard (“ἄϊεν” 25), and of course saw also (cf. on 70).

[27] Zeus absents himself intentionally, in order that he may not appear to connive at the rape (cf. on 3).

28-29. Cf. h. Apoll. 347-348.

[29] δέγμενος: generally explained as a perf. part., without reduplication, and with irregular accent (from *“δέγμαι”, probably an older form of “δέδεγμαι”. See Leaf on Il. 2.794, H. G. § 23). But it may be a present form; Leaf remarks that there is no reason for supposing that the affection of “χ” by “μ” is confined to aor. and perf. stems. Cobet altered to “δέχμενος”.

[35] μητέρα . . φῦλα are almost certainly objects, not (as Gemoll) subjects, of “ὄψεσθαι”.

[37] ἔθελγε μέγαν Νόον: Il. 12.255 θέλγε νόον” (in a bad sense). Gemoll does not accept Hermann's lacuna. He explains: “so long as she hoped that her mother and the other gods would see her, she trusted (that her cry would avail) and (she called out so that) the mountains echoed.” But if this is the meaning intended, the wording is most obscure. The lacuna seems necessary. The change in sense between 37 and 38, and the absence of protasis to “δ̓”, require at least another line. The case is different from those noted on 127. The sense of the lost passage, as Francke saw, is “but when she saw the earth opening to swallow her, then she despaired and shrieked loudly.”

[40] χαίταις: for the Attic dative cf. 205, 308, 441. Hermann reads “χαίτῃς ἀμβροσίῃς”.

[42] κυάνεον δὲ κάλυμμα=Il. 24.93; see on 182.

[43] ἐΠὶ τραφερ́ην τε καὶ ὑΓρ́ην= Il. 14.308, υ” 98, imitated in later epic, as Opp. Ven. i. 11. For the omission of “γῆν” and “θάλασσαν” cf. h. Aphr. 123.

[45] Ἤθελεν: (with neg.) “had no mind” (=had not the power). “ἐθέλειν” implies a desire to do what is, or seems to be, in one's power to do, and so is often practically equivalent to “δύνασθαι”. Cf. Il. 9.353, Φ” 366.

[46] Bücheler and Francke reject this line. The stress on οἰωνῶν is unusual, but Gemoll compares “ζώεινh. Aphr. 221, with accent [acutemacr] ¯. There are various emendations which give a usual but characterless verse. The line is modelled on Il. 22.438.

[47] ἐΝν̂ημαρ: it is generally assumed from this word that the fast at Eleusis lasted nine days. This is not improbable, and is supported by parallels; see Roscher die Enneadischen, etc. Fristen, 1903, p. 16 f. (Abhandl. Sachs. Gesellsch. xxi.), who compares a festival at Lemnos, where fire was put out for nine days ( Philostr. Her.19. 14); the Thesmophoria ( Ov. Met.x. 434); the fast of Clytia (ib. iv. 262); the Italian Bacchanalia (Livy xxxix. 9). Roscher is probably right in explaining the number as representing an ancient week, one-third of a lunar month (op. cit. p. 14 f.). There is, however, no other allusion to the length of the Eleusinian fast; and in the present passage “ἐννῆμαρ” may be purely conventional, to express a round number of days, with no special reference to the actual duration of the fast. A period of nine days or nights is common in Homer: Il. 1.53, Ζ 174, Ι 470, Μ 25, Ω 107, 610, 664, 784, η 253, ι 82, κ 28, μ 447, ξ” 314; Theog. 722, 724, h. Apoll. 91. The Sicilian festival of the two goddesses mentioned by Diod.v. 4 lasted for ten days.

Δηώ: first here for “Δημήτηρ”, then often in poetry. The form is usually regarded as hypocoristic (Mannhardt Myth. Forsch. p. 295, Preller-Robert i.^{2} p. 748, Pauly-Wissowa 2713).

[48] στρωφᾶτ̓=Il. 13.557. The form is probably late; see Leaf on Il. 15.666.

αἰθομένας . . ἔχουσα=Od. 7.101 (“ἔχοντες”). For the significance of the torches, which play so large a part in the myth and ritual of Demeter, see Introd. p. 10, Lenormant ii. p. 124 f. On the whole subject of fire-festivals see Frazer G. B. iii. p. 238-326, who thinks that the use of torches in such cases “appears to be simply a means of diffusing far and wide the genial influence of the bonfire or of the sunshine which it represents” (p. 313). He quotes many examples (p. 255, 313 f.) to shew that the avowed intention of torch-lighting is often to fertilise the fields, or to prevent blight, etc.

49-50. Compare the mourning of Demeter in Callim. h. Dem. 17αὐσταλέα ἄποτός τε καὶ οὐ φάγες οὐδ᾽ ἐλοέσσω”. This, however, may be independent of the hymn.

[50] βάλλετο: the editors quote Il. 11.536 and other passages where the act. “βάλλειν” has the meaning of “sprinkle,” “wet.” No other instance seems to occur of the middle “βάλλεσθαι” in this sense, unless we accept Hermann's “λουτρά τ᾽ ἐπιβαλοῦ χροΐ” (for the corrupt “ἐπὶ χροῒ βάλε” or “βάλλευ”) in Eur. Or.303.

[51] A formulaic line (only here) similar to Il. 6.175 ἀλλ᾽ ὅτε δὴ δεκάτη ἐφάνη ῥοδοδάκτυλος Ἠώς”, and Il. 24.785 (“φαεσίμβροτος”). φαινολίς: Ruhnken compares Sapph. fr. 95ἔσπερε πάντα φέρων ὄσα φαίνολις ἐσκέδασ᾽ αὔως”.

[52] σέλας: for a torch, Apoll. Arg. 3.293, 4.806. Here it is probably collective, “torchlight,” as the regular attribute of Hecate is a torch in either hand; cf. the plur. “χείρεσσι”. So “δάος”=“δαΐδας” in the formula “δάος μετὰ χερσὶν ἔχουσαι” (Il. 24.647 and elsewhere). For the attribute of Hecate see Roscher 1900 f. Farnell (Cults ii. p. 549 f.) thinks that the torch was originally the symbol of Hecate as a chthonian deity, not as the moon, with which, however, the hymn-writer plainly identified her (see on 25). For the connexion of Hecate with Demeter and Persephone see on 440.

[53] ἀΓγελέουσα: Hecate (or Artemis) was called “ἄγγελος” at Syracuse (Hesych. s.v. and Schol. on Theocr. ii. 12), but it is unlikely that there is here any allusion to this title.

The “news” which Hecate gives is that she heard Persephone's cry—a circumstance which certainly was unknown to Demeter. Hence “ἀγγελέουσα” needs no emendation, and the difficulties about this part of the narrative, and the inference based on them as to the composition of the hymn, are imaginary. Maass “ε. Ἶρις”, I. F. i. 164 accepts the continuity of the text (though reading “ἀγγέλλουσα”, which is virtually the same as the future).

[54] ὡρηφόρε: the hiatus is legitimate in the bucolic diaeresis; H. G. § 382 (2). On the epithet, “bringer of the seasons,” see Mannhardt Myth. Forsch. p. 227, who compares Anth. Pal. vi. 98. 1 “Δηοῖ λικμαίῃ καὶ ἐναυλακοφοίτισιν Ὥραις”.

[55] θεῶν οὐρανίων: not Homeric. For “θεός” a monosyllable cf. Od. 14.251 θεοῖσιν”. So Theog. 44 “θεῶν”, and perhaps Il. 1.18 θεοί”. Below, 259, 325.

[57] φων̂ης Γὰρ Ἤκους᾿: the exx. of “γάρ” lengthened by ictus are mostly before “οἱ” or “εὑ”: Il. 2.342, Ζ 38, Ι 377, δ” 826, etc. But cf. Il. 2.39, Τ” 49, where “γάρ” before a vowel appears to be established. “γάρ ῤ̔” would be simple, and the collocation of the two words is confirmed by the metre in Il. 13.352 and other passages, although in other places “ῥα” may have been inserted from mistaken metrical grounds. Of course “μέν” may have dropped out here, as perhaps in 122, in which case “ἄκους᾿” would naturally be altered to “ἤκους᾿”.

[58] ὅς τις ἔηΝ: parenthetical; see 119, and note on h. Herm. 208.

ὦκα λέγω Νημερτέα: the explanation given in J. H. S. xvii. p. 52 (=“λέγω πάντα σοὶ ὦκα εἶναι νημερτέα”) is improbable, as “νημερτής” must be closely connected in a predicative sense with “λέγω”; see Ebeling s.v. But the text may be correct: Hecate asseverates the truth of her statement by a common formula; cf. 433, Od. 11.137 τὰ δέ τοι νημερτέα εἴρω”; “I tell thee truly (all I know).” “ὦκα” is unusual with the present, but justified by the context, “and I tell it quickly.” Hecate wishes to spare Demeter disappointment, by confessing her ignorance at once. Hermann's lacuna (with “λέγοι”) seems therefore unnecessary.

[63] στὰν δ᾽ ἵπΠων Προπάροιθε=Il. 24.286, ο” 150.

[64] σύ Περ recurs 116, and Ludwich's conjecture is excellent on palaeographical grounds; cp. h. Herm. 308ἐνέχων δε” M=“ἕνεχ᾽ ὧδε”. The stroke to denote “ν” in “θεα_” (=“θεαν”) was no doubt taken for a circumflex.

[66] κούρηΝ τ̀ην ἔτεκον: the antecedent is attracted to the case of the relative, as in Il. 10.416, Ξ” 75, 371. H. G. § 267. Cf. Verg. Aen.i. 573urbem quam statuo vestra est. This “inverse attraction” (for “τὴν κούρην”) is slightly different from the attraction of a nominative absolute to the case of the relative, as in Il. 6.396 θυγάτηρ . . . Ἠετίωνος: Ἠετίων, ὃς ἔναιε, α” 50. In Od. 8.74 (“οἴμης τῆς”) the gen. may be partitive, or due to either of these forms of attraction.

[67] ἀδιν́ην: see Leaf on Il. 2.87. The word is often used with verbs or substantives expressing grief, where it seems to mean “loud” or “vehement.” The derivation, and consequently the original meaning are obscure (Leo Meyer Handbuch der gr. Etymologie, 1902): Göbel's suggestion (“” intens. and [root ]“δε” ‘move’) is as probable as any. The primary sense would then be “quick” or “busy.” Prellwitz Et. Wört. s.v. suggests a connexion with “ἀδήν”.

δἰ αἰθέρος ἀτρυγέτοιο=Il. 17.425. Elsewhere “ἀτρύγετος” is applied to the sea. The derivation and meaning are unknown. The ancients connected the word with “τρύγη”, i.e. “unharvested,” “barren,” or with “τρύειν”, “unconquered” (by tempests), see Ebeling. Modern scholars have generally adopted one of these derivations. Prellwitz s.v. sees in “-τρυγ-” the German Dorf, Eng. thorp, with the same general sense.

[70] καταδέρκεαι ἀκτίνεσσι: cf. Od. 11.16, where “καταδέρκεσθαι” (here intrans.) is more naturally constructed with an obj. acc. “καταδέρκεται” in M is a common scribe's error (e.g. Il. 10.82 ἔρχεαι ἔρχεται, 115 νεμεσήσεαι νεμεσήσεται”) assisted by the similar context in Od. 11.16, where the verb is in the third person. “ὄπωπεν” followed naturally.

[71] The writer has a reminiscence of Od. 3.93 κείνου λυγρὸν ὄλεθρον ἐνισπεῖν, εἴ που ὄπωπας” (cf. 65 “ ἔπει ἔργῳ” with Od. 3.99).

[76] μέγα ἅζομαι: unless with Ruhnken we insert “ς᾿” there is an hiatus, which however may be justified by Theog. 532 “ταῦτ᾽ ἄρα ἁζόμενος” (this is practically the MS. tradition, as the only variant is “ἄῤ” for “ἄρα”; see Rzach's note). Curtius “γρυνδζüγε” p. 162, Prellwitz s.v., and Fortunatov K. Z. xxxvi. 46 assume an initial yod which would produce hiatus. The same explanation is sometimes given of “ὡς” making position (H. G. § 397).

[77] οὐδέ is suspected by Wegener and Gemoll. Puntoni (p. 52) defends the text. “οὐδέ” may be illogical for “οὐ”, but it is quite natural after the parenthetic clause “δὴ γάρ κτλ.” (cf. 32). The sense of the passage is: “you shall know all (for I pity you); and you are to know that Zeus alone is to blame.” See further on vii. 56. Indeed “οὐδέ” is hardly to be distinguished from “οὐ” in several Homeric passages; see Fränkel in Album Grat. to Herwerden p. 61 f., who quotes Il. 19.420, Υ” 133 etc. (“οὐδέ τί σε χρή”). In Il. 16.225 (“οὐδέ τις ἄλλος”) the “δέ” has force.

[79] θαλερ́ην: the special epithet of a young husband or wife, like the “blooming” bride in English ballads; so with “γόνος”, h. Aphr. 104, and with “γάμος, ζ 66, υ” 74, h. Pan 35.

82-83. There is no reason to eject οὐδέ τί σε χρ̀η . . χόλον. Hermann altered Γόον to “χόλον” on the ground that the formula “ουδέ τί σε χρή” introduces a repetition of a previous statement (Il. 7.109, Τ” 67 etc.). But the duplication of “χόλον” is intolerable; and as “γόος” is the expression of “χόλος” there would be no difficulty, even if the present passage were from the old epic.

[85] ἀμφὶ δὲ τιμ́ην, “in respect of honour”; the wording, if somewhat prosaic, is correct. The order is like that of Hes. Op.74ἀμφὶ δὲ τήν γε”. The proposed alterations (“τιμῇ” or “τιμῇς”) rest on the analogy of h. Herm. 390ἀμφὶ βόεσσιν” and ib. 172 “ἀμφὶ δὲ τιμῆς” (so MSS., “τιμῇς” Gemoll). But for “ἀμφί” with acc. cf. h. Herm. 57, viii. 1, xxii. 1, xxxiii. 1. These exx. are all of “speaking about,” but Pind. Isthm.vii. 8Pind. Isthm., 9 has both dat. and acc. in a wider sense.

[87] τοῖς: rightly explained by Franke as demonstrative: Hades dwells among those over whom he is lord.

88-89. Cf. Scut. 341-342. τανύπτεροι is to be taken with οἰωνοί, not with “ἵπποι”. Nothing is said in this poem about winged horses, although Gemoll compares Eur. El.466.

[90] αἰΝότερον καὶ κύντερον: cf. 305 f., Od. 11.427.

[92] Νοσφισθεῖσα, “rejecting,” as in h. Herm. 562 and orac.ap. Hendess 119. 7 “νοσφισθεῖσα γέρα προτέρων τιμάς τε παλαιάς” (of Deo).

[94] ἀμαλδύνουσα: not, as in Homer, “destroying,” but “disguising.” Baumeister compares Apoll. Arg. 1.834, 4.112.

[95] βαθυζώνων, “low-girt,” i.e. girt over the hips. The epithet, which occurs in Il. 9.594, γ” 154, is apparently not synonymous with “βαθύκολπος”, as the ancient grammarians and most editors assume; see on h. Aphr. 257.

[96] Κελεοῖο: this is the usual tradition for the king's name; cf. Paus.i. 39. 1(Pamphos), Apollod.i. 5. 3. schol. on Eq.695; see further in Roscher ii. 1026 f. The schol. on Nicand. Alex. 130 calls the king Hippothoon (the eponymous hero of the Attic tribe) with Metanira as his wife. For other accounts see Förster p. 12. There was a cult of Celeus and his daughters at Eleusis ( Protrept. i. p. 39), and a shrine of Metanira ( l.c.).

[99] Παρθενίῳ φρέατι: for the metre of “φρέατι_” cf. 101, 248; La Roche Hom. Unter. i. p. 49, H. G. § 373. The local dative is amply supported by examples in H. G. § 145; it is here not harsher than “τραπέζῃ” “at table” (Od. 21.35). See further on 308 and h. Aphr. 173. Gemoll objects that the “Ionic” form is “φρείατι” (“φρήατι”), while in Attic “φρέατι” has a long. But Herodotus uses “φρέαρ”, and the hymn - writer might naturally adopt the epic quantity (“φρεία^τα Φ” 197). On the forms of the word see Brugmann Grundriss ii. p. 236, 342 f., Prellwitz s.v.

The “Maiden well” is not mentioned again in the hymn; it is most probably identical with the “Flowery well,” at which, according to Pamphos, Demeter sat; cf. Paus.i. 39. 1φρέαρ ἐστὶν Ἄνθιον καλούμενον. ἐποίησε δὲ Πάμφως ἐπὶ τούτῳ τῷ φρέατι καθῆσθαι Δήμητρα κτλ.” Frazer (l.c.) thinks it may be the spring called Vlika, about a mile and a half west of Eleusis, on the road to Megara. The well is not to be confused with the Callichorum, which was close to the precinct of Eleusis (see on 272), although the fame of this latter well led several ancient writers to identify it with the place where Demeter rested; cf. Callim. h. Dem. 16, Nicand. Ther.486, Apollod.i. 5. 1; in Orph. Arg. 729 a river in Asia is called both Parthenius and Callichorus, probably in view of this literary tradition. The accounts of Pamphos and the present hymn no doubt follow the ancient Eleusinian tradition; see further on 200. The last hemistich is a formula: Od. 7.131, ρ” 206.

[101] Γρηΐ Παλαιγενέϊ ἐΝαλίγκιος: the corn-spirit, in the form of the last sheaf, is often called the “Old Woman,” “Grandmother” etc.; see Frazer G. B. ii. p. 170 f. It has been suggested that in “γρηΐ” we have a survival of the otherwise nameless corn-spirit. Jevons even holds that the corn-goddess was known simply as “γραῦς”, and her daughter as “κόρη”, until the Athenians identified the two with Demeter and Persephone (p. 367, 378 f.). But it is difficult to believe that the Eleusinian goddesses were nameless until so late a period. Indeed, as far as regards the hymn, the metamorphosis of Demeter into an old woman need have no special significance; some disguise was necessary for the purpose of the story. Compare the account of Pamphos mentioned by Paus.i. 39. 1(“γραῒ εἰκασμένην”). For a similar disguise cf. Il. 3.386, of Aphrodite, which shews that the present passage may be due to epic influence.

[105] Ἐλευσινίδαο: son of Eleusis, the eponymous hero of the place, Paus.i. 38. 7.He was also called Eleusinus, Hyg. Fab.147, Serv. on Georg.i. 19, alibi.

[106] εὔηρυτον: (only here) formed, like “κοτυλήρυτον Ψ” 34, from “ἀρύω” which first occurs in Hesiod.

108-110. Pausanias causes a difficulty in this passage: in i. 38. 3 he states “καλοῦσι σφᾶς” (the daughters of Celeus) “Πάμφως τε κατὰ ταὐτὰ καὶ Ὅμηρος Διογένειαν καὶ Παμμερόπην καὶ τρίτην Σαισάραν”. Puntoni considers the lines interpolated, following Hermann, who, however, subsequently retained 108 reading “τρεῖς ὡσεί τε”. The name “Καλλιδίκη” in 146 would on this view have been substituted for another, unless the whole verse has been interpolated. An interpolation however is on general grounds highly improbable, and later than Pausanias' time out of the question; it would be more legitimate to suppose an early variant. Cf. n. crit. on 476. Gemoll thinks that the text of Paus. is corrupt, suggesting “καλεῖ δὲ σφᾶς <οὐ> κατὰ ταὐτὰ καὶ Ὅμηρος . . Διογένειαν κτλ.”, the gap being filled with the names Callidice etc. Preller, Baumeister, and others suppose that Paus. may have made a slip of memory, which seems the most probable solution of the difficulty.

[111] ἔγΝων: for the Homeric “ἔγνωσαν”. So Pind. Pyth.ix. 85(136). The correct form, however, seems to be “ἔγνον” which Cobet restores. Compare “ἔβα^ν”, and “ἔφα^ν” 118.

[113] Παλαιγενέων ἀΝθρώπων: a variation of the Homeric “τίς πόθεν εἶς ἀνδρῶν; Bücheler's “χαμαιγενέων” is no improvement.

[115] The form “πιλνᾶς” given by M being transitive (“πιλνᾷ” 3 sing. Hes. Op.510, “πίλναται” passive Apoll. Arg. 4.952), Voss's “πίλνασαι”, as preserving the sigma, seems preferable to Hermann's “πιλνᾷ”, which rests only on the analogy of “δαμνᾷ Ξ” 199 (called Doric in schol. T ad loc.). The syllable “αι” was omitted, as in “ἐπιβήσεσθ̓” 332, from the effect of the hiatus.

[119] αι<*> τινές ἐστε: parenthetical, as in 58. Demeter speaks as though she did not know their names.

[122] Δωσώ: the name is uncertain; Fontein's “Δηώ” is some way off “δώς”, and as Demeter (called “Δηώ” in 47) invents her story, it is natural that she should give a fictitious name. Brunck's “Δώς” is not elsewhere found as a proper name, but=“δόσις” in Hes. Op.356.This, however, requires the insertion of “μέν”, and preference may be given to Passow's “Δωσώ”, where the vowel could easily have been lost before “ἐμοί”. In either case there might possibly be a mystic allusion to the corn as a “gift” to men (see Pater p. 102).

[123] Νῦν αὖτε: for “νῦν δέ” as in Od. 22.6. Ruhnken's “νῦν δ᾽ αὖτε”, though of course common, is therefore needless. Κρ́ητηθεν: editors see an allusion to the early worship of Demeter in Crete, as if the writer wished to hint this fact, even in a fictitious story. For the Cretan cult see Diod.v. 77.The myth of Iasion (Od. 5.125) was localised in Crete, Theog. 970. Miss Harrison believes in Cretan influence at Eleusis (Proleg. p. 565 f.). But the explanation is unnecessary; the name of Crete would naturally occur to any one who wished to give a plausible account of his parentage or travels. In Od. 14.199 f. Odysseus invents a Cretan home. Cf. also Od. 13.256 f., Od. 19.172 f.

[126] Θορικόνδε: the town and deme of Thoricus (Therikó) was N. of Sunium, with a harbour now called Mandrí. See Leake Demi of Attica p. 68. It was one of the twelve independent cities of Attica until the time of Theseus (Strabo ix. p. 397). For its history and remains see Frazer on Paus.i. 31. 3.

κατέσχεθον: the construction “κατασχεῖν νηΐ” is not Homeric, but occurs in Herodotus and Attic (Francke).

[127] Hermann's lacuna is perhaps unnecessary, considering the elliptical style of this hymn generally; cf. 317, 446. Of course a step in the narrative is omitted. For the Homeric custom of landing for meals cf. Od. 14.346, ι 85, κ 56, ο” 499. This passage seems to be a reminiscence of that in “ξ”, where Odysseus escapes from the Thesprotian sailors.

[128] ἐΠηρτύνοντο: Francke objects to the verb, on the ground that it is not used by Homer in the middle, and should mean “fix on.” But the simple verb “ἀρτύνω” is found in the middle, with the sense here required “prepare”: cf. Il. 2.55=Il. 10.302 ἠρτύνετο βουλήν”. Homer, however, has “ἐντύνεσθαι” with “ἄριστον, δαῖτα, δεῖπνον”.

[129] δόρποιο: used in the proper sense of supper; cf. Od. 14.347 δόρπον ἕλοντο” with “ἑσπέριοι” 344. δεῖπΝον in 128 must therefore be general for any meal, or perhaps for the principal meal of the day, here supper.

[132] τιμ̂ης (for “ὤνου”) is not Homeric ( Herod. and Attic).

[133] Demeter feigns ignorance of the name of the country, although in 126 she mentions Thoricus. But Eleusis is sufficiently far from Thoricus to justify the word “ἀλαλημένη” and to give colour to her feigned ignorance of the place.

[137] The key to this difficult passage is “τέων”, which is of course interrogative. To follow “οἰκτείρατε” it would have to be relative. Therefore rather than write “τέως” (un-Homeric in the sense of “until”) with Ruhnken, it seems better to assume a lacuna containing a verb to govern “τέων”, e.g. (on the analogy of the corresponding line 149) “τοῦτο δέ μοι σαφέως ὑποθήκατε, ὄφρα πύθωμαι”. The termination “-ωμαι” coming before “ἵκωμαι” and “ἐργάζωμαι” would account for the omission. The answer, 149 f., implies a question. Attempts have been made to give “τοκῆες” its full metrical value, but the synizesis is probably genuine; cf. “βασιλῆεςHes. Op.263, and perhaps “ἱππῆες Λ” 151. So “ἐπηετανός” (quadrisyll.) Hes. Op.607, h. Herm. 113.

[140] ἀφ́ηλικος: not in Homer. Cf. Moeris p. 82 “ἀφηλικεστέραν, πρεσβυτέραν Ἀττικῶς”. But “ἀπῆλιξ” is found in Herod.iii. 14(in compar.). In Il. 22.490 (a late passage) “παναφῆλιξ” has a different sense.

[144] δεσπόσυνον: first in Pind. Pyth.iv. 267(475). διδασκ́ησαιμι: for the form cf. Hes. Op.64ἔργα διδασκῆσαι. διαθ[ρ]ήσαιμι” and “διαθ[λ]ήσαιμι” are of course easier changes than Voss's “διδασκήσαιμι”, which also involves the alteration of “γυναικός” to “γυναῖκας”. The sense, however, is very near, and the corruption not greater than some of those known in M (p. xviii). “διασκήσαιμι” (cf. the variant in l.c.) would be little removed from “διαθήσαιμι”.

148-9=216-7. Cf. Solon fr. 5. 64δῶρα δ᾽ ἄφυκτα θεῶν γίγνεται ἀθανάτων”, Rhian. ap. Stob. 54 “φέρομεν δὲ θεῶν ἑτερόρροπα δῶρα

ἀφραδέϊ κραδίῃ”. The early editors doubted the mood of “τέτλαμεν”, and Brunck's alteration was to suit an infin. (“τετλάμεν”). The indic. is certainly right; cf. Od. 20.311.

[151] Cf. Scut. 105 “δ̀ς Θήβης κρήδεμνον ἔχει ῥύεταί τε πόληα. κρ́ηδεμνα”: applied to the walls of Troy, a ‘diadem,’ Il. 16.100, ν” 388. Compare the epithet “ἐϋστέφανος”. So Il. 2.117πολίων κάρηνα”. See also vi. 2.

[153] Τριπτολέμου: for Triptolemus and the other princes cf. 474 f. According to Paus.i. 14. 2Triptolemus was the son of Trochilus or (the Athenian version) of Celeus. Apollodorus (i. 5. 2) calls him the eldest son of Celeus and Metanira, but mentions other genealogies, i.e. that of Panyasis (son of Eleusis and Demeter) and that of Pherecydes (son of Oceanus and Ge). Hyginus fab.147 and Serv. on Georg.i. 19 give a different parentage (Eleusinus and Cothonea or Cyntinia). For the later myth of Triptolemus see Preller-Robert i.^{2} p. 770 f., Harrison M. M. A. A. p. xlix f. (and Eumolpus). The derivation “τρίς, πολεῖν” must now be abandoned, as Triptolemus had no early connexion with the plough (Kern de Tript. Aratore, 1887; cf. Lehrs Aristarch.^{2} p. 459, von Wilamowitz Aus Kydathen p. 132); for the name cf. Neoptolemus etc.

For Dioclus cf. Plut. Thes.10(a king of Eleusis). In 474Plut. Thes., 477 the form is “Διοκλῆς”; Ruhnken compares the double “Ἴφικλος, Ἰφικλῆς” and others. Polyxeinus and Dolichus appear to be abstracted from titles of Pluto; for Polyxeinus (whose name is not elsewhere mentioned in connexion with Eleusis) cf. on 9 “Πολυδέκτῃ”. Dolichus is certainly an epithet of Pluto; cf. von Prott in Ath. Mitth. xxiv. p. 251 [“πλουτο]νι δ[ολι]χοι”. Elsewhere he is a son of Triptolemus (Dolichius), Eusth. 306 on Il. 2.625, Byz.Steph. ; Herodian (“π. μον. λέξ”. p. 10) quotes a line “Εὔμολπος Δόλιχός τε καὶ Ἱπποθόων μεγάθυμος”.

Eumolpus, like Triptolemus, is here only one of the Eleusinian chiefs; his fame as the first hierophant and founder of the priestly family is later than the hymn.

The genitives depend on “ἄλοχοι, τῶν πάντων” being explanatory.

[154] ἀμύμονος: as Pausanias in his citation (see crit. note) expressly says that Homer calls Eumolpus “ἀγήνωρ”, Ruhnken and others would exchange the epithets in 154, 155, reading “ἀγήνορος Εὐμόλποιο” and “πατρὸς ἀμύμονος”. But Pausanias' quotation is probably a casual error, influenced by the next line.

[156] Πορσαίνουσι: probably intrans., “manage in the house.” Ruhnken takes “δώματα” as an object, joining “κατά” with the verb.

[157] Πρώτιστον is sound. For this feminine form in comparative and superlative adjectives cf. Od. 4.442 ὀλοώτατος ὀδμή”, Theog. 408 (“Λητὼ”) “ἀγανώτατον ἐντὸς Ὀλύμπου”, Pind. fr. 152γλυκερώτερος ὀμφά”. For exx. in prose see Kühner-Blass i. p. 554 n.

[159] θεοείκελος: the gods, when they are disguised as mortals, often shew a nobility which excites admiration; cf. the disguise of Apollo (h. Apoll. 464 f.), of Aphrodite (h. Aphr. 92 f.), and of Dionysus (vii. 17 f.).

[160] εἰ . . ἐΠίμεινον=Od. 17.277, where of course “ἐθέλεις” (the proper Homeric form) is found. Hermann is probably right in restoring it here; cf. 137. For the later “θέλω” see on h. Apoll. 46.

[164] τηλύγετος: M. and R. on Od. 4.11 summarise Savelsberg's view (Rhein. Mus. 1853) that this word=adolescens, “grown big” (*“τῆλυς” “great”), and is applied to boys and girls from the age of about thirteen to twenty or more. Leaf on Il. 3.175 approves. This explanation takes no account of the present passage, where Demophon is quite an infant. Fick Wörterbuch i.^{4} 440 connects the word with “τᾶλις” a bride: Prellwitz s.v. sees in the latter part the root of “ὐγιής” etc. It is of course possible that the writer was ignorant of the real meaning, and understood the word as referring to an only son, or to one born to his parents in old age, as the ancients variously explained. Francke and Gemoll think that the sense “lateborn” could not have been here meant, as “ὀψίγονος” follows; but pleonasms are quite in the manner of this poem; cf. “πολυεύχετος” and “ἀσπάσιος” 165, and the synonyms in 124.

[165] Πολυεύχετος: only here, for “πολυάρητος” in Homer and below 220.

[168] θρεπτ́ηρια: see on 223. δοίη: sc. Metanira: cf. 223 “δοίην”.

[170] κυδιάουσαι: for the occasional retention of the original “-άω” etc. see H. G. § 55. Instances in the hymns are h. Aphr. 266, vii. 14, 41.

[172] ᾡς, “according as”; so 295, 416. Ruhnken's “ὅσς᾿” is quite needless.

[174] It is noticeable that here and in 401 M represents the diphthong “ει” by “η”; cf. also h. Apoll. 9. “ἤαρος” may be a genuine form (i.e. a correct transcription of a prae-Euclidean E), or it may be a confusion with “ἦρος, ἠρινός”. Homer only uses “ἔαρος, Ζ 148, τ” 518 (but see Agar in J. P. xxviii. 1901, p. 80 f.). For “ἦρος” cf. 455.

[176] The picture of girls raising their dress to run is not found in Homer or Hesiod. The action, as Francke notes, is commonly represented in art from the seventh century, and (although Gemoll rejects the idea) it is quite possible that the writer may have been influenced by such works of art (Francke p. 26). At all events, the pictorial touch is rather after the manner of a later poet. Baumeister compares

ἂν δὲ χιτῶνας
λεπταλέους λευκῆς ἐπιγουνίδος ἄχρις ἄειρον

. (Compare this description of maidens running by the side of the chariot with the simple statement in Od. 13.84, 319.)

[177] ἀμφὶ . . ἀΐσσοντο: borrowed from Il. 6.509 (of a horse). So “κυδιόων Ζ” 509=“κυδιάουσαι” 170, and Il. 6.400 παῖδ᾽ ἐπὶ κόλπῳ ἔχουσα”=187 (“ὑπό”).

[178] κροκηΐῳ: only here; for the form (=“κροκέῳ”) cf. “κουρήϊον” (“ἄνθος”) 108, also “ἅπαξ εἰρ”. For the colour cf. Ars amor. i. 530 croceas irreligata comas.

[182] κατὰ κρ̂ηθεν: Il. 16.548, where see Leaf, Od. 11.588, Theog. 574, and “ἀπὸ κρῆθενScut. 7. The stem “κρη-” appears in “κρήδεμνον, κρήνηH. G. § 107, n. 5. The covered head, and the “κυάνεος πέπλος” are, of course, signs of mourning; cf. Demeter “Μέλαινα” at Phigalia Paus.viii. 42, Pauly-Wissowa 2734.

[183] θεᾶς may be restored, as in 210 M gives “θεᾷ”. For the confusion of “η” and “α” in the MS. see 147.

186=Od. 1.333 (“στῆ ῥα”) and elsewhere. τέγεος, properly any roofed space, is here the “μέγαρον”.

[187] ὑΠό: we should expect “ἐπί”, as in Il. 6.400 (Gemoll). But the variation is trivial; in Od. 15.469 ὑπὸ κόλπῳ” is used though with a slightly different sense.

188-211. Preller brackets these lines as interpolated, and others eject the whole or part of the passage. Preller's reasons are quite inadequate, as Baumeister, Gemoll and others point out.

188-189. Objection has been needlessly raised to this account of Demeter's miraculous entrance, in spite of which Metanira does not seem to recognize her divinity (cf. 213 - 215). She seems, indeed, to suspect that her visitor is something out of the common (190), just as Demeter appears “θεοείκελος”, i.e. “noble,” to the girls (159). But when her momentary fear has gone, she is ready to accept Demeter as a mortal. Compare Anchises' original scruples (h. Aphr. 92f.), and his acceptance of Aphrodite's denial of divinity. Even more striking is the indifference to a miracle shown by the Tyrrhenian captain in the hymn to Dionysus; see vii. Introd. p. 228, and notes on h. Dem. 159, h. Apoll. 465.

[188] μελάθρου κῦρε κάρη=h. Aphr. 173. Gemoll thinks that the present passage was borrowed from the h. Aphr. while Abel reverses the debt. In both places the words seem equally suitable. Gemoll argues that “μέλαθρον” is properly used of the roof-timbers in the h. Aphr., but improperly here for the lintel; but this is hypercritical. Indeed, we may suppose the goddess to have just crossed the threshold and to be standing actually in the “μέγαρον”.

[189] “Πλ̂ησεν κτλ.”: miraculous light marks the presence of the gods: cf. h. Apoll. 444 (of Apollo), Eur. Bacch. 1083 (Dionysus), Ov. Fast. i. 94lucidior visa est quam fuit antc domus”; so infra 278.

[191] κλισμοῖο: on the “κλισμός” see Helbig H. E. pp. 118, 122. It was more luxurious than the “πηκτὸν ἕδος” (=“δίφρος” 198) which Demeter accepted. Matthiae compares Athen. v. 4 and Od. 19.55 f.

[193] φαεινοῦ: epithet of “θρόνος, Λ” 645. The “κλισμός” is “πολυδαίδαλος Ω” 597, and “ποικίλος α” 132, i.e. inlaid, or studded with silver (“ἀργυρόηλος”). In Il. 8.436 the epithet “χρύσεος” is ideal, for the chairs of gods.

[194] The last hemistich=h. Aphr. 156. Cf. Verg. Aen.xi. 480oculos deiecta decoros.

[195] Ἰάμβη: the episode of Iambe and Demeter is related by Apollod.i. 5. 1γραῖά τις Ἰάμβη σκώψασα τὴν θεὸν ἐποίησε μειδιᾶσαι. διὰ τοῦτο ἐν τοῖς θεσμοφόροις τὰς γυναῖκας σκώπτειν λέγουσι”: Nicand. Alexiph. 130; cf. Diod.v. 4, E. M. and Hesych. s.v. The scholia on Nicand. l.c., Hephaest. p. 169, Eustath. p. 1684 attribute the invention of the iambic metre to Iambe. The connexion is absurd, although it may have been present in the mind of the writer of this hymn. As Gemoll notes, there is no proof that the Eleusinian raillery was uttered in iambic or any other metre; it was no doubt impromptu. The schol. on Nicand. Ther.484 mentions Ambas as a son of Metanira who laughed at the sacred rites; this suggests a connexion with Iambe, whose similarity to “ἴαμβος” must be accidental. Iambe's jesting is here a mythological explanation of the banter which was a feature of the Eleusinia. No doubt the jesting was part of the primitive festival, although the literary references mostly mention the practice in connexion with the Athenian period of the Eleusinia. According to the schol. on Plut. 1014 the Athenian women abused one another, on their way to Eleusis in carriages; cf. also Suidas s.v. “τὰ ἐκ τῶν ἁμαξῶν”. There was a similar custom at the “στήνια” (Athens): see Preller-Robert i.^{2} p. 778. So Epidaurian women railed at each other at the parallel festival of Damia and Auxesia ( Herod.v. 83; cf. Frazer on Paus.ii. 30. 4). In these cases the raillery was peculiar to women, who were so intimately connected with agricultural rites. But at the Eleusinia there was also a custom known as “γεφυρισμός”, in which men and women alike seem to have abused and jested with the procession at a bridge on the Eleusinian road. See Ran. 384 f., Strabo ix. p. 400, E. M. p. 229, Hesych. s.v. “γεφυρισταί”, Svoronos p. 297. There was a general “αἰσχρολογία” in the Sicilian festival ( l.c.). The custom is probably due to the widespread idea that abuse of a person or his belongings brings good luck (by avoiding the “φθόνος θεῶν” or the evil eye, etc.). Frazer (G. B. i. p. 97 and on Paus.i. 37. 3) quotes, among other examples, Theophr. Hist. Plant. viii. 3, Quaest. Conv. vii. 2. 2; a Greek sower of cummin must curse to avoid failure of the crop.

The raillery of Iambe is akin to the indecencies associated with Baubo (Babo), who was actually worshipped at Paros (see inscr. quoted on 491) and certainly figured in the Eleusinian cult of Demeter ( Harp. s. v. “Δυσαύλης”, Protrept. ii. 77).

[199] Cf. Il. 5.879 ταύτην δ᾽ οὔτ᾽ ἔπεϊ προτιβάλλεαι οὔτε τι ἔργῳ”.

[200] ἀΓέλαστος: this has been referred to the tradition that Demeter sat upon an “ἀγέλαστος πέτρα”: Apollod.i. 5. 1, schol. on Eq.782, Suidas s.v. “Σαλαμῖνος”, Hesych. s.v. The situation of the stone cannot now be identified. Apollodorus places it by the Callichorum, but this is no authority, as he does not seem to follow the local tradition in regard to the resting-place of Demeter (see on 99). The stone is mentioned in a fourth-century inscr. (“Ἐφ. Ἀρχ”. 1883 p. 115); it was probably near Athens, and unknown in the old Eleusinian myth; see Svoronos p. 247 f. In any case it should be noted that the word “ἀγέλαστος” has no immediate connexion with the “ἀγέλαστος πέτρα”, for Demeter is now sitting “ἐπὶ δίφρου” (198) in the house.

The latter hemistich=Od. 4.788.

ἄπαστος: Callimachus (h. Dem. 8), who says nothing of Iambe, makes Demeter break her fast in the evening: “ἕσπερος ὅς τε πιεῖν Δαμάτερα μῶνος ἔπεισεν”. This supports the theory that the Mystae fasted only till sunset (cf. the Mohammedan Ramadan; see Ramsay p. 126 n. 5).

202-205 bracketed by Matthiae and others, needlessly. Hermann (Epist. cv) objects to “μιν” followed by “πότνιαν ἁγνήν”, but this apposition is quite Homeric; see Il. 21.249, ζ” 48, and cf. the frequent use of the pronominal “” in apposition with a proper name. He is also offended by the inelegancy of 204 and by “ὀργαῖς”, 205. Francke thinks that “πρίν γ᾽ ὅτε δή” in 202 was written by an imitator of 195; but the writer of 195 may surely have repeated himself.

[204] ἵλαον σχεῖν θυμόν: Hes. Op.340ὥς κέ τοι ἵλαον κραδίην καὶ θυμὸν ἔχωσιν”. The metre (with the last three feet spondaic) is not common, except in stereotyped endings, as in “θνητοῖς ἀνθρώποις” (or the genitive of this formula) 11, 22, 29, 45, 55, 73, 403, and often in Homer. In 195, 202 “Ἰάμβη κέδν᾽ εἰδυῖα” the older epic form was of course “κέδνα ϝιδυῖα” (Od. 1.428 etc.). In 302 “ξανθὴ Δημήτηρ” is formulaic (=Il. 5.500); so 452 “κρῖ λευκόν” =Od. 4.604. With the present line cf. 417, 421, 474. The number of “spondaic” verses (i.e. with the last two feet spondaic) is much greater in this hymn than the proportion in the first book of the Iliad (e.g.) or in the hymn to Apollo (see Schürmann de h. in Cer. aetate etc. p. 55 f., Francke p. 23, and see generally Eberhard Metr. Beob. i. p. 10 f., La Roche Wiener Studien xx. p. 70 f.).

[205] ὀργαῖς, “humour,” “mood,” a sense common both in sing. and plur. See L. and IambeS. , who was Demeter's companion as long as she remained in Celeus' house, “pleased her afterwards also,” not merely for the moment. The double dat. (“οἱ . . ὀργαῖς”) presents no difficulty; for the “σχῆμα καθ᾽ ὅλον καὶ μέρος” in the dat. compare Il. 1.24, Θ 129, Ν” 82, Scut. 221, Herod.vii. 16.

[207] οὐ Γὰρ θεμιτόν: cf. schol. on Nicand. Alex. l.c. δὲ θεὸς οὐκ ἐδέξατο, λέγουσα μὴ θεμιτὸν εἶναι πιεῖν αὐτὴν οἶνον ἐπὶ τῇ θλίψει τῆς θυγατρός”. Jevons (p. 379 f.) thinks that wine is here a surrogate of blood and was for this reason excluded from the non-animal sacrifices to cereal deities. For wine as akin to blood see Frazer G. B. i. p. 358 f., and for bloodless offerings to Demeter or other deities of vegetation cf. e.g. Paus.viii. 42. 11.So the Eleans did not pour wine to the Despoinae. But Demeter and Persephone did not as a rule object to animal sacrifice: pigs were offered at the Attic Thesmophoria, and at Thebes ( Paus.ix. 8. 1); see Schömann Griech. Alterth.^{4} ii. p. 232 f. And, since human blood seems, at least originally, to have been shed during the Eleusinia (see on 265), the goddesses can hardly have objected to wine as its substitute. It need hardly be noted, in fact, that abstention from wine would be natural in any fast, such as took place in the Eleusinia.

[208] The passage refers to the “κυκεών”, the institution of which the hymnwriter, according to his wont, ascribes to Demeter herself. The drinking of this mixture of meal and water was the actual means of communion with the goddess. and belonged therefore to the most sacred part of the ritual in the “τελεστήριον”. The mystae received certain objects from the hierophant and answered “ἐνήστευσα, ἔπιον τὸν κυκεῶνα, ἔλαβον ἐκ κίστης, ἐγγευσάμενος ἀπεθέμην εἰς κάλαθον, καὶ ἐκ καλάθου εἰς κίστην” ( Protrept. 18, Arnob. v. 26; see Lobeck Aglaoph. i. p. 25, Harrison Prolegomena p. 155).

For the “κυκεών” in Homer see Il. 11.624 f., Od. 10.234 f. In the latter passage it is called “σῖτος”, being compounded of “ἄλφιτα”. but it is always drunk (“ἔκπιον κ” 237). So Eusth. 870. 65 “εἰ καὶ μεταξὺ βρωτοῦ καὶ ποτοῦ κυκεὼν εἶναι δοκεῖ, ἀλλὰ μᾶλλον οἷα ζωμός τις ῥοφητὸς ἦν”, comparing Il. 11.640 f. Cf. Ar. Pax712οὔκ, εἴ γε κυκεῶν᾽ ἐπιπίοις βληχωνίαν”, schol. on Nicand. Alex. 128 f. (“ἔπιε”).

On the sacramental eating of corn see Frazer G. B. ii. p. 318 f.

[211] ὁσίης ἕνεκεν, “to observe the rite,” as practised by the mystae. The expedients to bring the apodosis into this line are violent. “ἐπέβη” is far removed from “ἕνεκεν”, which gives admirable sense and is defended by I. T. 1461 “ὁσίας ἕκατι”. Another suggestion, “πίε πότνια”, is equally rash. The lacuna has Puntoni's support; it must contain the verb of drinking. The missing verse may have run somehow as follows: “ἔκπιεν: δὲ λαβοῦσα δέπας θέτο ἔνθ᾽ ἀνάειρε”.

Πολυπότνια: not in early epic, but cf. Thesm. 1156, Apoll. Arg. 1.1125, Orph. h.xl. 16 (of Deo). The writer of this hymn is fond of compounds with “πολυ-”; cf. 9Orph. h., 17Orph. h., 18Orph. h., 28Orph. h., 31 etc.

[213] χαῖρε: not here a salutation at meeting, but a courteous form of address or congratulation after some incident has occurred: Baumeister compares Od. 18.122 (after pledging a guest in wine, = “your health”), Od. 11.248, θ” 408, 413.

ἐΠεὶ οὔ σε κακῶν κτλ.”; cf. h. Aphr. 132οὐ μὲν γάρ κε κακοὶ τοιόνδε τέκοιεν”, and a close parallel in

οὐ σέ γέ φημι κακῶν ε<*>`ξ
ἔμμεναι οὐδὲ κακοῖσιν ἐοικότα φύμεναι αὐτόν:
οἷόν τοι μέγα εἶδος ἐπιπρέπει

(possibly an imitation of this passage; but “κακῶν ἔξ” is in Il. 14.42 and for “εἶδος ἐπιπρέπει” Gemoll compares Od. 24.252).

[214] αἰδώς, “dignity,” a sense not in Homer.

216-217. Cf. 147-148.

[217] Ζυγός: only the neut. in Homer. For the phrase cf. Hes. Op.815ἐπὶ ζυγὸν αὐχένα θεῖναι βουσί”, Theog. 1023 “ὑπὸ ζυγὸν αὐχένα θήσω”, where the gender is indeterminate, but is probably neuter. Callimachus (fr. 467) is the first writer who certainly uses “ζυγός” in the sense “yoke,” but Plato (Tim. 63 B) has the masc. for “balance.”

221-223=166-168, with small variations.

[223] δοίηΝ is certainly to be retained; the mother would reward the nurse with “θρεπτήρια”, when the child grew up. This is not to be confused with the “θρεπτήρια” in Hes. Op.188, of the return made by the child to his parents in their old age; so “θρέπτρα” (the Homeric form) in Il. 4.478, Ρ” 302.

[227] κοὔ: objections have been raised to the crasis, which, however, is perfectly tolerable; cf. n. on 13.

228-230. ἐΠηλυσίη, “witchcraft,” is certain (cf. h. Herm. 37), but “ὑποταμνόν” and “ὑλοτόμοιο” are puzzling. The former has been explained as a “cut herb,” used in sorcery, but the formation hardly allows such a meaning. Voss's “οὔτε τομαῖον” (sc. “φάρμακον”) is too violent. The same editor altered “ὑλοτόμοιο” to “οὐλοτόμοιο” (a non-existing word), i.e. herbs cut for harmful purposes. In the Class. Rev. 1895, p. 13 it was suggested that “ὑποταμνόν” and “ὑλοτόμοιο” are superstitious paraphrases for the worm (“ἕλμινς” or “σκώληξ”), and that Demeter knows of a remedy against this children's complaint. For such paraphrases cf. Aratus 959 “σκώληκες”,

κεῖνοι τοὺς καλέουσι μελαίνης ἔντερα γαίης”, and Hesiod's “φερέοικος” “snail,” “ἀνόστεος” “cuttle-fish,” “ἴδρις” “ant.” See A. Cook B. “Descriptive Animal Names in Greece,” Class. Rev. 1894, pp. 381 f., where a large number of similar substantives or epithets are collected. If this view is correct, the translation will be: “neither shall witchcraft hurt him, nor the Undercutter (Borer); for I know an antidote far stronger than the Woodcutter.” This involves the accentuation “ὑποτάμνον”, a participle used as a substantive, like “ἀμείβοντες, ἀμφιφῶν, Ἔμπουσα, κελέοντες”. The objection is that “ὑλοτόμοιο”, the wood-cutter appears unsuitable as a paraphrase for the parasitic worm. In Hermath. i. p. 142 Davies retained “ὑποταμνόν”, and suggested “οὐλοτόμοιο” from “οὖλα” “gums,” i.e. gumcutting. But as Tyrrell notes, these words are strangely formed if they denote a process. “οὐλοτόμοιο” should be active, and mean “gum-cutter.”
Davies is, however, probably right in seeing an allusion to “teething,” the first inevitable trouble of childhood. It may therefore be suggested that the “ὑποτάμνον” and “οὐλοτόμος”, or gum-cutter, is a worm, which, according to the belief of many peoples, causes toothache. Although teething itself could hardly be attributed to a worm, the incidental aches could be referred to that agency, i.e. the absence of a worm would result in easy teething. This explanation would be more certain, if we accept the correction “οὐλοτόμοιο”, but it may still hold good with the retention of “ὑλοτόμοιο” (a general word for a worm), as suggested above.

For the worm as the cause of toothache cf. Shakespeare Much Ado iii. 2. 28; the belief is very common, e.g. in Scotland, County Folk-Lore iii. (Orkney), p. 140; India, Crooke Popular Religion and Folk-Lore of N. India i. p. 151 (where women of the gipsy tribes know charms to extract the worm); Finland, Abercromby Pre- and Proto-historic Finns i. p. 328. Dyer Folklore of Shakespeare p. 273 f. gives parallels from Germany and China. In the Geopon. xii. 27 and 35 the same remedies are assigned to worms and toothache.

[231] The story of Demeter nursing Demophon has a parallel in Paus.ii. 5. 5: the children of Plemnaeus, a legendary King of Aegialea, in Sicyon, died at birth, until Demeter took pity and under the guise of a strange woman reared up a child named Orthopolis. On the close connexion between the growth of children and vegetation see the interesting chapter in Mannhardt Myth. Forsch. p. 351 f. “Kind und Korn.” For Demeter as a goddess of healing see Rubensohn in Ath. Mitth. xx. p. 360 f. In the hymn, Demophon is in no present danger; Demeter only promises to keep him in good health. According to Nicand. Ther.485 a lizard (“ἀσκαλαβώτης”) had wounded Metanira's child; in Ovid Fast.iv. 446 f. the child (Triptolemus) is dying.

θυώδεϊ δέξατο κόλπῳ=Il. 6.483 (“κηώδεϊ”), of Andromache; hence “θυώδης” does not refer to the divinity of Demeter, who sheds a superhuman fragrance only when she appears as a goddess (see on 277).

[232] χερσίν τ̓: the “τε”, to which many editors object, seems genuine. Demeter receives the child in her bosom and her arms (not “places the child with her hands in her bosom”).

[234] Δημοφόωνθ̓: Apollod.i. 5. 1 follows this version of the story. He mentions however Triptolemus as the elder son of Celeus, and relates the gift of the winged chariot. Demophon was finally ousted altogether by the greater fame of Triptolemus.

[236] The abruptness of the text is impossible, and Hermann's supplement is recommended by the homoeoteleuton.

[237] For the story cf. Apollod.i. 5. 1βουλομένη δὲ αὐτὸ ἀθάνατον ποιῆσαι, τὰς νύκτας εἰς πῦρ κατετίθει τὸ βρέφος καὶ περιῄρει τὰς θνητὰς σάρκας αὐτοῦ”.

inque foco pueri corpus vivente favilla
obruit, humanum purget ut ignis onus.

Similarly Thetis wished to make Achilles immortal, but was prevented by Peleus: cf. Apollod.iii. 13. 6 and

῾α πασσαγε ωηιξη, ας ρυηνκεν ποιντεδ ουτ, μαψ βε δεριϝεδ φρομ τηε ηψμν̓ μὲν γὰρ βροτέας αἰεὶ περὶ σάρκας ἔδαιεν
νύκτα διὰ μέσσην φλογμῷ πυρός: ἤματα δ᾽ αὖτξ
ἀμβροσίῃ χρίεσκε τέρεν δέμας, ὄφρα, πέλοιτο
ἀθάνατος καί οἱ στυγερὸν χροΐ γῆρας ἀλάλκοι.

[238] καταπΝείουσα: cf.

iungere dignata est os puerile suo.
pallor abit, subitasque vident in corpore vires.
tantus caelesti venit ab ore vigor.

[239] κρύπτεσκε: so Apollod.iii. 13. 6κρύφα Πηλέως εἰς τὸ πῦρ ἐγκρυβοῦσα”. For the purifying effect of fire on human beings cf. Rohde Psyche p. 29, Mannhardt A. W. F. p. 52f., Frazer G. B. iii. p. 312, who says “to the primitive mind fire is the most powerful of all purificatory agents.” He compares the custom of modern Greek women who leap over the midsummer bonfire, crying “I leave my sins behind me.” The myth of Demophon suggests, if it does not prove, that the Eleusinian children were purified by passing over fire (Jevons p. 365, Introd. p. 10). For such customs in the case of children see Frazer G. B. iii. p. 239 f. Modern Greeks still believe that newborn babies are protected from evil by the presence of fire; see Rodd Customs and Lore of Modern Greece p. 107 f. For the cognate idea of carrying fire over the field see on 48.

Ἠΰτε δαλόν: this may mean “she hid him in fire as a brand is kept alight” (in the ashes); for which see Od. 5.488 f. and n. on h. Herm. 234. More probably, however, we should understand “she wrapt him in flames like a lighted torch.”

[240] λάθρα^ occurs only in a doubtful fragment of Euripides (1117 v. 28 Dind.); it is corrected in Hel.835(“λάθρ᾽ οὐδαμοῦ”). “ἑῇ” for “φίλῃ” was read by Zenodotus in Il. 3.244, but the alteration seems too violent here; much more so “κρύβδα φίλων”.

[241] Προθαλ́ης, “early - growing,” only here; for the form cf. “ἀμφιθαλής Χ” 496, and “εὐθαλής” common in poetry after Homer.

The last hemistich = Il. 24.630 with “γάρ” (for “δέ”) which Voss wrongly restores here. The sense requires “δέ”, and the hiatus in the bucolic diaeresis is legitimate.

[242] ἀΓ´ηρων: see on h. Aphr. 214.

[244] ἐΠιτηρ́ησασα: she watched to see how the nurse made the child thrive, and thus broke the taboo. The magic could only be worked in secrecy, although the writer implies rather than expresses this (258 f.). In fact it is doubtful whether he understood the real nature of the taboo in the myth; he lays stress only upon Demeter's anger (251, 254), as if she renounced her design of her own will. In the Achilles legend, Apollodorus (l.c.) is more explicit: “Θέτις κωλυθεῖσα τὴν προαίρεσιν τελειῶσαι”. Apollonius vaguely states that Thetis left Peleus, as soon as she heard him cry, and rushed into the sea, “χωσαμένη” (Il. 4.877); the schol. on Nub. 1068 similarly says “ δὲ λιτηθεῖσα ἐχωρίσθη”. Curiosity in seeing a forbidden sight is punished in the classical myth of Cupid and Psyche; for other examples of this world-wide motive see Hartland Science of Fairy Tales pp. 270 f.

[245] κώκυσεν: the language of Apoll. Arg. 4.872 is similar: “ἧκε δ᾽ ἀϋτὴν
σμερδαλέην ἐσιδὼν μέγα νήπιος

ἄμφω Πλ́ηξατο μηρώ: cf. Il. 12.162 13. 198.

[246] ἀάσθη: for the quantity of the first vowel cf. Od. 13.68 α?ασαν, Λ 340 α?άσατο δὲ μέγα θυμῷ”. In 258 the “α” is short, for which cf. Il. 16.685, Τ” 113, 136, h. Aphr. 253.

[248] The trochaic caesura in the fourth foot is not uncommon, when the caesura is preceded by a monosyllable (“μέν, δέ, γε”, etc). Instances like that in 17 (where see note) are different. For the quantity of the “ι” in “πυρί” see on 99. No emendation is necessary.

[252] ἔτικτε: the omission of the mother's name is awkward, as Demeter is the subject of the main sentence; but there is no real difficulty, especially as “τῆς” and “τῇ” immediately precede.

[253] ἀΠὸ ἕο θ̂ηκε: cf. Il. 12.205 ἀπὸ ἕθεν ἧκε, ι 461 ἀπὸ ἕο πέμπε”. Here Cobet reads “ἀπὸ ἕθεν ἧκε”, which Gemoll approves, as “θῆκε” with “-δε” is remarkable; it may be added that a verb expressing violent action would seem more appropriate to Demeter's anger: cf. Apoll. Arg. 4.674τὸν μὲν ἄρ᾽ ἁρπάγδην χαμάδις βάλε κεκληγῶτα”.

[254] ἐΞανελοῦσα Πυρός: Apollodorus, seemingly following a different tradition, says “τὸ μὲν βρέφος ὑπὸ τοῦ πυρὸς ἀνηλώθη”. In Ovid's account, the mother takes the child from the fire.

κοτέσασα: in Homer “κοτεσσαμένη”.

[256] The editors compare Orph. fr. xxxii.μηδαμὰ μηδὲν
εἰδότες, οὔτε κακοῖο προσερχομένοιο νοῆσαι
φράδμονες, οὔτ᾽ ἄποθεν μάλ᾽ ἀποστρέψαι κακότητος
οὔτ᾽ ἀγαθοῦ παρεόντος ἐπιστρέψαι τε καὶ ἔρξαι
ἴδριες, ἀλλὰ μάτην ἀδαήμονες ἀπρονόητοι

”. The resemblance can hardly be accidental, but it by no means follows that the Orphic poet read “φράδμονες” here, as Bücheler infers (so Tyrrell). For the quantity of the first syllable in “ἀφράδμονες” cf. Od. 15.444 ἐπιφράσσετ᾽ ὄλεθρον”: Hes. Op.655προπεφραδμένα”: Theog. 160 “ἐπεφράσσατο”: h. Apoll. 388ἐφράζετο”, and regularly “Ἀφροδίτη”. La Roche Hom. Unters. i. p. 10, H. G. § 370.

[258] Ν´ηκεστον: the reading of M “μήκιστον” might possibly be defended as a superlative of “μέγ᾽ ἀάσθης”: cf. also Eur. Hipp.818τὰ μάκιστ᾽ ἐμῶν κακῶν”. But Voss's correction, based on Hes. Op.283, is easy and highly probable, if not certain.

[259] “ἴστω Γάρ κτλ.”: cf. Il. 15.36-38, ε” 184-186 (with M. and R.'s note); Leaf on Il. 2.755. On the position of “ὅρκος” (the object of the oath) before “Στυγὸς ὕδωρ” cf. Apoll. Arg. 3.714 f.

[262] θάνατον . . ἁλύξαι=Il. 21.565, Od. 17.547, χ” 66. Hence Huschke's “γῆρας” should not be received, although Apollonius has “γῆρας ἀλάλκοι” (see on 237).

265-267. The text is certainly sound (with the sole correction of “συναυξήσους᾿” to “συνάξους᾿”, for which cf. Il. 2.381, Ξ” 149, 448): “when Demophon is a man, the Eleusinians will always be fighting with one another.” Editors have assumed a lacuna before 265 and after 267, or at all events after the lines. It was supposed that the lost passage or passages referred to the death of Demophon, or to his leadership in the war, or mediation between the parties. This supposition is quite gratuitous; 265 simply marks the time, “when he has grown to manhood,” and has no closer connexion with the preceding or succeeding lines. There is no trace in myth or history of an Eleusinian civil war; hence Matthiae (followed by Baumeister) substituted “Ἀθηναίοισι” for “ἐν ἀλλήλοισι”, assuming that Demophon was the leader of the Eleusinians in their war against Athens. The corruption is most improbable, not to mention the further difficulty that tradition made Eumolpus, not Demophon, the leader of the Eleusinians ( Thuc.ii. 15, Isocr. Paneg. 19, Apollod.iii. 15. 4, in Leocr. 24, Paus.i. 38. 3). There are so few allusions to early Eleusinian history in Greek literature, that it would not be surprising if mention of a civil war were found in this passage only. But Creutzer was no doubt right in explaining the lines by reference to the “βαλλητύς”, or sham fight, which is expressly connected with Demophon by Hesychius s.v., “ἑορτὴ Ἀθήνησιν ἐπὶ Δημοφῶντι τῷ Κελεοῦ ἀγομένη”. Lobeck (Aglaoph. p. 206) quotes an anonymous verse in Artemid.i. 8ταύροις ἐν Ἰωνίᾳ παῖδες Ἐφεσίων ἀγωνίζονται καὶ ἐν Ἀττικῇ παρὰ ταῖς θεαῖς ἐν Ἐλευσῖνι: κοῦροι <*>αθηναῖοι περιτελλομένων ἐνιαυτῶν”; but it is not clear whether this line has any connexion with the “βαλλητύς”. According to A. Mommsen and Lenormant the “βαλλητύς” took place at the end of the festival. It may, however, have been a ceremony during the initial stage of purification (see Introd. p. 10). The rite was like that at Troezen ( Paus.ii. 32. 2, called “λιθοβόλια”). See Gruppe G. Myth. p. 901. Similar customs are quoted by Bather in J. H. S. xiv. 253, Jevons p. 292. It need not be supposed that the origin of such “λιθοβόλια” was always the same; in the present case the mystae may have stoned one another to draw blood as a means of communion with the Corn-goddess, or the blood may have been thought to increase the fertility of the land. The latter idea is probably at the root of some, if not all, of the numerous parallel examples which shew that fights, either sham or more serious, have taken place to ensure a good harvest. This, as a European custom, was first clearly demonstrated by Mannhardt B. K. p. 548 f.; for instances from savage tribes see Frazer on Paus.ii. 30. 4.As often, the meaning of the rite was lost at Eleusis, where the mock-battle was supposed to commemorate an early civil war.

[265] ὥρῃσιν: the editors (mostly adopting Fontein's “τοῦ γε”), understand this as “in his riper years.” But “τῷ γε” is to be retained and “ὥρῃσιν” taken in the proper sense of the plural, “when the years revolve for him in their seasons.” Cf. h. Aphr. 102ὥρῃσιν πάσῃσι”, infra 399, h. Apoll. 350.

[267] αἰὲν . . Ἤματα Πάντα: Baumeister, understanding the reference to be to an actual war, is obliged to explain this as an epic formula vaguely indicating a “long time.” But it has its regular meaning “for ever”; the “βαλλητύς” takes place every year.

[268] τιμάοχος: only here and in h. Aphr. 31, which Gemoll claims to be the original passage.

[269] ἀθανάτοις is made necessary by similar formulas: e.g. 11, 21, 45, 403; hence Stoll's “ἀθανάτων” must be rejected. There remains the difficulty of “ὄνειαρ”, which can scarcely be a disyllable with synizesis; in Hes. Op.462 the MSS. have “εἴαρι πολεῖν”, but Pollux (i. 223) rightly gives “ἔαρι” (“ε_α_”). The synizesis of “ηε” is no authority for that of “εια” (see on 137). It seems best therefore to remove the diphthong, with Ilgen, and read “ὄνεαρ”, the form accepted by Schulze Quaest. Ep. p. 228 and Solmsen K. Z. 32, 292, who calls it “sprachlich tadellos.” This could be a trisyllable by the correction of “θνητοῖσιν” to “θνητοῖς τ̓”; but it is nearer to the manuscript to read “θνητοῖσί τ᾽ ὄνεαρ”. For the synizesis compare (besides Hes. Op.462 quoted above) Hes. Op.492μήτ᾽ ἔαρ γιγνόμενον”, Mimnerm. 2 and Chaerem. fr. 42 (“ἔαρος” a trochee). If “ὄνειαρ” is to be retained, with its full value u--, it must contain the whole of the fourth foot; this involves the lengthening of the last syllable by position, as is done by the conjectures of Ruhnken and others. The legitimacy of this use was the subject of a discussion in the Class. Rev. Dec. 1896, Feb.-Apr. 1897. The result was entirely to justify the use in Homer and Hesiod, although undisputed examples are not common in early epic, and very rare in later hexameters. For the most recent discussion on the subject see Leaf Il.vol. ii. App. p. 634 f.

[270] There is no proof that there was a “temple” of Demeter at Eleusis, apart from the hall of initiation, which cannot properly be called a “νηός”. Strabo, it is true, speaks of a “ἱερόν” as well as the “μυστικὸς σηκός” (ix. p. 395), but the word “ἱερόν” need not imply a building; it may=“τέμενος”. As Frazer remarks (on Paus.i. 38. 6, p. 511) “no later writer” (than the hymn) “and no inscription yet discovered speaks of such a temple.” Various attempts have been made to identify this supposed temple with some of the pre-Persian remains discovered by the excavations of the Greek Archaeological Society. Frazer (l.c. p. 509) doubtfully suggests that it may have been on the site of the later hall of initiation, where walls of Eleusinian marble have been unearthed. Remains of another early building, probably a temple, have been discovered north of the hall, and separated from it by a rock-cut staircase, leading up to the terrace. This building has also been thought to be the old temple of Demeter. It is possible that the “νηός” served also as a hall of initiation, which would of course be sacred to Demeter. In this case the building may be identified with the walls abovementioned, which belong to a building older than the age of Pisistratus; but it is impossible to judge of the form of this building from these scanty remains, or to conjecture how far it was a prototype of a later hall (probably built by Pisistratus), and of the enlarged Periclean hall. See Philios p. 65, 74, who also identifies the “νηός” with the primitive “τελεστήριον”; Svoronos (p. 345 f.) places the “νηός” on the brow of the hill, but this seems negatived by “ὑπαὶ πόλιν”.

[271] Πόλιν αἰΠύ τε τεῖχος: i.e. the acropolis, the fortifications of which (“τεῖχος”) have been traced on the low hill above the hall of initiation. The actual town lay at the foot of the hill, and extended to the sea.

[272] Καλλιχόρου: see on 99; this well was not identified until 1892, when excavations shewed it to be situated by the great Roman propylaea, just outside the precinct. The well-mouth is surrounded by concentric circles, which no doubt served as marks for the Eleusinian woman who danced round the water in honour of the goddess ( Paus.i. 38. 6). For references to the discovery see Philios p. 57 f., and Svoronos p. 252.

[274] εὐαγέως: the adv. in Apoll. Arg. 2.699, etc. “εὐαγής” is not found in early epic. For exx. of “εὐαγής, εὐαγέως” in ritual see Dieterich de hymnis Orph. 1891, p. 34. ἱλάσκοισθε: for the opt. after “ὑποθήσομαι” cf. Od. 17.250, H. G. § 306. The mood expresses a less certain result than would be indicated by “ἱλάσκησθε”, which Schäfer reads.

[275] μέγεθος καὶ ει<*>῀δος=h. Aphr. 82.

[276] Περί τ᾽ ἀμφί τε: cf. Il. 2.305 ἀμφὶ περὶ κρήνην”, h. Apoll. 271ἀμφιπεριφθινύθει”, Theocr. vii. 142περὶ πίδακας ἀμφὶ μέλισσαι”.

κάλλος ἄητο: modelled on Hes. Sc.7 f. “τῆς καὶ ἀπὸ κρῆθεν . . . τοῖον ἄηθ᾽ οἷόν τε πολυχρύσου Ἀφροδίτης”.

[277] “ὀδμ́η κτλ.” Fragrance is a sign of divinity: cf. Theognis 9, P. V. 115, Eur. Hipp.1391, Verg. Aen.i. 403, Ov. Fast.v. 375.

[278] φέγΓος: see on 189. With this passage cf. Bacchyl.xvii. 102ἀπὸ γὰρ ἀγλα

ῶν λάμπε γυίων σέλας
ὥστε πυρός” (of the Nereids).

[279] κατεν́ηνοθεν: properly sing. Hence Ruhnken and others read “ξανθὴ δὲ κόμη”. But, as Franke well remarks, the writer may easily have taken the archaic form for a plural. There is no reason to suppose a genuine schema Pindaricum, with Baumeister.

[280] αὐΓ῀ης: for Ruhnken's simple correction cf. Soph. Phil.1190αὐταῖς” MSS., while the scholia preserve “αὐγαῖς”.

[281] Γούνατ᾽ ἔλυντο: Il. 7.16 λύντο δὲ γυῖα”: Il. 13.85 γυῖα λέλυντο”, and often “λύτο γούνατα”.

[283] ἀΠὸ δαπέδου: Hesych. ii. 253 quotes the parallel form “ζάπεδον”, which occurs in Xenophanes i. 1, and an inscr. from Paros (I. G. A. 401 = Roberts Epigr. 17); “δάπεδον” therefore stands for the original “δjάπεδον” (or for “δϝάπεδον” Prellwitz Et. Wört. s.v. “δα-”) and the metre is not due to false analogy (as Gemoll supposes), but was, at least originally, justified by pronunciation. In Od. 11.598 Aristotle Rhet.iii. 11 read “ἐπὶ δάπεδόνδε” for the vulgate “ἔπειτα πεδόνδε”. La Roche Hom. Unters. i. p. 49.

[284] ἐλειν́ην: the Attic form is accepted by most editors after Ruhnken; it does not occur elsewhere in epic. Rutherford (New Phryn. p. 160) rejects “ἐλεεινός” in Attic prose; the form is due to late usage.

[285] εὐστρώτων: only here and in h. Aphr. 157ἐς λέχος εὔστρωτον”.

[289] ἐλούεον: called an “impossible” form by Gemoll. It is a false formation, but is not to be ejected on that account. Cf. Schulze Quaest. Ep. p. 65 n. 1, Smyth Ionic Dialect p. 535, Solmsen l.c. p. 13, K. Z. 29, 98. Ludwich needlessly objects to the washing of the child. The women perform one of the duties of a nurse, in place of Demeter. It is perhaps unnecessary to press the phrase further, and to point out that the child would be covered with wood-ash. This motive, however, is expressly mentioned in a very similar passage (of the Nymphs and Bacchus), Anth. Pal. ix. 331 “αι<*> Νύμφαι τὸν Βάκχον, ὅτ̓ ἐκ πυρὸς ἥλατο κοῦρος”,

νίψαν ὑπὲρ τέφρης ἄρτι κυλιόμενον”.

[291] τροφοί and τιθ̂ηναι, “those who cared for and nursed him,” are here synonymous. Cf. “τιθηνοίμην” (142) used by Demeter in her disguise as a “τροφός” (103). Tyrrell's suggestion “ἠὲ τιθήνη” is no improvement. Cf. Orph. h. x. 18 “τροφὸς ἠδὲ τιθήνη”.

[292] ΠανΝύχιαι: the origin of the “παννυχίς” is almost certainly indicated in this word (Preller). Gemoll notes that the ignorance of Celeus as to what has happened until the morning points to a mystery. Most of the sacred ceremonies during the whole course of the Eleusinia were carried on at night. In the very earliest period the worship of Demeter Thesmophoros at Eleusis, as elsewhere, was probably confined to women (Foucart p. 78, Jevons p. 379, Ramsay p. 127); and the hymn clearly shews the important part played by the women, even in a later stage of the Eleusinian religion. For women as mainly or exclusively concerned in agriculture see Jevons p. 239-242. Even when a share in agriculture falls to the lot of the men, the place of women in festivals concerned with sowing, reaping, etc. is often predominant; for examples see Frazer G. B. i. p. 35, ii. p. 203, etc.

[293] δείματι Παλλόμεναι: the same phrase in an oracle ap. Herod.vii. 140(Hendess 111. 10).

[296] Πολυπείρονα, “countless:” literally “with many boundaries,” formed on the analogy of “ἀπείρων”. Cf. Orph. Arg. 33 “πολυπείρονας οἴμους”.

[301] Matthiae thinks that the rest of the hymn, from this line, was put together from fragments of the hymn seen by Pausanias, but the vv.ll. in Paus. only point to natural and quasiclerical errors, see Preface p. xli.

[302] Ξανθ̀η Δημ́ητηρ = Il. 5.500. The epithet may have originally referred to the colour of ripe corn, as the “hair” of Demeter (cf. 454 “κομήσειν ἀσταχύεσσιν”, Euseb. P. E. v. 34 “οἱ δὲ ἐκόμων Δήμητρι”), although, of course, in the hymn Demeter is purely anthropomorphic; see Mannhardt Myth. Forsch. p. 234.

[305] ἐΠὶ χθόνα: for the accusative see on xxv. 3. The worship of Demeter and Cora in Triphylia was thought to be explained by the alternation of good and bad years (“τάχα διὰ τὰς ὑπεναντιότητας”) according to Demetrius of Scepsis ap. Strab. 344 “καὶ γὰρ εὔκαρπός ἐστι καὶ ἐρυσίβην γεννᾷ καὶ θρύον Τριφυλία: διόπερ ἀντὶ μεγάλης φορᾶς πυκνὰς ἀφορίας γίνεσθαι συμβαίνει κατὰ τοὺς τόπους”.

[308] ἀρούραις: for the local dat. (like “οὔρεσι” etc.) cf. Il. 5.137 ἀγρῷ”, and see on 99. Here the dat. is used with a verb of motion; H. G. § 145 (6). There is a different const. in Il. 10.353 ἑλκέμεναι νειοῖο βαθείης πηκτὸν ἄροτρον”.

[310] Cf. Hes. Op.180Ζεὺς δ᾽ ὀλέσει καὶ τοῦτο γένος μερόπων ἀνθρώπων”.

[312] θυσιῶν: so 368. The word is not Homeric (for “θυέων” which Hermann gratuitously read).

[314] Ἶριν . . . χρυσόπτερον=Il. 8.398; see on h. Apoll. 107. Iris is here employed as a messenger to gods on earth, while Hermes is sent to the underworld (335). Cf. Maass “ἾριςI. F. i. 157 sq.

[315] Πολύηρατον . . . ἔχουσαν= Theog. 908 (“ἔχουσα”).

[316] ὣς ἔφαθ̓: the use of this formula after an indirect speech is not Homeric, but occurs in Hes. Op.69, infra 448, Apoll. Arg.4.236, 1119. Wyttenbach's lacuna is not needed; cf. on 127.

[317] Cf. h. Apoll. 108.

[319] κυανόπεπλον: not in Homer, and in the hymns only here, and in 360, 374, 442 of Demeter. Theog. 406 it is a general epithet of Leto, with no special reference to mourning, as in this hymn (cf. 183).

[321] ἄφθιτα εἰδώς: only here, for “ἄφθιτα μήδεα εἰδώςh. Aphr. 43, where see note.

[325] Valckenär's addition of “πατήρ” is preferable to the other suggestions, as it retains “θεούςin synizesi, which is probably the cause of its omission, unless this is simply due to “haplography” in “-τα, πα-”.

[328] Hermann's “ἕλοιτο μετ᾽ ἀθανάτοισι θεοῖσι” (on the analogy of 444) does not account for “ἐλέσθαι” in place of “θεοῖσιν”. The suggestion “βόλοιτο” rests on Il. 11.319, where one family of MSS. (e) has “δὴ ἐθέλει” for “δὴ βόλεται. ἐθέλοι βόλοιτο” might produce “ἐθέλοιτο” which otherwise it is difficult to explain.

[331] θυώδεος: applied to Olympus in h. Herm. 322. It appears to be a favourite word with the writer of this hymn: cf. 231, 244, 288, 355, 385. The meaning here may be literally “fragrant with incense” (which ascends to heaven), or perhaps simply “sweetsmelling” as in 231. See further on h. Herm. 231.

[337] ἁΓν́ην: specially an epithet of Persephone: Od. 11.386, infra 439. She was worshipped as “Ἁγνή” in Messenia, Paus.iv. 33. 4; cf. the inscr. of Andania. “ἁγνή” is also frequent with Demeter, Hes. Op.465, supra 203, Archil. 120.So “ἁγναὶ θεαί” of both goddesses C. I. G. 5431, 5643. Rohde Psyche p. 192, Roscher i. p. 1813 f., Pauly-Wissowa 2754.

ἀΠό may here be retained, though Voss pointed out that in the Homeric formula the prep. is “ὑπό”: cf. Il. 21.56, Theog. 653.

[339] μεταλ́ηξειε: the spelling is philologically correct, as “λήγω” makes position in Il. 9.191, θ” 87. According to Didymus in schol. A, Aristarchus read the single liquid in the Homeric passages Il. 9.157, 261, 299. On the other hand it should be noted that M constantly neglects a double consonant; cf. 14, 40, 158, 313 in this hymn.

344, 345. Baumeister's despair at this passage still holds good. “ἐπ᾽ ἀτλήτων” might possibly be construed “in such intolerable circumstances” if the neglected position “ἀτλήτων” is permissible (there is no instance in Homer except “σχετλίη”, La Roche Homer. Unters. i. p. 4 and 16; but cf. Pind. Ol.viii. 20 and 77, Emped. 14). But it is hard to believe that epic, or any Greek usage admits of the translation. Of the conjectures, “ἀποτηλοῦ” is the best; if written “ἀποτληου”, the resulting word is not worse than M's other corruptions; e.g. “ἐπηλσίησι” for “ἐπηλυσίη” 228.

In the next line a word of the quantity uu- has perhaps fallen out, owing to “θεῶνin synizesi (cf. 325); this may have been “χαλεπήν” (with “βουλήν”), the dative “βουλῇ” having been written afterwards to ease the construction. “μητίσετο” is not a Homeric form for “μητίσατο”, which should probably be restored; cf. h. Apoll. 322, 325 a.

[348] M's reading “σε” is just possible, as “ἄγειν, ἐξάγειν” could mean “let go,” “turn out.” But the parallel passage 335 f. makes “με” practically certain.

[349] Ἐρέβευσφι: Franke's correction is easy (cf. Il. 9.572, Theog. 669, where some MSS. have “ἐρέβεσφι”), but perhaps unnecessary, if the peculiarities of our tradition of the hymn (“κατενήνοθεν” with plur. 278, “παύσειεν” neut. 351) are to be preserved. So the form “εἱστήκει” 452 is defensible.

[351] Παύσειεν is no doubt genuine although the act. for the middle “παύσαιτο” is remarkable. Compare, however, Hes. Sc.449παῦε μάχης”. In Od. 4.659 there is overwhelming MS. support for “μνηστῆρες . . . παῦσαν ἀέθλων”, where most editors read “μνηστῆρας”. So Ran. 580 “παῦε παῦε τοῦ λόγου”. Tyrrell considers the use to be a mark of lateness (p. 39).

[352] χαμαιγενέων ἀΝθρώπων = h. Aphr. 108 (where see note).

[357] μείδησεν: Hades “smiled,” anticipating the success of his plan to keep Persephone (372 f.). ὀφρύσιν: generally with “ὑπό” or “ἐπί” in Homer (“νεύειν” etc.), but cf. Od. 9.468, μ” 194. So without a prep. Pind. Pyth.ix. 65.

[362] The line is quite genuine, in spite of Bücheler's objection (imperite corrasa verba). Gemoll thinks that “μοι” is required, but the sense is quite clear without it. The object of Persephone's anger is plain from 344 “πόλλ᾽ ἀεκαζομένῃ”. Hades carefully avoids saying “come back” (as Gemoll thinks he ought to say); Persephone will find out in due time the necessity of returning. She has not yet eaten the pomegranate, and he therefore uses the ambiguous futures “ἔσσομαι” etc., which suit equally well the choice or the necessity of returning.

Περιώσιον ἄλλων = Pind. Isthm.iv. 3.

[365] δεσπόσσεις: not in early epic. Like “ἀδικεῖν” (367) it is chiefly Attic, but also found in Herodotus. The word may be suggested, as Baumeister notes, by the title “Δέσποινα”, under which Persephone was worshipped at many places, especially in Arcadia; Paus.viii. 37. 9, Immerwahr die Kulte u. Myth. Ark. i. p. 120.

[366] σχ́ησησθα: since there is no instance of the termination “-σθα” or “-θα” in a future, while the aorists “βάλησθα, πάθησθα, εἴπησθα” are Homeric (KühnerBlass ii. § 209. 3), it seems better to keep the spelling of M and regard “σχήσησθα” as the subjunctive of the otherwise late aorist “ἔσχησα”. The subjunctive will be of the nature of the type “δύσομαι εἰς Ἀίδαο καὶ ἐν νεκύεσσι φαείνω” (H. G. § 275 f.), which in Homer occurs constantly in combination with futures and is practically indistinguishable from them in meaning; see h. Apoll. 1. “σχήσεισθα” which most recent editors prefer is called a “verbildete Form” by Schulze K. Z. 33. 317.

[367] τῶν δ᾽ ἀδικησάντων: “those who have wronged thee” (by not paying due honour) will be punished all their days (i.e. by the Furies, for whose relation to Hades and Persephone see Il. 9.454 with Leaf's note and 571). There is no allusion to punishment after death, although the fate of the uninitiated is not happy in the underworld (cf. 481 f.); line 365 shews that the reference is here to the living.

[368] θυσίαισι: the Attic form (for “θυσίῃσι”) may well be original in this hymn.

[371] αὐτός, (Hades) “himself,” in contrast to Persephone; or possibly “with his own hands.”

[372] ῥοῖης κόκκον ἔδωκε: Apollodorus (i. 5. 3) follows: “ῥοιᾶς ἔδωκεν αὐτῇ φαγεῖν κόκκον”. In Ov. Met.v. 535 f. Persephone of her own accord picks the fruit in a garden, and eats seven seeds. There is a widespread belief that the living may visit the underworld and return safely, provided that they abstain from the food of the dead. The Finnish hero Wäinämöinen refuses to drink in Manala, the place of the dead (Kalevala xvi. p. 293). In Africa S. there is a similar story: a man visits spiritland and is warned to return before he meets one who will give him food (Leslie Among the Zulus and Amatongas p. 121). In New Zealand a Maori woman was thought to have come back from the dead, having by the advice of her father refused the food which the dead people offered her (Shortland Traditions of New Zealand p. 150). The last story is quoted by Tylor Prim. Cult. ii. p. 51, who gives a parallel among the Sioux of N. America. Several similar tales are collected by Hartland Science of Fairy Tales, ch. iii. (among the ancient Danes, in the Banks islands, and in the Hervey islands). Hartland remarks that there is the same objection to eating the food of the fairies (cf. Rhys Celtic Folklore i. p. 290; see also Folk-Lore viii. p. 380; County Folk-Lore iii. (Orkney and Shetland), p. 25, 27). Some other references are given by Frazer on Paus.viii. 37. 7; cf. also Folk-Lore x. p. 300 f. (Japan). The basis of the belief is the idea that a common meal unites the partakers in a close bond; hence the sanctity of the relation between host and guest in primitive society. By eating any food in the underworld, Persephone established a bond with the dead. But there is no doubt a special significance in the particular food—a pomegranate—although its precise meaning has been disputed. According to one view, the fruit, from the blood-red colour of the inside, is a symbol of blood and death. A pomegranate tree was planted over the graves of Menoeceus, a suicide ( Paus.ix. 25. 1), and the unlucky Eteocles (in the latter case by the Erinyes, Imag. ii. 29, i. 4). It was believed to have sprung from the blood of Dionysus Zagreus ( Protrept. ii. 19). The fruit was therefore appropriate to the dead. Probably, however, it is here rather symbolical of marriage and fertility, from the multitude of its seeds; cf. Herod.iv. 143ὅσοι ἐν τῇ ῥοιῇ κόκκοι”. It was the emblem of Hera, probably as goddess of marriage; the fruit expedited birth, N. H. xxiii. 107; cf. ib. 112 (of its flowers) sistunt potu menses feminarum. It was an attribute of Aphrodite (see Murr die Pflanzenwelt in d. Gr. Myth. p. 50 f., Roscher Lex. 2090, Preller-Robert i.^{2} p. 763). Pausanias (ii. 17. 4) refuses to discuss the meaning of the fruit in the hand of the Argive Hera. The mystae at Eleusis abstained from eating it (Porphyr. de Abstin. iv. 16) as did the Thesmophoriazusae ( l.c.), and the banqueters at the Haloa (schol. Lucian dial. meretr. vii. 4; see Harrison Proleg. p. 148). The Arcadians would not bring the pomegranate into the temple of Despoina ( Paus.viii. 37. 7). According to this view, the pomegranate would symbolise, not so much Persephone's general union with the dead, as her special union with Hades. In actual custom, the Greeks made wedding-cakes of sesame (“διὰ τὸ πολύγονον, ὥς φησι Μένανδρος” schol. Pax869).

For the pomegranate as an attribute of Persephone and Pluto in art see Preller-Robert i.^{2} p. 763 n. 2, Bötticher Baumkultus ch. 38.

It does not appear, however, that the writer of the hymn attached any particular meaning to the pomegranate (unless, like Pausanias, he was afraid to divulge a mystery). Apollodorus does not offer any explanation, while Ovid ( Met.v. 532) simply says sic Parcarum foedere cautum est.

[373] ἀμφὶ Νωμ́ησας: the sense is obscure, owing to the peculiar use of “νωμᾶν”. The meanings of the verb fall mainly under two heads (1) “distribute,” of food etc., (2) “wield” or “handle” (“α”) weapons etc., (“β”) of the mind, “turn over.” Hermann first read “ἀμφὶς νωμήσας” (after Santen) translating seorsum tribuens, i.e. apart from Hermes. Gemoll follows this view. Hermann afterwards retained “ἀμφί” (with “” for “” after Ruhnken) and understood “dividing it into two parts” (one of which he himself ate). Either “ἀμφίς” or “ἀμφί ” might bear this sense, but the participation by Hades in the food is not mentioned elsewhere in this or any other version of the myth. Nor is such participation required according to folklore; the living have only to eat the food offered by the dead, not share it with them, to prevent their return. Voss's explanation dum eam prope se traheret, is quite impossible; nor can we assume tmesis, “embracing her,” a sense which “ἀμφινωμᾶν” could not bear, although it might be used of a nurse “handling” a baby.

The most probable view is to take “νωμᾶν” figuratively. Ilgen translated “turning it (“”) over in his mind,” but a far better sense is given by retaining “” (as accented in M), and translating after Matthiae “peering round him,” = “παπτήνας”, cf. Il. 4.497 ἀμφὶ παπτήνας” (cf. also Il. 15.241 ἀμφὶ γινώσκων ἑτάρους”). For this sense of “νωμᾶν” cf. Herod.iv. 128νωμέοντες . . σῖτα ἀναιρεομένους” “observing them foraging.” Plat. Crat. 411Dτὸ νωμᾶν καὶ τὸ σκοπεῖν ταὐτόν”. Eur. Phoen.1255μάντεις δὲ μῆλ᾽ ἔσφαζον”, “ἐμπύρους τ᾽ ἀκμάς

ῥήξεις τ᾽ ἐνώμων” where the scholiast paraphrases “ἐπεσκόπουν” and “παρετήρουν”; perhaps “προσενώμαPhiloct. 716, and in an intermediate construction Eur. Phoen.1563τάδε σώματαὄμματος αὐγαῖς σαῖς ἐπενώμας”; schol. “ἀντὶ τοῦ διεσκόπεις”. Hades cast glances about him to see whether his action is seen by any one, especially Hermes, who was commissioned by Zeus to restore Persephone to the upper world, and would have thwarted his design. λάθρͅη: the rhythm and the parallel passage 411 (“αὐτὰρ λάθρῃ ἔμβαλέ μοι ῥοιῆς κόκκον”) shew that this word is to be taken with “ἔδωκε”, not “νωμήσας”. Eur. Itmay mean “without the knowledge of Hermes,” “secretly,” or perhaps “treacherously,” i.e. Persephone did not realize the result of eating. For the latter sense of “λάθρῃ” cf. Od. 17.80. See further on 413.

[379] διὲκ μεγάρων: Gemoll rightly notes that the realm of Hades is thought of as a huge house; cf. Il. 3.322 δῦναι δόμον Ἄιδος εἴσω” etc. Otherwise the entrance of horses into the “μέγαρον” would be impossible.

τὼ δ᾽ οὐκ ἄκοντε ΠετέσθηΝ: the common Homeric formula, with “ἀέκοντε”, which, however, is not to be read in the later hymn; cf. 413. With the passage generally cf. Il. 5.364-7.

[381] ὕδωρ has always “υ” short in thesi in early epic; hence Hermann suggested “οὔτ᾽ ἄρ᾽ ὕδωρ”. But Baumeister quotes Batr. 97, Apoll. Arg. 4.290 and other later passages in support of the text.

[382] οὔτ᾽ ἄκριες is remarkable, according to Gemoll, between “ἵππων ἀθανάτων” and “ἔσχεθον ὁρμήν”. He does not note, however, that “ἄκριες” (“ας”) always forms the fourth foot in epic; see Ebeling. The unusual position is moreover justified by the great stress laid on “ἄκριες”, cf. “αὐτάων” “over the very mountains” (383).

[384] στ̂ησε δ᾽ ἄγων: from Il. 2.558.

[386] Ἠΰτε μαινάς: the editors quote Il. 22.460 μαινάδι ἴση” (of Andromache). So Il. 6.389 μαινομένῃ ἐϊκυῖα”; cf. (of Demeter herself) Ov. Fast.iv. 457-8. In the hymn, as no doubt in Homer, “μαινάς” may be simply “a mad woman,” with no reference to the “maenads”; in any case this passage does not imply that there was as yet any Dionysiac influence at Eleusis.

ὕλͅη: Ruhnken's correction of “ὕλης” is in accordance with Homeric usage, which requires the singular; the genitive may have arisen from a mistaken view that “ὄρος ὕλης” could stand for “ὄρος ὑλῆεν”. The MS. reading, however, would be more easily explained if the dat. plur. “ὕλῃς” were original. This form is found in Anacreon fr. 51ὅστ᾽ ἐν ὕλῃς” (so Bergk; “ὕλαις” schol. Pind. Ol.iii. 52, “ὕλῃ” Athenaeus and Aelian). Otherwise the plur. does not seem to occur before Dion. de Thuc. 6; see Zachariae K. Z. xxxiv. p. 453 f. It seems safer to retain the singular.

[392] ΠαυομένΗ: M has “παομ” . ., but the confusion of “α” and “αυ” is common in MSS., e.g. “Νάστης Ναύστης Β 867, Φασιάδην Φαυσιάδην Λ 578, καλόν καυλόν Π 338, ἀγήν αὐγήν” Aratus 668, Herod.ii. 111.Ignarra's excellent correction “συνάξουσ᾽ ου” for “συναυξήσους᾿” (= “συνα”(“υ”)“ξης᾿”) in 267 rests partly on this principle.

[398] The corrections “ἐπάσω” or “τι πάσσαο” no doubt give the sense, but it is rather violent to suppose such a desperate corruption as “πτᾶσα” in M, especially when the scribe had no difficulty with “πάσσατ̓” 50 and “πάσασθαι 413. πτᾶσα” was first defended in the Class. Rev. March 1901, <“σύ γ̓”> being supplied to complete the line. The ellipse of the verb of the second protasis in a double condition is occasionally found: Il. 9.42 εἰ δέ τοι αὐτῷ θυμὸς ἐπέσσυται ὥς τε νέεσθαι, ἔρχεοεἰ δὲ καὶ αὐτοί, φευγόντων κτλ.” (Il. 9.262 εἰ δέ, σὺ μέν μευ ἄκουσον” is only similar in form). In later authors exx. are fairly common: Plat. Euthyd. 285C, Symp. 212 C. So “εἰ δ᾽ οὖνSoph. Ant.722.πτῆναι” is not Homeric, but “ἐξέπτη” occurs in Hes. Op.98, Batr.208 Batr., 211, “πτᾶσα” in “ηεροδ. π. διχρ”. 289. 24. The line thus gains in vividness: “but if so, you will have to fly back”; cf. Od. 11.208. If “ἰοῦσα” following “πτᾶσα” is awkward, it would be possible to read “ἐοῦς᾿”, as in 364Herod., 395.

[399] Ilgen's ὡρέων (for “ὀρέων” M) is nearest to the MS., and preserves an Ionic form and Homeric synizesis; cf. “ἐρέω 406, Κρονίδεω 414. εἰς ἐΝιαυτόν”: the supplement of m can hardly be an invention of the scribe, and the lexx. give instances of the distributive force of “εἰς”, “every year.” See L. and S. s.v. ii. 2.

The division of time is followed by Apollod.i. 5. 3.Περσεφόνη δὲ καθ᾽ ἕκαστον ἐνιαυτὸν τὸ μὲν τρίτον μετὰ Πλούτωνος ἠναγκάσθη μένειν, τὸ δὲ λοιπὸν παρὰ τοῖς θεοῖς”. The third part of the year is of course the winter season, when the corn is below the earth. The editors note the old division of the year into three seasons. According to another account (Ovid Fast.iv. 614, Met. v. 567, Hygin. fab.146) the year is divided into two equal periods of six months each. See Preller-Robert i.^{2} p. 763 n. 3, where it is remarked that Apollo was thought to spend six months in Delos and Lycia respectively, according to Delian tradition, whereas the Delphians believed him to be present among them for nine months.

[401] ὁΠπότε . . θάλλει: the pres. indic. with “ὁππότε” (“as soon as”) is rare; but cf. Od. 18.408 κατακείετε οἴκαδ᾽ ἴοντες”,

ὁππότε θυμὸς ἄνωγε”. The subj. “θάλλῃ” (“whenever”) is read by Voss and Gemoll.

[403] Here the construction is clearly broken, and a lacuna of a line is necessary.

[406] ἐρέω: disyll. in Hes. Op.202; but without synizesis below 416.

[409] ἐλθεῖν after “ἦλθε” (407) has been suspected; but the repetition is not offensive. The infin. depends on “ἦλθεν ἄγγελος”, which implies a command. Bücheler compares Il. 11.715 ἄγγελος ἦλθε . . θωρήσσεσθαι” and Il. 24.194.

[411] The repetition of “αὐτάρ” in one line is hardly possible; probably in the first place it has expelled another particle, which now can hardly be recovered. So Il. 18.203 αὐτὰρ Ἀχιλλεὺς ὦρτο διίφιλος ἀμφὶ δ᾽ Ἀθήνη”, where several MSS. have “αὐτὰρ ἀθήνη”. Ruhnken's “εἶθαρ” and Ilgen's “αὐτίκ̓” are equally near to “αὐτάρ”: the sense might be better given by “ἤτοι”.

[413] ἄκουσαν . . βίͅη . . ΠροσηΝάγκασσε. In 372 (“ἔδωκε φαγεῖν”) nothing is said of the compulsion on which Persephone here insists. Plainly Hades did not use actual force or compulsion of any kind, especially as Hermes was present. Persephone only means that she had no wish to eat, and could not refuse the food. Nor would it be unnatural for her to overstate the case, from a desire to avoid blame for her thoughtlessness. There is no reason with Mitscherlich and Bücheler to suspect the line as a late interpolation. For the pleonasm cf. the Homeric “βίῃ ἀέκοντος Α” 430 etc.

[417] The list of the Oceanids is borrowed, in the main, from Theog. 349 f., from which passage, together with the quotation of Paus.iv. 30. 4, the names in the text are restored. The writer has taken 16 out of the 41 names in Hesiod, adding Leucippe, Phaeno, Melite, Iache, and Rhodope. Of these, Melite appears as a Nereid in Theog. 246 and in the interpolated passage Il. 18.42. For the meaning of the names see Goettling-Flach on l.c., PrellerRobert i.^{2} p. 552.

[424] The verse has been needlessly suspected. In 5 only the Oceanids are mentioned; but this is quite natural, as they form the greater part of Persephone's companions. Nor is it an objection that Pallas and Artemis end the list; in fact they may well be considered to occupy the place of honour. Not to quote modern analogies, it may be pointed out that the list of nymphs in Theog. 349-361 is closed with the name of Styx “ δή σφεων προφερεστάτη ἐστὶν ἁπασέων”. Pallas and Artemis are present according to most versions: cf. Eur. Hel.1315, Diod.v. 3, Paus.viii. 31. 2, Achill. ii. 150, Claud. Rapt. Pros. i. 228, ii. 205 f. (where they try to defend Proserpine). Ovid does not mention either the Oceanids or other companions by name. For the epithet of Pallas cf. “Παλλάδι τ᾽ ἐγρεμάχῃ” in orac.ap. Hendess 79. 6.

For καί making position see on h. Aphr. 13.

[428] ὥς Περ κρόκον: this is difficult, but no doubt genuine; the emendations are all wild. The meaning might be “as (abundantly as) the crocus.” This, however, would be very prosaic; nor is there reason to suppose, with Ilgen, that the crocus was so much more abundant than the narcissus as to serve for a literary comparison. On the contrary, Aristotle (Mir. Ausc. 111) instances the local profusion of crocus on the promontory of Pelorias in Sicily as exceptional. Probably, therefore, the reference is to the colour of the miraculous flower, the hymn-writer having in mind the yellow Narcissus tazetta (see on 12). Sibthorp (Flora Graeca vol. iv. s.v.) quotes Dioscor. 4. 161 (158), where the tazetta is said to have “κοῖλον κροκοειδές”. For the comparison cf. 178 “κροκηΐῳ ἄνθει ὁμοῖαι”, of yellow hair. A similar expression in Theocr. v. 131πολλὸς δὲ καὶ ὡς ῥόδα κίσθος ἐπανθεῖ” also refers to colour. The “yellow” tazetta is thus distinguished from the N. poeticus, which Dioscorides also mentions.

[429] αὐτάρ, to which Ilgen and Gemoll object, is used in a continuative, not an adversative sense. Περὶ χάρματι, “for joy,” a use of “περί”, lit. “compassed by” not found in Homer but fairly common in later poetry. See L. and S. , and add to the exx. there quoted Apoll. Arg. 3.866ὀδύνῃ πέρι”.

[431] ἅρμασι χρυσείοισι: the short vowel before “χρ” is rare, according to La Roche, Hom. Unters. i. p. 41, who allows as a certain instance in Homer only Il. 23.186 ῥοδόεντι δὲ χρῖεν”. But the shortening is probable in several other passages, e.g. Il. 24.795, θ” 353. See Agar in Class. Rev. April 1901. In the Hymns cf. h. Apoll. 293, 439, h. Herm. 332, viii. 1, Orph. h. lv. 18.

[433] Cf. Od. 7.297 ταῦτά τοι ἀχνύμενός περ ἀληθείην κατέλεξα”, and Od. 3.254.

[434] The first hemistich = Il. 1.601, the second Il. 22.263, h. Herm. 391. The repetition of “θυμός” in three lines is ugly; Bücheler suspects a cento.

[437] Γηθοσύνας: Ruhnken's emendation is supported by Od. 20.8 ἀλλήλῃσι γέλω τε καὶ εὐφροσύνην παρέχουσαι”. So h. Herm. 312δὸς δὲ δίκην καὶ δέξο”. The plur. of “γηθοσύνη” is found in Apollonius.

438-440. The genuineness of this passage (suspected by Mitscherlich and others) is proved by the citation in Philodemus; see crit. n.

[439] κόρηΝ: elsewhere the writer uses the Homeric form; the form “κόρη” is the Attic official title of Persephone (in decrees). The form is also Aeolic; “κόραιSapph. fr. 62. 2.

[440] Hecate was closely associated with Demeter and Persephone. According to one tradition, she was the daughter of Demeter ( Eur. Ion1048, Schol. Apoll. Arg. 3.467, schol. Theocr. ii. 12). In art she often appears in scenes relating to the mission of Triptolemus, and, as “ἡγεμόνη”, in the “κάθοδος” or “ἄνοδος” of Persephone; see Roscher Lex. 1900 f., Preller-Robert i.^{2} p. 761 n. 1, and 763. Farnell (Cults ii. p. 511 f.) thinks that the connexion is due, in part at least, to her chthonian character. This is very probable; it is to be noted, however, that the moon is widely thought to influence vegetation (see Frazer G. B. ii. p. 154 f.), and this belief may possibly have contributed to the association of Heeate, as a moon-goddess, with Demeter or Persephone.

[441] μετ̓: the prep. can hardly go with “ἧκε”, as “μεθιέναι” is nowhere used for “send to fetch.” Hermann read “μέτ̓” “among them”; Gemoll objects to this anastrophe of “μετά” as not found in Homer with the dat. (Hoffmann Tmesis in der Il. i. 18). It might, however, be permissible in the hymn. This passage must be considered in connexion with Il. 15.144 θεοῖσι μετ᾽ ἄγγελος ἀθανάτοισι” and Il. 23.199 μετ᾽ ἄγγελος ἦλθ᾽ ἀνέμοισιν”, where Aristarchus read “μετάγγελος”, internuncia. Modern scholars are not agreed about the existence of “μετάγγελος”, but in Il. 23.199, at least, it seems required. Probably therefore we should read “μετάγγελον” here. Voss emended “τὰς δὲ μέτ̓”, “to fetch them.”

[442] Δημ́ητερα: M has “ἢν μητέρα”, a reading which is just possible, as Rhea was the mother of both Zeus and Demeter ( Theog. 453 f.); an object “αὐτάς” could be supplied from “ταῖς”, and the subject of “ἕλοιτο” is clear from the general sense. But “Δημήτερα” greatly simplifies the construction, and “κυανόπεπλος” is a standing epithet of Demeter in this hymn; cf. on 319. The mistake of M is natural, after “Ῥείην”, and it is noticeable that in the title of xiii. (to Demeter) M has “εἰς μητ . . ρα θεῶν” (corrected to “εἰς δήμητραν”). The scribe may also have had a reminiscence of 360 “μητέρα κυανόπεπλον” (of Demeter).

[445] “Νεῦσε κτλ.”: the construction, if correct, is highly elliptical; fully expressed the sentence would run “νεῦσε . . κούρην <ἰέναι> ὑπὸ ζόφον, <μένειν> δὲ παρὰ μητρί”. Hermann and Bücheler suppose a lacuna after 446.

[448] ὣς ἔφατ̓: see on 316. ἀΓγελιάων: Homer uses the dat. after “ἀπιθεῖν” (cf. 358), but the gen. is defensible, as “οὐκ ἀπίθησε” = “ἐπέκλυεν” (cf. Od. 5.150 Ζηνὸς ἐπέκλυεν ἀγγελιάων”).

[450] Ῥάριον: according to Herodian “π.μ.λ.” 35, Bekker An.693. 11Ράρος” (and therefore its derivatives) should be written with spir. lenis, “Ρ̓αρος”, but the authority is perhaps insufficient. For the Rharian plain cf. Paus.i. 38. 6τὸ δὲ πεδίον τὸ Ῥάριον σπαρῆναι πρῶτον λέγουσι καὶ πρῶτον αὐξῆσαι καρπούς, καὶ διὰ τοῦτο οὐλαῖς ἐξ αὐτοῦ χρῆσθαί σφισι καὶ ποιεῖσθαι πέμματα ἐς τὰς θυσίας καθέστηκεν”. “The plain Rharium seems to have been in the immediate vicinity of Eleusis, but on which side it would be difficult to determine” (Leake Top. Ath. ii. p. 159); Lenormant places it on the north side (Cont. Rev. 38. 134). For the word see coni. praec. 42; Marmor Parium 25, and an inscr. in “Ἐφ. Ἀρχ”. 1883 p. 119 f., which give the usual termination of the name as Raria or Rharia. Byz. Steph. also recognizes Rharion: “Ῥάριον: πεδίον ἐν Ἐλευσῖνι, καὶ ῥαρία γῆ”. Herod. l.c. quotes “Ρ̓αρίδος Δηοῦς”.

φερέσβιον: first in Theog. 693. Apollodorus in schol. Genev. on Il. 21.319 gives the word as “παρ᾽ Ὁμήρῳ”. See Preface p. l. On the word cf. Solmsen l.c. p. 20 f. οὖθαρ ἀρούρης = Il. 9.141; cf. also xxx. 9.

[451] ἕκηλον: not immotum ab aratro (as Baumeister translates), but “idle”; the “work” of the field being to produce crops. Cf. Apoll. Arg. 4.1247εὐκήλῳ δὲ κατείχετο πάντα γαλήνῃ”.

[453] Two seasons are described: spring, when the ears are green; and harvest-time, when the rich furrows are laden with the ripe ears, cut and lying on the ground, while other ears (“τὰ δ̓”) have already been bound into sheaves (Franke). Gemoll quotes

οἵ γε μὲν ἤμων
αἰχμῇς ὀξείῃσι κορωνιόεντα πέτηλα
βριθόμενα σταχύων, ὡσεὶ Δημήτερος ἀκτήν
οἳ δ᾽ ἄρ᾽ ἐν ἐλλεδανοῖσι δέον

. But the original is rather

δράγματα δ᾽ ἄλλα μετ᾽ ὄγμον ἐπήτριμα πῖπτον ἔραζε,
ἄλλα δ᾽ ἀμαλλοδετῆρες ἐν ἐλλεδανοῖσι δέοντο

. In the latter passage, as in the hymn, there are two distinct scenes in the harvesting: (1) reaping, (2) binding; but in the hymn the completion of each operation is described, whereas in the Iliad the operations are still in progress (compare “βρισέμεν”, which implies corn already cut, with the imperf. “πῖπτον”, and “δεδέσθαι” with “δέοντο”).

[455] Ἦρος: the form is found in Alcaeus fr. 45 and other lyric poets.

[456] On ἐλλεδανοῖσι cf. Solmsen Untersuchungen p. 244.

[462] m's supplement “κ᾽ ἐθέλῃσθα” was probably formed from “κ᾽ ἐθέλοιτο” 328, which is anomalous. “ἐθέλῃσθα”, however, is a correct form (Od. 3.92 etc.) and may stand; it is as good as “ἕλοιο”, which Ilgen reads from 444.

[471] For the gifts of husbandry and religion imparted by Demeter to Attica cf. Isocr. iv. 28.

[473] ἔβρις᾿: here with dative, in 456 with gen. The two constructions, as also the act. and pass. forms of the verb, appear to be about equally common.

[476] δρησμοσύνΗν is “ἅπαξ λεγ”. in this sense, and possibly “δρηστοσύνην” (Od. 15.321) is the correct form. But Hesych. and the E. M. recognize “δρησμοσύνη”, explaining by “θεραπεία, ὑπηρεσία”. The reading of M “χρησμοσύνη” might be defended, as the meaning “arrangement” seems possible; see L. and S. s.v., and cf. “χρηστήριον” in the sense of “victim.” Pausanias' variant Πᾶσι is to be preferred to “καλά. πᾶσι” naturally leads to another enumeration of names, and excuses the repetition in 476, to which many commentators object. If the text of M is correct, the addition of “σεμνά” to “καλά” would be very awkward. There is perhaps an echo in an inscr. “Ἐφ. Ἀρχ”. iii. 81 “ὄργια πᾶσιν ἔφαινε βροτοῖς” (of a priest).

[478] The reason for strict secrecy in the Eleusinia has been variously explained: in many cases secret rites belong to a conquered people, who wish to preserve their religious practices from their conquerors; this explanation has been applied to the Eleusinia, which may have been “Pelasgian” (Gardner p. 383 f.). But the cause may rather be due to the nature of religion: as Ramsay (p. 125) remarks, “it was a condition of their good effect that they (the Mysteries) should not hereafter be lightly spoken of”; cf. Strabo 467 “ κρύψις μυστικὴ τῶν ἱερῶν σεμνοποιεῖ τὸ θεῖον”. See further Jevons p. 360 f., who believes that the silence imposed on the initiated was not for concealment (there was little to conceal), but to prevent pollution.

Παρεξίμεν: Agar (Class. Rev. 1896, p. 388) revives Ruhnken's “παρεξέμεν”, not in the sense of “neglect,” but “divulge.” “παρεξίμεν” must mean “transgress,” “overstep,” and will stand if “ἀχέειν” means “give out.” See next note.

Πυθέσθαι: cf. Paus.i. 38. 7τοῖς οὐ τελεσθεῖσιν, ὁπόσων θέας εἴργονται, δῆλα δήπου μηδὲ πυθέσθαι μετεῖναί σφισιν”.

[479] ἀχέειν, “divulge.” The existence of this form was maintained by Buttmann (Lexilogus, Engl. tr. p. 178 f.) here, and in h. Pan 18, where the MSS. give “ἐπιπροχέουσα χέει”. It is apparently defended by Scut. 93 “ἣν ἄτην ἀχέων”, and Ion fr. 39ὕμνον ἀχέων” (MSS. “ἀχαιῶν”), Moschion fr. 187ἀχήσεται”. Zenodotus read the same form instead of “ἰάχων Σ” 160, and apparently supposed it to be an equivalent in sense (although the schol. understands “grieving”). Of the conjectures, there is nothing to be said for “χανεῖν”, and “ἠχέειν” would not become “ἀχέειν”. See generally Schulze K. Z. 29. 247 sq., who however does not admit “ἀχέειν” here, while he reads “ἀχέει” in h. Pan.

[480] This is the earliest allusion to the happiness of the initiated after death; cf. Pind. fr. 137ὄλβιος ὅστις ἰδὼν κεῖν᾽ εἶσ᾽ ὑπὸ χθόν̓: οἶδε μὲν βίου τελευτάν”,

οἶδεν δὲ διόσδοτον ἀρχάν”, Soph. fr. 719ὡς τρισόλβιοι
κεῖνοι βροτῶν, οἳ ταῦτα δερχθέντες τέλη
μόλωσ᾽ ἐς Ἅιδου: τοῖσδε γὰρ μόνοις ἐκεῖ
ζῆν ἐστί, τοῖς δ᾽ ἄλλοισι πάντ᾽ ἔχει κακά”, H. F. 613, Isocr. Panegyr. 28, Plato Phaed. 69 C, [Plato] Axioch. 371 D, Cic. Leg.ii. 14, Aristid. Or.xiii.Aristid. Or., xix.For other references see Lobeck Aglaoph. i. p. 69, Foucart Recherches, etc. p. 53; Dieterich Nekyia p. 64. In this passage, as in Pindar, Sophocles, Euripides and others, it seems to be distinctly claimed that mere initiation procures happiness in a future state; nothing, at all events, is said about the necessity of a virtuous life. Foucart (Recherches, etc. p. 65 f.) thinks that the object of the mysteries was essentially practical: the mystae were taught how to avoid the dangers which beset the soul in its descent to Hades. He proves that such practical instructions formed part of the Orphic religion (p. 66 f.); but it is a most improbable hypothesis that the “ἀπόρρητα” at Eleusis were a kind of “guide to Hades.” Orphic doctrines did not obtain a hold on the Eleusinia until a later period than the date of this hymn. In any case, however, it is clear that, in the general opinion of the early mystae, actual communion with the deities of the underworld was the main, if not the only, essential to salvation. That this belief persisted, is evident from the criticism of Diogenes: “τί λέγεις, ἔφη, κρείττονα μοῖραν ἕξει Παταικίων κλέπτης ἀποθανὼν Ἐπαμεινώνδας, ὅτι μεμύηται” (Plutarch de aud. poet. 4). See Rohde p. 271 f. The belief could, of course, be paralleled from the history of other religions. Serious and educated thinkers, at least in later times, believed that initiation in the Eleusinian or other mysteries was an incentive to virtue (e.g. Andoc. Myst. 31, Diod.v. 49; see Ramsay p. 125, Gardner p. 401); but Rohde (p. 275) considers that the language of Andocides (l.c.μεμύησθε . . . ἵνα τιμωρήσητε μὲν τοὺς ἀσεβοῦντας, σῴζητε δὲ τοὺς μηδὲν ἀδικοῦντας”) is quite exceptional.
ὄπωπεν: the word suggests the “ἐποπτεία”, but no doubt refers more generally to all the sights seen by “μύσται” and “ἐποπταί” alike (if the distinction between the two classes of initiated is as old as the hymn).

484 = Il. 23.142 (“ἂψ ἴμεν”).

[486] “μέγ̓ ὄλβιος κτλ.”: cf. xxx. 7 (with 489 cf. xxx. 12, and with 494 cf. xxx. 18).

[489] Plutus is son of Demeter and Iasion, Theog. 969 f. Cf. scolium in Athen. xiv. 694 “Πλούτου μητέρ᾽ Ὀλυμπίαν ἀείδω

Δήμητρα στεφανηφόροις ἐν ὥραις”,
σέτε, παῖ Διός, Φερσεφόνη”; see Preller-Robert i.^{2} p. 767 n. 5 and 780, Svoronos p. 387 f. The name of Plutus follows those of Demeter and Cora in a prayer, Thesm. 296. Demeter is “πλουτοδότειρα” in Orph. h. 40. 3.
ἄφενος: neuter, as always in Homer (in Il. 23.299 there is a variant “ἄφενον”). Only here in the Hymns.

[490] 490-495 are considered a later addition by Hermann and others.

For confusions caused by “ἄγ̓” or “ἄγε” cf. Il. 7.299, Σ” 314, h. Apoll. 165.

[491] The special cult of Demeter at Paros is attested by the title “Δημητριάς” applied to the whole island (Nicanor ap. Byz. Steph. s.v. “Πάρος”); cf. Herod.vi. 134.The island was colonized from Crete, one of the oldest centres of the cult (see on 123). According to the schol. on Av.1764, Archilochus composed a hymn to Demeter at Paros. The cult is also known by an inscr. from Paros (Ath. Mitth. xvi. p. 6), “δημητρι θεσμοφορωι και κορηι και διι ευβουλει και βαβοι” (= “Βαυβοῖ”). Cf. also Boeckh C.I.G. 2557, and B. C. H. i. p. 135. 54. An ear of corn and the head of Demeter are common types on the coinage; Head Hist. Num. p. 417. See further PaulyWissowa 2722 f.

Ἄντρωνα (“Ἀντρῶνας” in Demosth. x. 9, cf. Strabo 432 and Scylax 63 Müller): a Thessalian town, mentioned in the Catalogue Il. 2.697, opposite Oreus in Euboea, not elsewhere mentioned for the worship of Demeter. But in Il. 2.696 the neigh bouring Pyrasus is called “Δήμητρος τέμενος” (cf. Strabo 435), so that the cult no doubt prevailed along the Pagasaean gulf in very ancient times. There is thus no difficulty in the mention of these places by an early Attic or Eleusinian poet.

494, 495 = xxx. 18, 19. ᾠδ̂ης: the contracted form first in h. Apoll. 20. ὀΠάζειν: this correction of “ὄπαζε” (cf. Il. 21.217 ῥέζε ῥέζειν”, Hes. Op.611ἀπόδρεπε -εν -ειν”) is slighter than to write “πρόφρων δ̓” for “πρόφρονες” (on the analogy of xxx. 18). For the infin. in liturgy see Adami de poet. scenicis p. 243 and Smyth Greek Melic Poets p. 500, who compare Soph. Ant.1144, and the song of the Elean women “ἐλθεῖν, ἥρω Διόνυσε” (Smyth p. 154). On the general Homeric use of the infin. for imper. see Hentze in B. B. xxvii. 1902, p. 106 f.

[495] σεῖο. The writer returns to Demeter, the subject of the hymn, although the previous lines include Persephone in the invocation.


1 For a complete list of full accounts of the myth, or shorter allusions, both in poetry and prose, see Förster der Raub und die Rückkehr der Persephone (1874), pp. 29-98. The list includes Hesiod (Theog. 913-914), Archilochus, Lasus , Sophocles (Triptolemus), Panyasis, Pherecydes, among early poets. For prose cf. especially Diod.v. 3-5.

2 See below, p. 13.

3 Jevons' account of the primitive rites at Eleusis is here followed in the main outlines (op. cit. p. 365 f.); see also Lenormant, p. 852.

4 See Mannhardt Myth. Forsch. p. 224 f., Frazer G. B. ii. p. 168 f. On the duplication of Demeter and Persephone see especially G. B. ii. p. 218 f. This view explains the relation between Demeter and Core at Eleusis more easily than the old theory that Demeter was the Earth. It is not denied, of course, that Demeter became an Earth-goddess, at an early period. For the meaning of the name see Prellwitz Wiener Studien, 1902, xxiv. p. 525, who concludes for “Mutter Da,” “Δα-ματηρ”. Cf. also A. Cook Class. Rev. 1903, p. 176 f., Harrison Proleg. p. 271.

5 Protrept. ii. p. 12. For details see Lenormant and Ramsay. Many such dramatic exhibitions were developed from magical ceremonies intended to secure the revival of vegetation; see Frazer G. B. i. p. 227 f., iii. p. 164 f.

6 Frogs316 f. See Rohde Psyche p. 261 f. who holds the view that Iacchus was introduced by the Athenians.

7 This “δαίμων” is not to be confused with the male god of the Eleusinian triad—Hades, Demeter, Core. See on 2. On such triads see Usener Rhein. Mus. 58 (1903) p. 1 f.

8 For Attic. fuller lists see Gutsche Quaest. de hymn. in Cer. 1872, p. 19 f., Francke op. cit. p. 10 f.

9 Jevons (p. 363) is not justified in inferring from Herod.i. 30 that Eleusis held out until the time of Solon. Ramsay (p. 128) suggests that the religious systems of Athens and Eleusis were largely consolidated by Solon.

10 The language is of no help in determining the place of composition, although there appear to be a few Atticisms; Francke shews that there are also words proper to Ionic, Aeolic, and even Doric (p. 25).

11 See Maass Orpheus p. 178; his suggestion that the hymn belongs to North Greece has nothing to commend it. Fick (B. B. ix. p. 201) thinks that the author, if not an Athenian, was a Parian; the latter alternative has no probability.

12 To give an example, we are told that “ἔπειτα” in 47 implies that Demeter made two journeys.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 United States License.

An XML version of this text is available for download, with the additional restriction that you offer Perseus any modifications you make. Perseus provides credit for all accepted changes, storing new additions in a versioning system.

hide References (407 total)
  • Commentary references from this page (403):
    • Aeschylus, Seven Against Thebes, 860
    • Andocides, On the Mysteries, 31
    • Pseudo-Apollodorus, Library, 1.2.4
    • Pseudo-Apollodorus, Library, 1.5.1
    • Pseudo-Apollodorus, Library, 1.5.3
    • Pseudo-Apollodorus, Library, 3.13.6
    • Pseudo-Apollodorus, Library, 3.15.4
    • Aristophanes, Birds, 1764
    • Aristophanes, Knights, 695
    • Aristophanes, Knights, 782
    • Aristophanes, Peace, 712
    • Aristophanes, Peace, 869
    • Aristophanes, Thesmophoriazusae, 1156
    • Aristotle, Rhetoric, 3.11
    • Bacchylides, Epinicians, 3
    • Bacchylides, Dithyrambs, 17
    • Euripides, Bacchae, 1083
    • Euripides, Electra, 466
    • Euripides, Helen, 1301
    • Euripides, Helen, 1315
    • Euripides, Helen, 835
    • Euripides, Hippolytus, 1391
    • Euripides, Hippolytus, 818
    • Euripides, Ion, 1048
    • Euripides, Orestes, 303
    • Euripides, Phoenician Women, 1255
    • Euripides, Phoenician Women, 1563
    • Herodotus, Histories, 2.111
    • Herodotus, Histories, 3.14
    • Herodotus, Histories, 4.128
    • Herodotus, Histories, 4.143
    • Herodotus, Histories, 5.83
    • Herodotus, Histories, 6.134
    • Herodotus, Histories, 7.140
    • Herodotus, Histories, 7.16
    • Herodotus, Histories, 8.65
    • Herodotus, Histories, 8.77
    • Hesiod, Shield of Heracles, 222
    • Hesiod, Shield of Heracles, 288
    • Hesiod, Shield of Heracles, 449
    • Hesiod, Shield of Heracles, 7
    • Hesiod, Theogony, 23
    • Hesiod, Theogony, 319
    • Hesiod, Theogony, 435
    • Hesiod, Theogony, 447
    • Hesiod, Theogony, 454
    • Hesiod, Theogony, 785
    • Hesiod, Theogony, 913
    • Hesiod, Works and Days, 180
    • Hesiod, Works and Days, 188
    • Hesiod, Works and Days, 202
    • Hesiod, Works and Days, 263
    • Hesiod, Works and Days, 283
    • Hesiod, Works and Days, 340
    • Hesiod, Works and Days, 356
    • Hesiod, Works and Days, 462
    • Hesiod, Works and Days, 465
    • Hesiod, Works and Days, 492
    • Hesiod, Works and Days, 510
    • Hesiod, Works and Days, 607
    • Hesiod, Works and Days, 611
    • Hesiod, Works and Days, 64
    • Hesiod, Works and Days, 655
    • Hesiod, Works and Days, 69
    • Hesiod, Works and Days, 693
    • Hesiod, Works and Days, 74
    • Hesiod, Works and Days, 815
    • Hesiod, Works and Days, 98
    • Homer, Iliad, 10.302
    • Homer, Iliad, 10.353
    • Homer, Iliad, 10.416
    • Homer, Iliad, 10.82
    • Homer, Iliad, 11.319
    • Homer, Iliad, 11.536
    • Homer, Iliad, 11.624
    • Homer, Iliad, 11.640
    • Homer, Iliad, 11.715
    • Homer, Iliad, 12.162
    • Homer, Iliad, 12.205
    • Homer, Iliad, 12.255
    • Homer, Iliad, 13.352
    • Homer, Iliad, 13.557
    • Homer, Iliad, 13.734
    • Homer, Iliad, 13.85
    • Homer, Iliad, 14.308
    • Homer, Iliad, 14.348
    • Homer, Iliad, 14.42
    • Homer, Iliad, 15.144
    • Homer, Iliad, 15.241
    • Homer, Iliad, 15.36
    • Homer, Iliad, 15.666
    • Homer, Iliad, 16.100
    • Homer, Iliad, 16.225
    • Homer, Iliad, 16.548
    • Homer, Iliad, 16.685
    • Homer, Iliad, 17.425
    • Homer, Iliad, 18.203
    • Homer, Iliad, 18.42
    • Homer, Iliad, 18.552
    • Homer, Iliad, 18.567
    • Homer, Iliad, 19.258
    • Homer, Iliad, 19.420
    • Homer, Iliad, 1.18
    • Homer, Iliad, 1.24
    • Homer, Iliad, 1.249
    • Homer, Iliad, 1.53
    • Homer, Iliad, 1.601
    • Homer, Iliad, 21.217
    • Homer, Iliad, 21.249
    • Homer, Iliad, 21.319
    • Homer, Iliad, 21.56
    • Homer, Iliad, 21.565
    • Homer, Iliad, 22.263
    • Homer, Iliad, 22.438
    • Homer, Iliad, 22.460
    • Homer, Iliad, 22.490
    • Homer, Iliad, 23.142
    • Homer, Iliad, 23.186
    • Homer, Iliad, 23.199
    • Homer, Iliad, 23.299
    • Homer, Iliad, 24.194
    • Homer, Iliad, 24.286
    • Homer, Iliad, 24.630
    • Homer, Iliad, 24.647
    • Homer, Iliad, 24.785
    • Homer, Iliad, 24.795
    • Homer, Iliad, 24.93
    • Homer, Iliad, 2.117
    • Homer, Iliad, 2.238
    • Homer, Iliad, 2.305
    • Homer, Iliad, 2.342
    • Homer, Iliad, 2.381
    • Homer, Iliad, 2.39
    • Homer, Iliad, 2.55
    • Homer, Iliad, 2.558
    • Homer, Iliad, 2.625
    • Homer, Iliad, 2.696
    • Homer, Iliad, 2.697
    • Homer, Iliad, 2.755
    • Homer, Iliad, 2.794
    • Homer, Iliad, 2.87
    • Homer, Iliad, 3.175
    • Homer, Iliad, 3.244
    • Homer, Iliad, 3.322
    • Homer, Iliad, 3.386
    • Homer, Iliad, 4.478
    • Homer, Iliad, 4.497
    • Homer, Iliad, 4.877
    • Homer, Iliad, 5.137
    • Homer, Iliad, 5.364
    • Homer, Iliad, 5.500
    • Homer, Iliad, 5.853
    • Homer, Iliad, 5.879
    • Homer, Iliad, 6.175
    • Homer, Iliad, 6.2
    • Homer, Iliad, 6.389
    • Homer, Iliad, 6.396
    • Homer, Iliad, 6.400
    • Homer, Iliad, 6.483
    • Homer, Iliad, 6.509
    • Homer, Iliad, 7.109
    • Homer, Iliad, 7.16
    • Homer, Iliad, 7.299
    • Homer, Iliad, 8.398
    • Homer, Iliad, 8.436
    • Homer, Iliad, 8.494
    • Homer, Iliad, 9.141
    • Homer, Iliad, 9.157
    • Homer, Iliad, 9.191
    • Homer, Iliad, 9.262
    • Homer, Iliad, 9.353
    • Homer, Iliad, 9.42
    • Homer, Iliad, 9.454
    • Homer, Iliad, 9.572
    • Homer, Iliad, 9.594
    • Homer, Odyssey, 10.234
    • Homer, Odyssey, 11.137
    • Homer, Odyssey, 11.16
    • Homer, Odyssey, 11.208
    • Homer, Odyssey, 11.248
    • Homer, Odyssey, 11.386
    • Homer, Odyssey, 11.427
    • Homer, Odyssey, 11.588
    • Homer, Odyssey, 11.598
    • Homer, Odyssey, 12.47
    • Homer, Odyssey, 13.198
    • Homer, Odyssey, 13.256
    • Homer, Odyssey, 13.68
    • Homer, Odyssey, 13.84
    • Homer, Odyssey, 14.199
    • Homer, Odyssey, 14.251
    • Homer, Odyssey, 14.346
    • Homer, Odyssey, 14.347
    • Homer, Odyssey, 15.321
    • Homer, Odyssey, 15.444
    • Homer, Odyssey, 15.469
    • Homer, Odyssey, 17.250
    • Homer, Odyssey, 17.277
    • Homer, Odyssey, 17.547
    • Homer, Odyssey, 17.80
    • Homer, Odyssey, 18.122
    • Homer, Odyssey, 18.140
    • Homer, Odyssey, 18.408
    • Homer, Odyssey, 19.172
    • Homer, Odyssey, 19.55
    • Homer, Odyssey, 1.333
    • Homer, Odyssey, 1.428
    • Homer, Odyssey, 20.311
    • Homer, Odyssey, 20.344
    • Homer, Odyssey, 20.8
    • Homer, Odyssey, 21.35
    • Homer, Odyssey, 22.6
    • Homer, Odyssey, 24.252
    • Homer, Odyssey, 3.254
    • Homer, Odyssey, 3.92
    • Homer, Odyssey, 3.93
    • Homer, Odyssey, 3.99
    • Homer, Odyssey, 4.11
    • Homer, Odyssey, 4.442
    • Homer, Odyssey, 4.604
    • Homer, Odyssey, 4.659
    • Homer, Odyssey, 4.788
    • Homer, Odyssey, 5.125
    • Homer, Odyssey, 5.150
    • Homer, Odyssey, 5.272
    • Homer, Odyssey, 5.488
    • Homer, Odyssey, 5.59
    • Homer, Odyssey, 7.101
    • Homer, Odyssey, 7.131
    • Homer, Odyssey, 7.297
    • Homer, Odyssey, 8.74
    • Homer, Odyssey, 9.468
    • Homeric Hymns, Hymn 1 to Dionysus
    • Homeric Hymns, Hymn 13 to Demeter
    • Homeric Hymns, Hymn 2 to Demeter, 159
    • Homeric Hymns, Hymn 2 to Demeter, 208
    • Homeric Hymns, Hymn 2 to Demeter, 239
    • Homeric Hymns, Hymn 2 to Demeter, 265
    • Homeric Hymns, Hymn 2 to Demeter, 292
    • Homeric Hymns, Hymn 2 to Demeter, 420
    • Homeric Hymns, Hymn 2 to Demeter, 48
    • Homeric Hymns, Hymn 2 to Demeter, 480
    • Homeric Hymns, Hymn 2 to Demeter, 489
    • Homeric Hymns, Hymn 2 to Demeter, 49
    • Homeric Hymns, Hymn 2 to Demeter, 50
    • Homeric Hymns, Hymn 3 to Apollo
    • Homeric Hymns, Hymn 3 to Apollo, 1
    • Homeric Hymns, Hymn 3 to Apollo, 107
    • Homeric Hymns, Hymn 3 to Apollo, 108
    • Homeric Hymns, Hymn 3 to Apollo, 118
    • Homeric Hymns, Hymn 3 to Apollo, 123
    • Homeric Hymns, Hymn 3 to Apollo, 165
    • Homeric Hymns, Hymn 3 to Apollo, 271
    • Homeric Hymns, Hymn 3 to Apollo, 293
    • Homeric Hymns, Hymn 3 to Apollo, 322
    • Homeric Hymns, Hymn 3 to Apollo, 347
    • Homeric Hymns, Hymn 3 to Apollo, 350
    • Homeric Hymns, Hymn 3 to Apollo, 36
    • Homeric Hymns, Hymn 3 to Apollo, 388
    • Homeric Hymns, Hymn 3 to Apollo, 395
    • Homeric Hymns, Hymn 3 to Apollo, 407
    • Homeric Hymns, Hymn 3 to Apollo, 444
    • Homeric Hymns, Hymn 3 to Apollo, 46
    • Homeric Hymns, Hymn 3 to Apollo, 464
    • Homeric Hymns, Hymn 3 to Apollo, 465
    • Homeric Hymns, Hymn 3 to Apollo, 82
    • Homeric Hymns, Hymn 3 to Apollo, 9
    • Homeric Hymns, Hymn 3 to Apollo, 91
    • Homeric Hymns, Hymn 4 to Hermes, 113
    • Homeric Hymns, Hymn 4 to Hermes, 173
    • Homeric Hymns, Hymn 4 to Hermes, 208
    • Homeric Hymns, Hymn 4 to Hermes, 231
    • Homeric Hymns, Hymn 4 to Hermes, 234
    • Homeric Hymns, Hymn 4 to Hermes, 308
    • Homeric Hymns, Hymn 4 to Hermes, 312
    • Homeric Hymns, Hymn 4 to Hermes, 322
    • Homeric Hymns, Hymn 4 to Hermes, 332
    • Homeric Hymns, Hymn 4 to Hermes, 37
    • Homeric Hymns, Hymn 4 to Hermes, 390
    • Homeric Hymns, Hymn 4 to Hermes, 391
    • Homeric Hymns, Hymn 4 to Hermes, 562
    • Homeric Hymns, Hymn 4 to Hermes, 57
    • Homeric Hymns, Hymn 5 to Aphrodite
    • Homeric Hymns, Hymn 5 to Aphrodite, 102
    • Homeric Hymns, Hymn 5 to Aphrodite, 108
    • Homeric Hymns, Hymn 5 to Aphrodite, 123
    • Homeric Hymns, Hymn 5 to Aphrodite, 13
    • Homeric Hymns, Hymn 5 to Aphrodite, 132
    • Homeric Hymns, Hymn 5 to Aphrodite, 156
    • Homeric Hymns, Hymn 5 to Aphrodite, 157
    • Homeric Hymns, Hymn 5 to Aphrodite, 173
    • Homeric Hymns, Hymn 5 to Aphrodite, 214
    • Homeric Hymns, Hymn 5 to Aphrodite, 221
    • Homeric Hymns, Hymn 5 to Aphrodite, 253
    • Homeric Hymns, Hymn 5 to Aphrodite, 257
    • Homeric Hymns, Hymn 5 to Aphrodite, 259
    • Homeric Hymns, Hymn 5 to Aphrodite, 266
    • Homeric Hymns, Hymn 5 to Aphrodite, 284
    • Homeric Hymns, Hymn 5 to Aphrodite, 31
    • Homeric Hymns, Hymn 5 to Aphrodite, 43
    • Homeric Hymns, Hymn 5 to Aphrodite, 82
    • Homeric Hymns, Hymn 5 to Aphrodite, 92
    • Pausanias, Description of Greece, 1.14.2
    • Pausanias, Description of Greece, 1.31.3
    • Pausanias, Description of Greece, 1.37.2
    • Pausanias, Description of Greece, 1.37.3
    • Pausanias, Description of Greece, 1.38.3
    • Pausanias, Description of Greece, 1.38.5
    • Pausanias, Description of Greece, 1.38.6
    • Pausanias, Description of Greece, 1.38.7
    • Pausanias, Description of Greece, 1.39.1
    • Pausanias, Description of Greece, 2.14.3
    • Pausanias, Description of Greece, 2.30.4
    • Pausanias, Description of Greece, 2.32.2
    • Pausanias, Description of Greece, 2.35.5
    • Pausanias, Description of Greece, 2.5.5
    • Pausanias, Description of Greece, 4.1.8
    • Pausanias, Description of Greece, 4.30.4
    • Pausanias, Description of Greece, 4.33.4
    • Pausanias, Description of Greece, 8.31.2
    • Pausanias, Description of Greece, 8.37.7
    • Pausanias, Description of Greece, 8.37.9
    • Pausanias, Description of Greece, 8.42
    • Pausanias, Description of Greece, 8.42.11
    • Pausanias, Description of Greece, 9.23.2
    • Pausanias, Description of Greece, 9.23.4
    • Pausanias, Description of Greece, 9.25.1
    • Pausanias, Description of Greece, 9.31.9
    • Pausanias, Description of Greece, 9.8.1
    • Pindar, Isthmean, 4
    • Pindar, Isthmean, 7
    • Pindar, Olympian, 3
    • Pindar, Olympian, 8
    • Pindar, Pythian, 1
    • Pindar, Pythian, 9
    • Plato, Phaedo, 69c
    • Plato, Cratylus, 411d
    • Plato, Symposium, 212c
    • Plato, Euthydemus, 285c
    • Plato, Timaeus, 63b
    • Sophocles, Antigone, 1130
    • Sophocles, Antigone, 1144
    • Sophocles, Antigone, 722
    • Sophocles, Oedipus at Colonus, 1590
    • Sophocles, Oedipus at Colonus, 683
    • Sophocles, Philoctetes, 1190
    • Thucydides, Histories, 2.15
    • Pindar, Pythian, 4
    • Callimachus, Hymn to Demeter
    • Apollonius Rhodius, Argonautica, 1.1125
    • Apollonius Rhodius, Argonautica, 1.1169
    • Apollonius Rhodius, Argonautica, 1.165
    • Apollonius Rhodius, Argonautica, 1.834
    • Apollonius Rhodius, Argonautica, 2.1214
    • Apollonius Rhodius, Argonautica, 2.699
    • Apollonius Rhodius, Argonautica, 2.821
    • Apollonius Rhodius, Argonautica, 3.1213
    • Apollonius Rhodius, Argonautica, 3.1219
    • Apollonius Rhodius, Argonautica, 3.293
    • Apollonius Rhodius, Argonautica, 3.467
    • Apollonius Rhodius, Argonautica, 3.714
    • Apollonius Rhodius, Argonautica, 3.866
    • Apollonius Rhodius, Argonautica, 3.873
    • Apollonius Rhodius, Argonautica, 4.1119
    • Apollonius Rhodius, Argonautica, 4.112
    • Apollonius Rhodius, Argonautica, 4.1247
    • Apollonius Rhodius, Argonautica, 4.236
    • Apollonius Rhodius, Argonautica, 4.290
    • Apollonius Rhodius, Argonautica, 4.674
    • Apollonius Rhodius, Argonautica, 4.806
    • Apollonius Rhodius, Argonautica, 4.869
    • Apollonius Rhodius, Argonautica, 4.872
    • Apollonius Rhodius, Argonautica, 4.952
    • Theocritus, Idylls, 17
    • Theocritus, Idylls, 2
    • Theocritus, Idylls, 25
    • Theocritus, Idylls, 5
    • Theocritus, Idylls, 6
    • Theocritus, Idylls, 7
    • Ovid, Metamorphoses, 10.434
    • Ovid, Metamorphoses, 4.11
    • Ovid, Metamorphoses, 5.385
    • Ovid, Metamorphoses, 5.392
    • Ovid, Metamorphoses, 5.532
    • Ovid, Metamorphoses, 5.535
    • Ovid, Metamorphoses, 5.567
    • Vergil, Aeneid, 11.480
    • Vergil, Aeneid, 1.403
    • Vergil, Aeneid, 1.573
    • Vergil, Eclogues, 1
    • Vergil, Eclogues, 10
    • Lucan, Civil War, 6.740
    • Cicero, De Legibus, 2.14
    • Plutarch, Theseus, 10
    • Statius, Achilleis, 2
    • Diodorus, Historical Library, 5.3
    • Diodorus, Historical Library, 5.3.4
    • Diodorus, Historical Library, 5.4
    • Diodorus, Historical Library, 5.49
    • Diodorus, Historical Library, 5.77
    • Ovid, Fasti, 1
    • Ovid, Fasti, 4
    • Ovid, Fasti, 5
  • Cross-references in notes from this page (4):
    • Aristophanes, Frogs, 316
    • Herodotus, Histories, 1.30
    • Hesiod, Theogony, 913
    • Diodorus, Historical Library, 5.3
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: