^{page 154 note 1} Edited by Lobel, E. in Oxyrhynchus Papyri, xxiv, no. 2390 fr. 2Google Scholar; now fr. 5 of Alcman in Page's Poetae Melici Graeci.

^{page 154 note 2} Cf. Lobel, ad loc.; Page, C.R. N.S. ix (1959), 20–21Google Scholar; Barrett, , Gnomon xxxiii (1961), 689Google Scholar; Bowra, , Greek Lyric Poetry, 2nd ed.(1961), pp. 25–26.Google Scholar A discussion has also been promised by Peek, W. (Philologus civ [1960], 180).Google Scholar

^{page 154 note 3} In later literature compare Ananke in Simmias' Pteryges, Natura in Claud, de raptu Proserp. 1. 248 (cf. Ov. M. 1. 21).

^{page 155 note 1} On the development of abstracta froi nomina agentis and other adjectival formations cf. Usener, H., Götternamen, pp. 364–75Google Scholar; Kretschmer, P., Glotta xiii (1924), 101ff.Google Scholar

^{page 155 note 2} I quote from the translation of Griffith, R. T. H., The Hymns of the Rigveda (1897).Google Scholar

^{page 155 note 3} I quote from the translation of E. A. Speiser in Ancient Near Eastern Texts, edited by J. B. Pritchard, 2nd ed., 1955 (hereafter abbreviated A.N.E.T. ²), p. 68.

^{page 156 note 1} in Lobel's transcript is a mistake, For the masculine see Barrett, loc. cit., note 5.

^{page 156 note 2} But I have doubts about Barrett's seems not to be used in invocations of the Muse in the early period.

^{page 155 note 1} The most sensible account is that by Kirk, G. S. in Kirk-Raven, , Presocratic Philosophers, pp. 48–72.Google Scholar

^{page 155 note 2} The ‘seven nooks’ of the Suda title are notoriously discrepant from the five nooks in which, in Pherecydes' cosmogony, Chronos' seed was distributed so as to produce the ‘five-nook generation of gods’ (Damasc. Pr. 124 b = Diels-Kranz 7 A 8). I find it hard to believe that these five nooks were somehow augmented by two further nooks in a separate category, and the most reasonable course is surely to alter to with Preller (Rh. Mus. iv [1846], 378), or better to (see below).

^{page 155 note 3} seems to have been taken in the way I advocate by Wilamowitz, Kl. Schr. v (2), p. 129, since he prints it with a small initial and and with capitals.

^{page 155 note 4} Cf. Alcmaion of Croton fr. 1. As the name of Pherecydes' father was known and not disputed in antiquity, he probably named him, as do Alcmaion and Antiochus in their sphragides. Hecataeus, Herodotus, and Thucydides only give their city. Against the assumption that Pherecydes began with a sphragis of this kind is the fact that Diogenes Laertius, obviously following a library catalogue, quotes fr. 1 as die beginning of the work. But it may be diat the sphragis was ignored in the pinax entry; to include it might have led to confusion where there were two works by the same author. There is reason to think that this has been done in the case of the Triagmoi of Ion of Chios, of which Harpocration cites from a catalogue etc. It is probably the same with fr. 1 of Heraclitus, which Aristotle and Sextus mention as the beginning of his book, though they are not following a catalogu.

^{page 155 note 5} Gell, 18. 9. 5; Birt, T., Das antike Buch-wesen (1882), pp. 482–4Google Scholar, and Kritik und Hermeneutik (Handb. d. Altertumsw. I, 3 [1913]), p. 295.Google Scholar I owe these references to my wife. Cf. also [Arist.] De plant. 22. 822^b33 A.P. 11. 356 and 16. 88

^{page 156 note 1} This is the accentuation of the manuscripts of Diogenes (with v.l. Zeùs) and Clem. Strom. 6. 9, and it is supported by ancient grammarians. See Cook, A. B., Zeus ii. 351 n.2.Google Scholar

^{page 156 note 2} She appears to be a hypostasis from the attested on an inscription from Myconos, , B.C.H. vii (1883), 398.Google Scholar In cult the epithet is more often borne by Zeus and Demeter. Ge is called Chthonie by ‘Musaeus’ fr. 11 Diels; a Chthonie also appears in Empedocles' catalogue of personified opposites, paired with Heliope (fr. 122. 1).

^{page 156 note 3} Against Kirk-Raven, p. 56.

^{page 156 note 4} If Pherecydes used this expression, as seems likely, it will have been in the form rather than as Damascius has it. Cf. Hdt. 1. 136 , 144 , 2. 10 and 4. 47 6. 89 9. 83 (v. 1. ), and Debrunner, A., Griechische Wortbildungslehre, p. 69.Google Scholar In the title of the book would be more easily corrupted to than would.

^{page 156 note 5} Euphorion fr. 101 Powell and other sources, cf. Roscher, iii. 1030.

^{page 157 note 1} Weltenmantel und Himmelszelt (1910) i.: 322 ff. A considerable part of this book bears on the interpretation of Pherecydes; it contains vast learning, which is however often injudiciously used. His attempt to find psephological symbolisms in the words and names used by Pherecydes (pp. 334 ff.) has found no followers.

^{page 157 note 2} See Σ^B γ 67.

^{page 160 note 1} In frr. 236 and 239 Zeus-Helios is equated with Dionysus, and in fr. 237 Dionysus with Phanes.

^{page 161 note 1} This division of words, which I have proposed in Gött. Gel. Anz. 1963, 170, is preferable on purely metrical grounds to the usual division besides being easier to supplement:

^{page 161 note 2} Paus. 8. 41. 4. The locals for some reason regarded her as a form of Artemis. Cf. Wilamowitz, , Glaube der Hellenen i. 221.Google Scholar

^{page 161 note 3} En. El. 1. 140–2, 2. 27–29, 3. 31–33, 89–91. The translation ‘Sphinx’ is marked as uncertain by Speiser. In the case of the Viper, Dragon, Great-Lion, and Mad-Dog, a variant has the plural in one or other place where the list occurs.

^{page 161 note 4} Besides their several appearances in Enuma Eliˇ, no extant text of which is earlier than 1000 B.C., these monsters were depicted on door-panels of sanctuaries of Marduk and his consort Sarpanîtu, according to an inscription of Agum II, who restored them in the fifteenth century B.C. (Heidel, A., The Babylonian Genesis, 2nd ed. [1951], p. 13).Google Scholar

^{page 161 note 5} It was first compared with Pherecydes by Jensen, P., Kosmologie der Babylonier (1890), PP. 303f.Google Scholar

^{page 161 note 6} For a full account of the Norse myth see Olrik, Axel, Ragnarök (1922)Google Scholar. Although set in the future instead of the past, it shows other similarities with the Near Eastern Succession Myth, and may be an actual offshoot of it carried north by the Goths from the shores of the Black Sea. The Iranian myth is related in the Bundahish, translated by West, E. W. in volume v of Sacred Books of the East, ed. Müller, F. Max, Oxford, 1879–1910.Google Scholar

^{page 164 note 1} Text and translation of all Norse poems cited can be found in Vigfusson-Powell, , Corpus Poeticum Boreale, Oxford, 1883 (two volumes).Google Scholar

^{page 164 note 2} Another myth connects Typhon with the bed of a river. Strabo 750— Cf. Malalas, p. 197 Dind.

^{page 164 note 3} Tertull, . De corona militis 7Google Scholar: Satumum Pherecydes ante omnes refert coronatum, Iouem Diodorus post deuictos Titanas. It is implied that the occasion in Pherecydes was similar to that in Diodorus (6 fr. 4). He may have explicitly made it the aition for the wearing of crowns; and it may have been suggested to him by the actual wearing of crowns by the victors in the ritual battle as he was acquainted with it.

^{page 164 note 4} Cf. Zeller-Nestle, , Gesch. d. gr. Phil. i⁶. 116.Google Scholar

^{page 164 note 5} Bodleian MS. Gr. class, f. 48 (P), no. 32312 in the Supplementary Catalogue; fr. 2 Diels. First edited by Grenfell and Hunt, Greek Papyri, Second Series, 1897, pp. 21–23. The text given above is based on a recollation of the original.

^{page 165 note 1} Compare Orph. fr. 238, where Zeus-Dionysus puts on a robe in the likeness oi the sun, above it a deerskin dappled to represent the stars, and round his waist a belt representing Oceanus. In a Sumerian poem (Falkenstein-Soden, , Sumerische u. Akkadischi Hymnen u. Gebete [1953], pp. 67–68Google Scholar) the goddess Inanna says of her father Enlil, ‘The heavens he has put upon my head as a crown, The earth as sandals upon my feet.’

^{page 165 note 2} and can be near-synonyms, e.g. Hes. Th. 393–6, 426–7, 449.

^{page 165 note 3} There were male weavers and embroiderers in Greece by the fourth century, cf. Pl. Phd. 87 b, Rep. 369 d, Aeschin. 1. 97.

^{page 166 note 1} Il. 21. 195–7, Pind. fr. 326, etc. Rivers might be mentioned apart from earth, cf. Hes. Th. 109, Wilamowitz on Eur. H.F. 1296.

^{page 166 note 2} Pherecydes uses for ‘house’ earlier in the same fragment, but the more poetic word, which is once used by Herodotus (2. 62), suits the more cosmological significance to be assumed here.

^{page 167 note 1} For the robe of sovereignty cf. Enuma Eliš 4. 29, Nonn. D. 1. 480, 2. 571, 3. 197.

^{page 167 note 2} Cf. U. Holmberg, op. cit.

^{page 167 note 3} Cf. Eisler, op. cit. passim; Gantar, loc. cit.

^{page 167 note 4} The world tree usually rises above the earth, its branches supporting or representing the sky. There are however cases where it supports the earth, see Eisler, pp. 765 f., Holmberg, pp. 53 f.

^{page 167 note 5} We must assume that the tree is a tree, and not a loom or the mast of a ship; see Kirk-Raven, p. 63.

^{page 168 note 1} Eisler, pp. 592, 594, figs. 70–71; cf. Danthine, Héléne, Le palmier-dattier et les arbres sacrés dans l'iconographie de l'Asie occidentale ancienne (1937) ii, figs. 333, 363, 429–33, etc.Google Scholar

^{page 168 note 2} Eisler, p. 595, fig. 73; but see rather the detail in Barnett, R. D., Assyrian Palace Reliefs (not dated, appeared in 1959), pl. 114.Google Scholar

^{page 168 note 3} Barnett, R. D. and Falkner, M., The Sculptures of Tiglath-Pileser III (London, 1962), pl. LIII and p. 13Google Scholar; differently interpreted on p. xxiii and in the caption to the plate.

^{page 169 note 1} Cited by Origen, c. Cels. 6. 42, from Celsus. The manuscripts give (cf. Kühner-Blass i. 294–5), This is not to be taken as evidence for a change of Zas into Zeus. The same modernization has taken place in part of the tradition of Diogenes at fr. 1.

^{page 169 note 2} Compare also the spring flowing from the Lake of Memory, from which the initiate is bidden to drink when he arrives in Hades, in the gold plates from Petelia and Eleu-thernae in Crete, Orph, fr. 32 (a) and (b).

^{page 170 note 1} A very similar method of interpretation is seen in the passage of Porphyry cited above. Those who think that the may have been an actual emission of semen, sc. that of Chronos (Kirk-Raven, p. 59, Schwabl, , R.-E. Supp. ix. 1460Google Scholar) can have paid no attention to the purport of the passage.

^{page 170 note 2} The fragment was righdy assigned to Pherecydes of Syros by Wilamowitz, , Kl. Schr. v (2), p. 133 n. 1Google Scholar; less confidently by Jacoby, (F.G.H. i. 429.39), who later changed his mind (Mnem. 1947, 60. n. 137).Google Scholar

^{page 170 note 3} But see Guthrie, W. K. C., Orpheus and Greek Religion, pp. 247f.Google Scholar

^{page 170 note 4} I doubt whether D.L. 1. 119 (fr. 12 Diels) belongs here.

^{page 171 note 1} In Hesiod's cosmology, where are the roots of earth and sea, there also are their ‘springs’, Th. 728, 736 ff. The oak of Zeus at Dodona, at the foot of which the oracular spring of Zeus Naios issued forth, would be another Greek parallel if it were true that it was thought of as a type of the world-tree; but the statement in Eisler, repeated by Holmberg, that its roots were said to reach down into Tartarus, has no ancient foundation. It derives from a misunderstanding (which a little checking of references would have corrected) of C. Boetticher, , Baumkultus der Hellenen (1856), p. 112.Google Scholar

^{page 171 note 2} For the addition of wings to unexpected things compare also the winged bed which conveys the sun round Oceanus at night in Mimn. 10. 5 ff.; and on the popularity of wings in archaic art cf. Nilsson, , Minoan-Mycenaean Religion², pp. 507f.Google Scholar; Eitrem, , R.-E. vi A. 886f.Google Scholar

^{page 173 note 1} It is far-fetched to interpret the testimonia as referring to worlds successive in time and not coexistent, and quite in admissible to reject them altogether as Kirk does (C.Q. N.s. v [1955], 28–32Google Scholar and Presoc. Phil., pp. 122 f.) from inability to imagine what empirical considerations could have led Anaximander to the idea of innumerable worlds. He wished to explain the world in terms of universal natural processes. Whatever process caused this world to form out of the Infinite would naturally be likely to form other worlds elsewhere and at different times. They are not visible to us, but their existence can be inferred from that of our world, although they are quite unconnected with it. That Anaximander reasoned like this ought not to strain our credulity.

^{page 174 note 1} The fact that none of the doxographers speak of a vortex in connexion with Anaximander's cosmology certainly indicates that this detail was unknown to Theophrastus; but it is likely enough that Theophrastus' information was second-hand and incomplete. I am not convinced by Hölscher's argument (Hermes lxxxi [1953] 266Google Scholar) that a vortex is excluded by the spontaneous, non–mechanical way in which the shell of fire at first surrounds the earth ‘like the bark round a tree’ (ps.-Plut. Strom. 2).

^{page 174 note 2} Tannery, P., Pour l'histoire de la science hellène² (1930), p. 74.Google Scholar

^{page 175 note 1} His explanation of the annual Nile flooc (Aet. 4. 1. 1, cf. Hdt. 2. 20) might be seer as a translation into physical theory of the Egyptian myth according to which the flood represented Osiris' victory over Seth. Seth manifested himself in the drought and the hot south wind from the desert (cf. Seippe, G. Der Typhonmythos, Diss. Greifiwald 1939, p. 109Google Scholar), Osiris in the new water. The essence of Thales' theory too was that the flood was delayed by wind (and so built up its strength): but of course it had to be the north winds, the Etesians, that held back the northward-flowing Nile.

^{page 176 note 1} It is sad to see scholars still treating the ancient chronographers' date for Arctinus as if it could be based on civic archives of some kind.