^{page 113 note 1} Adam, , Republic of Plato (1902)Google Scholar, adopts προσφερ⋯σ, the reading of the second hand in A.

^{page 113 note 2} T⋯ν λειμ⋯να 614E and τῷ λειμ⋯νι 616B both refer to the same place as is described at 614c as τ⋯πον τιν⋯ δαιμ⋯νιον and the place of judgement.

^{page 113 note 3} The position of the souls in the universe when they see the light first, and their position again when at the end of a day' journey they are κατ⋯ μ⋯σον τ⋯ φ⋯σ, are vexed questions. Adam locating them on the true surface of the earth when they first see the light (note on 616B 11), then plunges them into subterranean regions at the very centre of the earth when they are said to be κατ⋯ μ⋯σον τ⋯ φ⋯σ. His reasons for doing so are given in notes on 616B 13, 621B 10, and App. VI. to Bk. X. But it is to noted that the words κατ⋯ μ⋯σον τ⋯ φ⋯σ need not be pressed so as to make the souls come to the centre of the earth when they come to the of the light. ‘The middle of the light’ may mean the point at which the shaft of light pierces the surface of the earth, the centre of the universe. (Stewart, , Myths of Plato, pp. 152 and 167Google Scholar); or Plato's words may, I think, be interpreted to mean simply that the souls enter the light and so are in the midst of it (cf. the interpretation of SirHeath, T., Aristarchus of Samos, p. 152Google Scholar). Neither need the fact that the souls shoot upwards to birth (621B) imply that they saw the vision of Ananke' spindle from a subterranean region, ἄνω is explicable if it is assumed with ProfessorStewart, (o.c., pp. 111, 165, 168)Google Scholar that the souls were moving on the surface of the earth in the antipodal hemisphere. I think that Stewart and Heath are undoubtedly right in arguing against a subterranean position for the souls when they see the vision. I am not absolutely certain, however, that the souls remain always on the surface of the earth in the antipodal hemisphere. If κατ⋯ μ⋯σον τ⋯ φ⋯ς simply as means ‘the light,’ it is conceivable that they advanced to celestial regions.

^{page 114 note 1} It is possible to take ἄνωθ;εν not with τεταμ⋯νον but with καθ;ορ⋯ν (see Adam, note ad loc.)—i.e. ‘they saw from above stretching through all the heaven and earth.’ This would make the point of view of the souls definitely celestial.

^{page 114 note 2} Aὐτο⋯ is ambiguous as Adam points out (note ad loc.), and may refer either to φ⋯σ or to οὐρανοῡ. I take the pronoun as referring to φ⋯σ. The general sense of the passage is quite clear, as Adam points out: for the next sentence clearly shows that it is the heaven that is bound and that the light is a band thereof. But if αὐτο⋯ refers to οὐρανο⋯, the meaning may be that the heaven is bound by many chains, of which the light is one (should the prefix in σ;⋯νδεσμν be pressed?).

^{page 114 note3} The existence of a straight part of the light was denied by Boeckh, (Kleine Schriften III., pp. 297 sqq.)Google Scholar. and by Martin, (Mémoires de l' Académic dis Inscriptions et Belles Lettres XXX., pp. 93 sqq.)Google Scholar. Boeckh held that the souls from a place outside the universe saw in the distance a half-circle of the Milky Way, which because of their position appeared to them straight like a pillar; Martin that the souls saw above their heads a halfcircle of the Milky Way, but thought that it was really a straight light like a pillar, accounting for the dip at each horizon as an effect of perspective (!). But there is nothing in Plato' words which would lead us to believe that the appearance was different from the reality, and both scholars find difficulty in dealing with πι⋯ παντ⋯ς τος οορανο⋯ κα⋯ γασ τεταμανον.

^{page 114 note 4} Note on 616B II.

^{page 114 note 5} See Adam in the note cited above.

^{page 114 note 6} Adam, note on 616c 14 sub fin.

^{page 115 note 1} P. 143 (Hiller).

^{page 115 note 2} II., p. 199, 31 sqq.(Kroll).

^{page 115 note 3} Dr. A. B. Cook also reminds me that a world axis which has breadth need not be surprising in a myth which is steeped in Pythagorean doctrine, seeing that the Pythagoreans thought of lines as having breadth, just as they thought of points as having magnitude (Burnet, , Early Greek Philosophy³, p. 290)Google Scholar.

^{page 115 note 4} Some of the ancients supposed that the ὐποζμματα were wooden planks (Procl. in Remp. Comm. II., p . 200, 25 Kroll and scholium on this passage, p. 381; cf. scholium, on Aristophanes, Knights 279Google Scholar, repeated under heading ὐποζώματα in Suidas). But it is proved that the ὐποζώματα were ropes not planks by the fact that they occur among the σκ⋯νη κρεμαστ⋯) detachable parts, as opposed to the σκ⋯νη χ⋯λινα, wooden gear, in inscriptions giving inventories of triremes and their gear belonging to the Peiraeus. See Torr, C., Ancient Ships, p. 41, note 100 and references given in his note 103 on p. 42Google Scholar.

^{page 115 note 5} Pp. 41–42. The view of Adam and Torr is also maintained by Boeckh, , Urkunden über das Seewesen des Attischen Staates, pp. 133–138Google Scholar; Breusing, , Die Nautih der Alten, pp. 170–184Google Scholar; Cartault, , La Trière Athénienne, p. 56Google Scholar; Graser, , De Veterum Re Nauali, § 70Google Scholar.

^{page 115 note 6} Harvard Studies in Classical Philology XXXIV., pp. 63 sqq.: ‘The Hypozomata of Ancient Ships.’ The suggestion had already been put forward by Warre, , J.H.S. V., p. 216Google Scholar.

^{page 116 note 1} See Cecil Torr, Ancient Ships, Plate I., Nos. 4 and 5.

^{page 116 note 2} The fact that the pull was carried down to the keel by a device such as that of the smaller ropes passed under the keel at stem and stern in the Egyptian ships could hardly justify the name, I think.

^{page 116 note 3} For the ὐποζώματα, according to this view, are undergirders in the sense that they occupied a place on the lower part of the outside of the ship, under the walls of the ship where they projected, where the structure began to narrow downwards towards the keel. Cf. Graser, , o.c., § 82Google Scholar.

^{page 116 note 4} MrTorr, , however, holds that this rope truss would have been unnecessary on Greek or Roman war-ships, ‘which had decking enough to hold the stem and stern together’ (o.c., p. 42)Google Scholar.

^{page 116 note 5} Space does not allow of the detailed consideration of the evidence on which the generally accepted view is based. It may be briefly indicated here: (i.) There is the bronze relief of the forepart of a trireme, of which Adam, gives a photograph (o.c., Vol. II., p. 443)Google Scholar. Some scholars have, however, supposed that the horizontal bands encircling the prow of this ship and interpreted by Adam as hypozomata are mere-ornaments. (ii.) The literary evidence is supplied by Athenaeus V. 37. 2O3E-sqq., and by a comparison of Vitruvius X. 15. 6 with Athenaeus Mechanicus, p. 6. The Tεσσαρακοντ⋯ρησ να⋯σ described in Athenaeus V. 203E had a length of 280 cubits and a breadth of 38 cubits, and it took 12 hypozomata of 600 cubits length each. It is significant that 2x280+38 is roughly equivalent to 600, and we have thus an indication that hypozomata encircled galleys from stem to stern. The battering-ram described in the Vitruvius passage was a long beam of timber tapering to a head or rostrum of hard iron. Lengthwise from the rostrum to the other extremity of the beam were stretched three ropes eight fingers thick ‘ila religati quemadmodum naues a puppi adproram continentur.’ It is significant that Athenaeus Mechanicus in his description of the same ram says: ὑποζζννυται δ⋯ ⋯λος ⋯ κρ⋯ος ὅπλοισ ⋯κταδακτ⋯λοις τρισ⋯. For Mr. Brewster's criticism of the usual interpretation of these passages see his article referred to above.

The question of ‘trapping’—i.e. passing a cable vertically under the hull of a ship—has been considered in relation to the problem of the nature of ὐποζώματα but it is rightly urged, e.g. by Adam, that while ‘frapping’ was undoubtedly known to the ancients, (ὑποζωνν⋯τεσ in Acts XXVII. 17 probably refers to a device of this kind), it was a device employed in an emergency, while the ὐποζώματα of triremes were part of the regular equipment of the vessels. Cf. the arguments adduced by Breusing, , o.c., pp. 172 sqqGoogle Scholar.

^{Page 117 note 1} Proclus, , In Remp. Comm. II., p. 194, 19 sqqGoogle Scholar.(Kroll), gives it as the opinion of some of his predecessors that the light was meant to represent the Milky Way or the circle of the Zodiac. One of these was Numenius of Apamea, see p. 130. It seems probable from a consideration of the Somnium Scipionis (Cic. de Rep. VI. 16) that Cicero cr his authority interpreted Plato's light as the Milky Way.

^{page 117 note 2} See above for ancient opinions that the light represented the axis of the cosmos.

^{page 117 note 3} Aliter Adam, for whom τ⋯ ἄκρα τ⋯ν δεσμ⋯ν come at the centre of the earch (note on 616c 17). But ‘the ends of its bands’ may be interpreted as the ends of the binding—that is, the circular—portion of the light, and so placed at the pole.

^{page 118 note 1} Cf. Zeller, , Die Philosophie der Griechen⁵ I., p. 435, note 2Google Scholar.

^{page 118 note 2} [Philolaos], fr. 12 (Diels, , F.V.S.³ 32B 12)Google Scholar. καί τ⋯ μ⋯ν τ⋯ς σφα⋯ρασ σώματα πἱ;ντε ⋯ντ⋯, τ⋯ ⋯ν τᾷ σφα⋯ρᾳ π⋯ρ ≺κα⋯≻ ⋯δωρ κα⋯ γ⋯ κα⋯ ⋯⋯κα⋯ τ⋯ς σφα⋯ρας ⋯λκ⋯σ π⋯μπτον. The so-called fragments of Philolaos and the opinions attributed to him have, of course, to be used with great caution for evidence of early Pythagorean doctrine. But the word ⋯λκ⋯σ may safely be said to be a trace of very ancient Pythagorean terminology. See Burnet, , E.G.P.³, p. 293Google Scholar. He interprets σώματα in this passage as=regular solids and the fifth σώμα as the dodecahedron, which, being used for the construction of the whole universe (Plato, Tim. 55c), is thus termed the ‘hull of the sphere.’ Gundermann, , Rhein. Mus. N.F. LIX., p. 145Google Scholar, explains σώματα as ‘bodies’ in the sense of ‘elements,’ and the fifth element, which is the ship of the sphere, is identified by him with αἰθ;ἠρ (cf. Diels, , F. V S.³, on [Philolaos] fr. 12)Google Scholar. The significance of the term ⋯λκ⋯σ is not affected by the difference of opinion about the exact interpretation of the passage.

^{page 118 note 3} Aet. II . 4. 15 (F. V.S.³ 32A 17) (opinion attributed to Philolaos), τ⋯ δ⋯ ⋯γεμονικ⋯ν ⋯ν τῷ μεσαιτ⋯τῳ πνρ⋯, δπερ τρ⋯πεωσ δ⋯κην προυπεβ⋯λετο τ⋯σ το⋯ παντ⋯ς σφα⋯ρασ≻ ⋯ δημιου γ⋯ς θ;ε⋯ς. Zeller⁵, p. 416, note 1, ‘das ⋯γεμονικ⋯ν stoisch und der Demiurg platonisch ist, aber die Vergleichung des Centralfeuers mit dem Kiel des Weltganzen doch ursprünglich scheint.’

^{page 118 note 4} E.G.P.³, P. 111 and pp. 297 sqq.

^{page 118 note 5} De Cael. II. 13. 293A 20 sqq., ⋯ναντιως ο⋯ περ⋯πτ⋯ν 'Iταλ⋯αν, καλο⋯μενοι δ⋯ Πυθ;αγἱ;ρειοι λ⋯γοσιν. ⋯π⋯ μ⋯ν γ⋯ρ το⋯ μ⋯σου π⋯ρ εἶνα⋯ φασι, τ⋯ν εν τῷν ἄστρων ο⋯σαν, κ⋯κλψ φερομ⋯νην περ⋯ τ⋯ μ⋯σονν ⋯κτα τε κα⋯ ⋯μ⋯ραν ποιε⋯ν, etc. This system is ascribed to philolaos in Aet. II. 7. 7 (Diels, , F. V.S.³ 32A 16)Google Scholar.

^{page 119 note 1} Heiberg, pp. 511 sqq.

^{page 119 note 2} Zeller⁵, I., pp. 420 sqq.

^{page 119 note 3} This is pointed out by SirHeath, T., Aris tarchus of Samos. p. 250Google Scholar: ‘The earth in the system described by Simplicius is not in motion, but at rest. For Simplicius, so far from implying that the earth rotates, thinks it necessary to explain how the Pythagoreans to whom he refers could, notwithstanding the earth's immobility, call it a “star,” and count it, exactly as Plato does, among the “instruments of time.”’.

^{page 119 note 4} Plato, , Tim. 40c 1Google Scholar.

^{page 119 note 5} The doctrine that the moon is the antichthon, which is apparently ascribed to the same Pythagoreans, looks a late one. But this need not necessarily show that the doctrine of the fire in the earth's interior is late also. Later generations of Pythagoreans might easily have combined the original doctrine of fire inside a central, spherical earth with new-fangled notions about the antichthon. In the view which I hold to be the original Pythagorean one, that of a central fire in the bowels of a spherical earth situated in the centre of the cosmos, there is nothing to correspond to the antichthon but the antipodes (Alex, . Polyhistor. ap. Diog. Laert. VIII. 25Google Scholar: The Pythagoreans taught γ⋯νεσθ;αι … κ⋯σμον ἔμφυχον, νοερ⋯σφαιροειδ⋯, μ⋯σην περι⋯χοντα τ⋯ν κα⋯ αὐτ⋯ σφαιροειδ⋯ κα⋯ περιοικονμ⋯νην, εἶναι δ⋯ κα⋯ ⋯ντ⋯ποδασ κα⋯ τ⋯ ⋯μῖν κ⋯τω ⋯κε⋯νοις ἄνω). Is i unreasonable to suppose that the later notion of the separate antichthon developed from the idea of antipodes?

^{page 120 note 1} Fr. 52 (F.V.S.³ 21B 52), and compare fr. 62 (F.V.S.³ 21B 62).

^{page 120 note 2} Fr. 615 (Pearson) (Philodemus de piet, p. 23): κα⋯ Σοφοκλ⋯ς ⋯<ν 'Iν⋯>χφ τ⋯ν γ⋯ν μ<ητ⋯>ρα τ⋯ν θ;ε⋯ν φη<σ⋯ν>, ⋯ν Tριπτολ⋯μψ δ⋯ κα⋯ ‘Eστ⋯ανεἶναι.

Euripides, fr. 944 (Nauck²):

κα⋯ Tαῖα μ⋯τερ ‘Eστιαν δ⋯’ οἱ; σοφο⋯

βροτ⋯ν καλο⋯σιν ⋯μ⋯νην ⋯ν α⋯θ;⋯ρι.

^{page 120 note 3} See Martin, , Mémoires de l' Académie des Inscriptions et Belles Lettres XXVIII., Pt. I., pp. 335 sqqGoogle Scholar. Martin, following Nägelsbach, attributes the identification to the Orphics, p. 349. Cf. οἱ; σοφο⋯ βροτ⋯ν in the fragment of Euripides.

^{page 120 note 4} P. 30, Heiberg, (Annales intern. d'Histoire, 1900)=Theol. Arith., p. 6Google Scholar, Ast (Diels, , F.V.S³ 18A 44)Google Scholar.

^{page 120 note 5} E.G.P.³, pp. 108 sqq.

^{page 120 note 6} Hippasos made Fire the first principle according to Aristotle, , Met. A. 984A 7Google Scholar. Moreover, Light and Darkness appear under the heads of Limit and Unlimited respectively in the Pythagorean table of opposites (Met. A. 986A 25).

^{page 121 note 1} Cf. Ross, note ad loc. He identifies the One with the Limit here.

^{page 121 note 2} Cf. Ross, notead loc.: ‘The One is thought of as being in the centre of a shapeless mass of air or vapour and gradually introducing shape and limit into it, working from within outwards.’.

^{page 121 note 3} Aristotle, De Cael. 295A 13 sqq.; Ps. Plut. strom., fr. 10 (Diels, , Dox. Gracec., p. 582)Google Scholar.

^{page 121 note 4} Aet. III. 15. 7; Diog. Laert. IX. 21.

^{page 121 note 5} See what is said above about the fire in the earth in Empedocles' doctrine. The reference to Parmenides in the passage from Anatolius is to the cosmology in the second part of Parmenides' poem. I am reserving the discussion of this cosmology to a later place. It is significant that in the Anatolius passage the words immediately following those quoted above are: κα⋯ δ⋯ Eὐριπ⋯δησ ὡσ ‘Aναεαγ⋯ρου γεν⋯μενοσ μαθ;ητ⋯ς οὔτω τ⋯σ γ⋯σ μνησθ;ε⋯σ ‘⋯στ⋯αν δ⋯ σ' ο⋯; αοφο⋯ βροτ⋯ν νομ⋯ ζονσιν.’.

^{page 122 note 1} Diels, F.V.S.³ 32A 16.

^{page 122 note 2} It does not seem possible to draw any certain conclusions from Aet. I. 14. 2, οἱ ⋯π⋯ Πνθ;αγ⋯ρον σφαιρικ⋯ τ⋯ σχ⋯ματα τ⋯ν τεττ⋯ρων στοιχε⋯ων, μ⋯νον δ⋯ τ⋯ ⋯νώτατον π⋯ρ κωνοειδ⋯σ, or from Aet. II. 20. 12, reflection. Cf. Burnet, , E.G.P.³, p. 298 noteGoogle Scholar.

^{page 122 note 3} Ancient tradition relates that Parmenides had associated with the Pythagorean Ameinias or that he was a Pythagorean. For the evidence see Burnet, , E.G.P.³, p. 170Google Scholar. Professor Burnet's view is that the second part of the poem is ‘a sketch of contemporary Pythagorean cosmology.’ MrCornford, F. M., From Religion to Philosophy, p. 217Google Scholar, ‘the Pythagorean character of the Way of Opinion is recognized.’ Zeller, , o.c., p. 572Google Scholar, grants the existence of Pythagorean doctrine in the Way of Opinion. Cf. Gomperz, , Greek Thinkers (Eng. trans.) I., p. 182Google Scholar.

^{page 123 note 1} Simp. Phys. (Diels) 39. 12 and 31 10=Parmendies, fr. 12; F.V.S.³ 18B 12.

^{page 123 note 2} F.V.S.³ 18B IO, 1. 5 sqq.

^{page 123 note 3} Bumet, O.C., pp . 189–190Google Scholar.

^{page 123 note 4} Simp. Phys. 39. 17 (continuation of the passage referred to above), τα⋯την κα⋯ θ;ε⋯ν αἰτ⋯αν εἶνα⋯ φησι λ⋯γων ‘πρώτιστον μ⋯ν ῎Eρωτα θε⋯ν μητ⋯σατο π⋯ντω’ … κα⋯ τ⋯σ ψυχ⋯ς π⋯μπειν ποτ⋯ μ⋯ν ⋯κ το⋯ ⋯μφανο⋯σ, etc. (F.V.S.³ 18B 13).

^{page 123 note 5} Cf. Burnet, , E.G.P.³, p. 190Google Scholar. He accepts the identification, as also does Gilbert, , ‘Die δα⋯μων des Parmenides’ in A rchiv. für Geschichte der Philosophie, N.F. XIII., pp. 25 sqqGoogle Scholar. Cf. Cornford, F. M., o.c., p. 222Google Scholar.

^{page 123 note 6} ‘´Hντινα κα⋯ δα⋯μοα κνβερν⋯τιν κα⋯ κληρο⋯χον ⋯πονομ⋯ζει δ⋯κην τε κα⋯ 'Aν⋯γκην, cf. Aet. I. 25. 3. Παρμεν⋯δης κα⋯ δημ⋯κριτοσ π⋯ντα κατ’ ⋯ν⋯γκην. τ⋯ν αὐτ⋯ν δ⋯ εἶναι εἱ;μαρμ⋯νην κα⋯ δ⋯κην κα⋯ πρ⋯νοιαν κα⋯ κοσμ⋯ποιον.

^{page 124 note 7} The third interchangeable term is Mοῖρα, cf. 1. 37 of the same fragment.

^{page 124 note 8} Gilbert, , o.c., makes the second identification. Diels, (Parmeindes Lehrgedicht, P. 51)Google Scholar separates δ⋯κη πολ⋯ποινος from the goddess of the proem; δ⋯κη is only the priestess of the temple of light. Neither does he identify the goddess of the proem and the δσ⋯μων κνβερν⋯τισ.

^{page 124 note 1} That the noun to be supplied with στειν⋯τεραι is στεφ⋯ναι is clear from Aet. II. 70 1: Παρμεν⋯δησ στεφ⋯νασ εἶναι περιπεπλεγμ⋯νας, ⋯παλλ⋯λους, τ⋯ν μ⋯ν ⋯κ το⋯ ⋯ραιο⋯, τ⋯ν δ⋯ ⋯κ το⋯ πνκνο⋯. μικτ⋯σ δ⋯ ἄλλα ⋯κ φωτ⋯ς κα⋯ σκ⋯τονς μεταξὺ το⋯των. The reference is to rings or bands of light or darkness or light and darkness mixed, which are conceived as running round the sky and encircling the central earth. See Burnet, , E.G.P.³, pp. 187 sqqGoogle Scholar.

^{page 124 note 2} See references above.

^{page 124 note 3} Burnet, , E.G.P.^sup, p. 189Google Scholar; ct. Gilbert, , o.c., p. 42Google Scholar. I have shown above that there is no objection to the view that the early Pythagoreans conceived of their central, spherical earth as containing a core of fire.

^{page 124 note 4} Aet. II. 7. 7, where the names given to the central fire by [Philolaos] are given as ⋯στ⋯;δι⋯ς οἶκομητ⋯ρ θ;ε⋯ν, βωμῸς κα⋯ συνοχ⋯ κα⋯ μ⋯τρον φ⋯σεωσ. These names are surely to be regarded as early and not as belonging to the late Pythagoreans only. Their character attests their antiquity.

^{page 124 note 5} It is interesting to notice that in Orphic Hymn XXVII. (Abel) to Rhea the mother of the gods, who is in that hymn identified with Hestia, the language is in the highest degree reminiscent of that used of the δα⋯μων κυβερν⋯τις in Parmenides' poem:

'Aθ;αν⋯των θε⋯τιμε θ;ε⋯ν μ⋯τετρ⋯φε π⋯ντων, τ⋯δε μ⋯λοις, κρ⋯ντειρα θε⋯, σ⋯ο, π⋯τνι', ⋯π' εὐχαῖ…

αὐτ⋯

γαῖαν ἔχεις θνηοῖσι τροφ⋯ς παρ⋯χουσα προσηνεῖς.

⋯κ σ⋯ο δ' ⋯θαν⋯των τε γ⋯νος θνητ⋯ν τ' ⋯λοχε⋯θη …

'Iστ⋯η αὐδαχθεῖσα …

The two equations Rhea=Earth and Hestia =Earth seem to have operated here.

^{page 124 note 6} T⋯ν δ⋯ σνμμιγ⋯ν τ⋯ν μεσαιτ⋯την ⋯π⋯σαις <⋯ρχ⋯ν> τε κα⋯ <α⋯τ⋯αν> κιν⋯σεως κα⋯ γεν⋯σεωσ ὑπ⋯ρχειν, ἧντινα κα⋯ δα⋯μονα κυβερν⋯τιν κα⋯ κληρο⋯χον ⋯πονομ⋯ξει δ⋯κην τε κα⋯ 'Aν⋯γκην.

^{page 124 note 7} I. 11. 28 (F.V.S.³ 18A 37).

^{page 125 note 1} Cf. Iambl, . Theol. Arith. 60Google Scholar: τ⋯ν 'Aν⋯γκην οὺθε⋯λογοι τῇ το⋯ παντ⋯ς οὐρανο⋯ ⋯ξωτ⋯τῃ ἄντυγι ⋯πηχο⋯σι.

^{page 125 note 2} I.c., pp. 27 sqq.

^{page 125 note 3} In Parmenides Lehrgedicht, p. 107. In the note on p. 161 in F.V.S.³ he simply states two views—(a) that of Berger, who placed her in the sun and (b) that of Simplicius and Gilbert.

^{page 125 note 4} The character of the Milky Way corresponds to the description given of the mixed crowns or the mixed bands in Aet. II. 7. 1 (μικτ⋯ς δ⋯ ἄλλας ⋯κ φωτ⋯ς κα⋯ σκ⋯τους μεταξὺ το⋯των); cf. Aet. III. 1. 4, Παρμεν⋯δης τ⋯ το⋯ πυκνο⋯ κα⋯ το⋯ ⋯ραιο⋯ μῖγμα γαλακτοειδ⋯ς ⋯ποτελ⋯σαι χρ⋯μα. Aet. II. 20. 8A would suppot the statement that the Milky Way was intermediate between sun and moon. Burnet inclines not to believe in the’ which he thinks arise from a confused interpretation of fr. 12 by the authority responsible for the statement in Aet. II. 7. I. But he says (p. 191): ‘Whether we believe in the “mixed bands” or not makes no difference …; for the statement of Aetios that she was in the middle of the mixed bands undoubtedly implies that she was between earth and heaven.’

^{page 125 note 5} Cf. MrCornford's, F. M. illuminating remarks in From Religion to Philosophy, P. 222 and notesGoogle Scholar.

^{page 126 note 1} E.g., according to Heracleitus the life of the human soul depended on the maintenance in it of the due measures of fire. See Diels, , F.V.S.³ 12B 36Google Scholar (cf. 12B 26 and 118, and Burnet, , E.G.P.³, p. 151)Google Scholar. For the part played by fire in producing life in Empedocles' doctrine see Diels, , F.V.S.³ 21B 62Google Scholar.

^{page 126 note 2} l.c., I., pp. 416 sqq.

^{page 126 note 3} De Caelo II. 13. 293A 20 sqq., cited above.

^{page 126 note 4} Cf. τῇ ⋯ξωτ⋯τῃ ἄντυγι in the passage from Iamblichus quoted above.

^{page 126 note 5} Cf. the crystalline vault of Empedocles, Aet. II. 11. 2.

^{page 126 note 6} These are the circles, the fire bursting out from which makes the Morning Star, the Sun, the Milky Way, the other stars, and the Moon. Aet. II. 7. 1 compared with II. 15. 7, II. 20. 8A, III. I. 4. See Diels, . F.V.S.³, p. 161 noteGoogle Scholar.

^{page 127 note 1} The intervals between the wheels of fire in heaven may have been identified with the musical intervals of the scale (Burnet, , E.G.P.³, p. 110)Google Scholar. Burnet suggests ad loc. that the doctrine of the ‘harmony of the spheres’ began in some such way. If, as has been suggested recently by MrCornford, (C.Q. XVI., pp. 145 sqq.)Google Scholar, the doctrine that soul is a harmony belongs to early Pythagoreanism, then the fiery world soul may have been regarded as constituting an ⋯ρμον⋯α

^{page 127 note 2} E.G.P.³, p. 191.

^{page 127 note 3} The theory of wheels or bands goes back, of course, to Anaximander. Burnet, suggests (E.G.P.³, p. 188)Google Scholar that Pythagoras adopted the theory from him.

^{page 127 note 4} Cf. Zeller⁵ I., p 435.

^{page 127 note 5} Porphyry, . de antro nymph. 28Google Scholar. Pythagoras is given as authority for the statement that souls συν⋯γεσθαι εἰς τ⋯ν γαλαξ⋯αν. Cf. the view of Heracleides of Pontus, a Pythagorizing Platonist (Stob, . Ecl. I. 41. 39)Google Scholar.

But the notion that the Milky Way is the place or path of souls is probably older even than Pythagoras and the Pythagoreans. Cf. the instances of the occurrence of the idea amon primitive peoples given in Tylor, , Primitive Civilization⁴ I., p. 359Google Scholar, and DrCook, A. B., Zeus II., pp. 37 sqqGoogle Scholar.

^{page 127 note 6} Simpl, . Phys. 39. 17 quoted aboveGoogle Scholar.

^{page 127 note 7} The spindle revolves ⋯ν τタῖς τ⋯ς 'Aν⋯γκης γ⋯νασιν (617B). This might suggest that the whole heaven is embraced and surrounded by 'Aν⋯γκη. But see Adam's noted ad loc. for another interpretation of the phrase, which it must be confessed lacks explicitness if an exact statement of 'Aν⋯γκη's position in the universe is sought from it. Again the souls pass ultimately ὑπ⋯ τ⋯ν τ⋯ς 'Aν⋯γκης θρ⋯νον (621A), but their whereabouts at this point in the story is not clear.

^{page 128 note 1} Note on 616B 14 ad fin.

^{page 128 ntoe 2} Cf. Timaeus 36D sqq., especially the words ⋯ δ' (i.e. ⋯ ψυχ⋯) ⋯κ μ⋯σου πρ⋯ς τ⋯ν ἔσχατον οὐραν⋯ν π⋯ντῃ διαπλακεῖσα κὑκλῳ τε αὐτ⋯ν ἔξωθεν περικαλὑψασα …

Cf. again the language used of 'Aν⋯γκη in the Orphic Theogony of Hieronymus and Hellanicus (Lobeck, , Aglaophamus, p. 485)Google Scholar, διωργυιωμ⋯νην ⋯παντ⋯ τῷ, τ⋯ν περ⋯των αὐτο⋯ ⋯φαπτομ⋯νην, I owe this last passage to DrCook, A. B. (see Zeus II., Part II., p. 1022)Google Scholar.

^{page 128 note 3} Zeller⁴ II., pp. 1034 sqq.

^{page 128 note 4} Cf. Philoponus, , De an., quoted in Diels, , Dox. Graec., p. 214Google Scholar, and Plut, . de lat. uiu. VI. 1130BGoogle Scholar.

^{page 128 note 5} Cf. Zeller⁵ I., p. 435. and Adam, note on 616B sub fin.

^{page 129 note 1} Pp. 44 sqq. I should like to express here my great sense of obligation to DrCook, , who put at my disposal part of Volume II. of Zeus, when it was still in proof, in 06, 1923Google Scholar.

^{page 129 note 2} Zeus II., p. 169.

^{page 130 note 1} Cf. ib. 746 sqq.; Aesch. P.V. 430; Eur. Ion 1; Apollodorus, , Bibl. I. 8. 21 (Wagner)Google Scholar, See Pauly-Wissowa, , art. Atlas, p. 2122Google Scholarfin..

^{page 130 note 2} Eχει=upholds, supports; cf. Butcher and Lang translation ad loc., Merry and Riddell note ad loc. But an alternative interpretation is ‘guards.’ So Gruppe, O., Griechische Mythologie, p. 382Google Scholar; art. Atlas in P.-W., p. 2123. But ⋯ρειδων in the Aeschylus passage is in favour of the former interpretation.

^{page 130 note 3} The variation between the singular ‘pillar’ and the plural ‘pillars’ in the Atlas tradition is probably to be explained by a twofold notion of the supports of heaven. There is first the notion of a single sky-prop, for which the proper place is the centre of the flat earth. There is also the notion of heaven supported at its extremities on pillars four in number (Orphic Eὐχ⋯ πρ⋯ς Mουσαῖον 39. κ⋯σμου τε μ⋯ρη τετρακ⋯ονος αὐδ⋯, cf. Ibycus ap. schol. on Ap. Rhod. III. 106); cf. Zeus II., pp. 125 sqq., p. 56, note 2. DrCook, (op. cit., pp. 140 sqq.)Google Scholar holds that belief in a sky resting on four pillars is not inconsistent with belief in one central prop. Both notions of the supports of heaven seem reflected in the Atlas legend. For not only does he in one instance support a single, in the other support several pillars, but he himself, while generally located at the extremities of the earth (Hes, . Theog. 518, 746 sqq.Google Scholar; Aesch, . P.V. 350Google Scholar; Eur, . Hipp. 742 sqq.Google Scholar; Verg, . Aen. IV. 481, VI. 795 sqq.Google Scholar; Apollois dorus II. 120 [Wagner]), was in some versions at any rate placed in the middle of the earth under the central point of the overhanging heavens (Eur, . Here. Fur. 403 sqqGoogle Scholar.; cf. the island of Calypso, Atlas' daughter, described in Od. I. 50 as situated ὅθ;ι ⋯μφαλ⋯ς ⋯στι θ;αλ⋯σσης.

^{page 130 note 4} P.-W., p. 2123; cf. Gruppe, , op. cit., p. 382Google Scholar. When the stage of rationalization of legend is reached the blending of the two ideas is easy.Atlas is interpreted as a ofty mountain in North-West Africa, and the mountain is called κ⋯ων το⋯ οὐρανο⋯ (Hdt. IV. 184).

^{page 131 note 1} For evidence of the same interpretation see Scholia, on Hes. Theog. 507, 509, 517Google Scholar, Aesch, . P.V. 428Google Scholar and Eur, . Hipp. 747Google Scholar; and compare Hesychins, ⋯τλασ· ⋯τολμος, ⋯παθ;⋯ς, κα⋯ ⋯ δι ι ο ⋯ σ α ε ὐ θ; ɛ ῖ α ἕως τ⋯ν π⋯λωérature Pythagoricienne, p. 124. where he quotes the passage from Eustathius which is given below

^{page 131 note 3} 1389. 59.

^{page 131 note 4} Cf. Procl, . in Remp. comm. II., p. 200, 5 sqq., KrollGoogle Scholar.

^{page 131 note 5} In view of DrCook's, most interesting thesis that the omphalos at Delphi, which marked the centre of the earth, was originally topped by a pillar symbolizing the sky-god and representing the central support of the sky (Zeus II., pp. 169 sqq.)Google Scholar, a passage from the myth of Thespesius of Soli in Plutarch, , De ser. num. uind, 566DGoogle Scholar, raises some interesting conjectures. This myth is on many grounds comparable with the myth of Er, and like it is full of Orphic-Pythagorean doctrine. Now when Thespesius, whose wanderings in the world of the dead appear to be entirely aerial and celestial (Stewart, , Myths of Plato, p. 379)Google Scholar, had seen the plain of Lethe and the great mixing-bowl which his guide informed him was the oracle of Night and the Moon, he was not able to rise yet further and view the true oracle of Apollo situated higher in the heavens; but his guide tried to show him, though without success, τ⋯ φ⋯ς ⋯κ το⋯ το⋯ τρ⋯ποδος … (i.e. the tripod of the celestial oracle of Apollo) δι⋯ τ⋯ν κ⋯λπων τ⋯ς Θ;⋯μιδος ⋯περειδ⋯μενοε⋯ς τ⋯ν Π⋯ρνασον. The light reaches down from the sky, and passing through the womb of Themis is set in Parnassus—i.e. in the earthly Delphi. In view of the general similarity between the two myths, and in view further of the interest of the Pythagoreans in Delphi, is it too bold to suppose that the light here and the straight light like a pillar in the myth of Er are one and the same, and that both conceptions go back ultimately to the notion of the sky-god's column which stretches from the heaven, which it supports, down to earth, in which its base is set?

^{page 132 note 1} Cf. again Timaeus 34B : ψυχἠν ε⋯ς τ⋯ μ⋯σον αὐτο⋯ θ;ε⋯ς δι⋯ παντ⋯ς τε ἔτεινεν κα⋯ ἔτι ἔξωθ;εν τ⋯ σ⋯μα αὐτῇ περιεκ⋯λυψεν.

^{page 132 note 2} The channel by which it passes is χ⋯σμα διαμπερ⋯ς τετρημ⋯νον like the χ⋯σμα which forms Plato's Tartarus in the Phaedo myth 112A.

^{page 132 note 3} Nicomachus of Gerasa, , ap. Phot, bibl., p. 143A 30 sqqGoogle Scholar. … κα⋯ Aτλαντα (αὐτ⋯ν τερατολογο⋯σι).⋯ξων τ⋯ ⋯στιν κα⋯ ἤλιος κα⋯ πυρ⋯λιος κα⋯ μορφὼ δ⋯ κα⋯ ζαν⋯ς πὑργος, κα⋯ σπερματ⋯της λ⋯γος, 'Aπ⋯λλων τε κα⋯ προφ⋯της κα⋯ λ⋯γιος. Delette says with regard to this reference (op. cit,, p. 143), ἅξων désigne en effet le même être qu'Atlas,’ quoting the passage from Eustathius given above.

^{page 132 note 4} Fr. 2O4 (Rose), from Simpl, . comm. in De Cael. (Heiberg, p. 51)Google Scholar. Note that this occurs in the same passagewhich vouches for the Pythagorean doctrine of a central fire hidden in the heart of a central earth.

^{page 132 note 5} Arist, . Dt Cael. II. 13. 293b 2Google Scholar.

^{page 132 note 6} Simpl. I.c.

^{page 132 note 7} Aet. II. 7.7 (opinions of Philolaos).

^{page 133 note 1} See Preuner, in art. Hestia in Roscher, , p. 2620Google Scholar, in the section where Hestia is discussed in her capacity as goddess of the sacred fire: ‘Ohne Zweifel haben wir es hier mit einer indogermanischen religiösen Grundanschauung zu thun, wonach das Feuer in der aufsteigenden Flamme und dem zum Himmel aufwirbelnden Rauch die Gaben der Menschen, die in ihm verbrannt wurden, zu den Himmlischen, vor allem zum Himmelsgott selbst, zu Zeus, hinaufzutragen scheint.’ Might this conception of the flame mounting into the sky have been transferred to the Pythagorean Hestia or Central Fire?

^{page 133 note 2} An interesting parallel may in conclusion be noted. Dr. A. B. Cook has called my attention to the striking likeness between the Pythagorean circumambient Ananke and the Egyptian skygoddess Nut, who is represented as with her own body forming the arch of the sky (see fig. 34 in Erman, A., Handbook of Egyptian Religion (trans. Griffiths, ), p. 29Google Scholar; and Lanzone, , Dizionario di Mitologia Egizia, Tav. CLVI. sqq.)Google Scholar. Moreover, Egyptian mythology said that originally ‘Nut still lay upon her brother Keb (the earth). Therefore her father Shu thrust himself between them and raised her into the heights, and with her he raised into the heights all the gods that had hitherto been created, and Nut took possession of them, counted them, and made them into stars’ (Erman.I.c.). If Nut resembles Ananke, then Shu resembles Atlas, and we seem to have an extraordinarily close parallel to the Pythatogorean conception which united the goddess of circumambient fire with Atlas and his pillar.