556. I have translated Diels’s conjecture ἀπεστέγασεν δέ οἱ αὐγάς, | ἔστ’ ἂν ἴῃ καθύπερθεν. The MSS. have ἀπεσκεύασεν and ἔστε αἶαν.

557. Reading στείροις with Diels, Hermes, xv. loc. cit.

558. Retaining εἴδεος (i.e. ἴδεος), which is read in the MSS. of Simplicius. Cf. above, p. 243, n. 549.

559. That Empedokles regarded milk as putrefied blood is stated by Aristotle (de Gen. An. Δ, 8. 777 a 7). The word πύον means pus. There may be a punning allusion to πυός, “beestings,” but that has its vowel long.

560. Said of women in reference to births in the seventh and ninth months.

561. Of the membrane round the fœtus.

562. Reading ἴδεα ποιπνύουσα with Diels.

563. See Beare, p. 16, n. 1, where Plato, Tim. 45 b 4 (τοῦ πυρὸς ὅσον τὸ μὲν κάειν οὐκ ἔσχεν, τὸ δὲ παρέχειν φῶς ἥμερον), is aptly quoted. Alexander ad loc. understands κατὰ βηλόν to mean κατ’ οὐρανόν, which seems improbable.

564. On fr. 99, see Beare, p. 96, n. 1.

565. This passage is quoted by Aristotle (de Respir, 473 b 9), who makes the curious mistake of taking ῥινῶν for the genitive of ῥίς instead of ῥινός. The locus classicus on the subject of the klepsydra is Probl. 914 b 9 sqq. (where read αὐλοῦ for ἄλλου, b 12). The klepsydra was a metal vessel with a narrow neck (αὐλός) at the top and with a sort of strainer (ἠθμός) pierced with holes (τρήματα, τρυπήματα) at the bottom. The passage in the Problems just referred to attributes this theory of the phenomenon to Anaxagoras, and we shall see later that he also made use of a similar experiment (§ 131).

566. This seems to be the experiment described in Probl. 914 b 26, ἐὰν γάρ τις αὐτῆς (τῆς κλεψύδρας) αὐτὴν τὴν κωδίαν ἐμπλήσας ὕδατος, ἐπιλαβὼν τὸν αὐλόν, καταστρέψῃ ἐπὶ τὸν αὐλόν, οὐ φέρεται τὸ ὕδωρ διὰ τοῦ αὐλοῦ ἐπὶ στόμα. ἀνοιχθέντος δὲ τοῦ στόματος, οὐκ εὐθὺς ἐκρεῖ κατὰ τὸν αὐλόν, ἀλλὰ μικροτέρῳ ὕστερον, ὡς οὐκ ὂν ἐπὶ τῷ στόματι τοῦ αὐλοῦ, ἀλλ’ ὕστερον διὰ τούτου φερόμενον ἀνοιχθέντος. The epithet δυσηχέος applied to ἰσθμοῖο is best explained as a reference to the ἐρυγμός or “belching” referred to at 915 a 7 as accompanying the discharge of water through the αὐλός. Any one can produce this effect with a water-bottle. If it were not for this epithet, it would be tempting to read ἠθμοῖο for ἰσθμοῖο. Sturz conjectured this, and it is actually the reading of a few MSS.

567. On fr. 101, see Beare, p. 135, n. 2.

568. That the reference is to dreams, we learn from Simpl. de An. p. 202, 30.

569. Bernays conjectured ῥῆμα, “decree,” for χρῆμα, but this is not necessary. Necessity is an Orphic personage, and Gorgias, the disciple of Empedokles, says θεῶν βουλεύμασιν καὶ ἀνάγκης ψηφίσμασιν (Hel. 6).

570. I retain φόνῳ in v. 3 (so too Diels). The first word of v. 4 has been lost. Diels suggests Νείκεϊ, which may well be right, and takes ἁμαρτήσας as equivalent to ὁμαρτήσας. I have translated accordingly.

571. According to Porphyry, who quotes this line (de Antro Nymph. 8), these words were spoken by the “powers” who conduct the soul into the world (ψυχοπομποὶ δυνάμεις). The “cave” is not originally Platonic but Orphic.

572. This passage is closely modelled on the Catalogue of Nymphs in Iliad xviii. 39 sqq. Chthonie is found already in Pherekydes (Diog. i. 119).

573. I have retained ἀλλόγνωτι as nearer the MSS., though a little hard to interpret. On the subsequent history of the Orphic chiton in gnostic imagery see Bernays, Theophr. Schr. n. 9. It was identified with the coat of skins made by God for Adam.

574. This is the best μετοίκησις (Ael. Nat. an. xii. 7).

575. The dwellers in the Golden Age.

576. The MSS. of Porphyry have γραπτοῖς τε ζώοισι, which is accepted by Zeller and Diels. The emendation of Bernays (adopted in R. P.) does not convince me. I venture to suggest μακτοῖς, on the strength of the story related by Favorinus (ap. Diog. viii. 53) as to the bloodless sacrifice offered by Empedokles at Olympia.

577. These lines were already referred to Pythagoras by Timaios (Diog. viii. 54). As we are told (Diog. ib.) that some referred the verses to Parmenides, it is clear that no name was given.

578. On frs. 138 and 143 see Vahlen on Arist. Poet. 21. 1547 b 13, and Diels in Hermes, xv. p. 173.

579. Cf. Emp. frs. 27, 28, with Parm. fr. 8.

580. For the history of the term στοιχεῖον see Diels, Elementum. Eudemos said (ap. Simpl. Phys. p. 7, 13) that Plato was the first to use it, and this is confirmed by the way the word is introduced in Tht. 201 e. The original term was μορφή or ἰδέα.

581. Cf. Chap. I. § 27.

582. Arist. Phys. Δ, 6, 213 a 22 (R. P. 159). Aristotle only mentions Anaxagoras by name in this passage; but he speaks in the plural, and we know from fr. 100 that the klepsydra experiment was used by Empedokles.

583. In antiquity the Homeric Allegorists made Hera Earth and Aidoneus Air, a view which has found its way into Aetios from Poseidonios. It arose as follows. The Homeric Allegorists were not interested in the science of Empedokles, and did not see that his αἰθήρ was quite a different thing from Homer’s ἀήρ. Now this is the dark element, and night is a form of it, so it would naturally be identified with Aidoneus. Again, Empedokles calls Hera φερέσβιος, and that is an old epithet of Earth in Homer. Another view current in antiquity identified Hera with Air, which is the theory of Plato’s Cratylus, and Aidoneus with Earth. The Homeric Allegorists further identified Zeus with Fire, a view to which they were doubtless led by the use of the word αἰθήρ. Now αἰθήρ certainly means Fire in Anaxagoras, as we shall see, but there is no doubt that in Empedokles it meant Air. It seems likely, then, that Knatz is right (“Empedoclea” in Schedae Philologicae Hermanno Usenero oblatae, 1891, pp. 1 sqq.) in holding that the bright Air of Empedokles was Zeus. This leaves Aidoneus to stand for Fire; and nothing could have been more natural for a Sicilian poet, with the volcanoes and hot springs of his native island in mind, than this identification. He refers to the fires that burn beneath the Earth himself (fr. 52). If that is so, we shall have to agree with the Homeric Allegorists that Hera is Earth; and there is certainly no improbability in that.

584. Arist. de Gen. Corr. Β, 1. 329 b 1.

585. Ibid. Β, 6. 333 a 16.

586. Ibid. Α, 8. 325 b 19 (R. P. 164 e). This was so completely misunderstood by later writers that they actually attribute to Empedokles the doctrine of στοιχεῖα πρὸ τῶν στοιχείων (Aet. 1. 13, 1; 17, 3). The criticism of the Pythagoreans and Plato had made the hypothesis of elements almost unintelligible to Aristotle, and a fortiori to his successors. As Plato put it (Tim. 48 b 8), they were “not even syllables,” let alone “letters” (στοιχεῖα). That is why Aristotle, who derived them from something more primary, calls them τὰ καλούμενα στοιχεῖα (Diels, Elementum, p. 25).

587. We know from Menon that Philistion put the matter in this way. See p. 235, n. 527.

588. Arist. Met. Α, 4. 985 a 31; de Gen. Corr. Β, 3. 330 b 19 (R. P. 164 e).

589. Cf. Introd. § VIII.

590. Arist. Met. Α, 10. 1075 b 3.

591. Theophr. Phys. Op. fr. 3 (Dox. p. 477); ap. Simpl. Phys. p. 25, 21 (R. P. 166 b).

592. Arist. Met. Α, 4. 985 a 21; Γ, 4. 1000 a 24; b 9 (R. P. 166 i).

593. Plato, Laws, x. 889 b. The reference is not to Empedokles exclusively, but the language shows that Plato is thinking mainly of him.

594. Arist. de Gen. Corr. Β, 6. 334 a 1; Phys. Θ, 1. 252 a 5 (R. P. 166 k).

595. Ibid. Α, 8. 324 b 34 (R. P. 166 h).

596. Arist. de Gen. Corr. 326 b 6.

597. This is the view of Zeller (pp. 785 sqq.), but he admits that the external testimony, especially that of Aristotle, is wholly in favour of the other. His difficulty is with the fragments, and if it can be shown that these can be interpreted in accordance with Aristotle’s statements, the question is settled. Aristotle was specially interested in Empedokles, and was not likely to misrepresent him on such a point.

598. Arist de Gen. Corr. Β, 6. 334 a 6: τὸν κόσμον ὁμοίως ἔχειν φησίν ἐπί τε τοῦ νείκους νῦν καὶ πρότερον ἐπὶ τῆς φιλίας.

599. Arist. de Caelo, Γ, 2. 301 a 14: ἐκ διεστώτων δὲ καὶ κινουμένων οὐκ εὔλογον ποιεῖν τὴν γένεσιν. διὸ καὶ Ἐμπεδοκλῆς παραλείπει τὴν ἐπὶ τῆς φιλότητος· οὐ γὰρ ἂν ἠδύνατο συστῆσαι τὸν οὐρανὸν ἐκ κεχωρισμένων μὲν κατασκευάζων, σύγκρισιν δὲ ποιῶν διὰ τὴν φιλότητα· ἐκ διακεκριμένων γὰρ συνέστηκεν ὁ κόσμος τῶν στοιχείων (“our world consists of the elements in a state of separation”), ὥστ’ ἀναγκαῖον γενέσθαι ἐξ ἑνὸς καὶ συγκεκριμένου.

600. It need not mean that Empedokles said nothing about the world of Love at all; for he obviously says something of both worlds in fr. 17. It is enough to suppose that, having described both in general terms, he went on to treat the world of Strife in detail.

601. Arist. de Gen. Corr. Β, 6. 333 b 21 (R. P. 168 e); Met. Β, 4. 1000 a 29 (R. P. 166 i). Cf. Simpl. Phys. p. 1124, 1 (R. P. 167 b). In other places Aristotle speaks of it as “the One.” Cf. de Gen. Corr. Α, 1. 315 a 7 (R. P. 168 e); Met. Β, 4. 1000 a 29 (R. P. 166 i); Α, 4. 985 a 28 (R. P. ib.). This, however, involves a slight Aristotelian “development.” It is not quite the same thing to say, as Empedokles does, that all things come together “into one,” and to say that they come together “into the One.” The latter expression suggests that they lose their distinct and proper character in the Sphere, and thus become something like Aristotle’s own “matter.” As has been pointed out (p. 265, n. 586), it is hard for Aristotle to grasp the conception of irreducible elements; but there can be no doubt that in the Sphere, as in their separation, the elements remain “what they are” for Empedokles. As Aristotle also knows quite well, the Sphere is a mixture. Compare the difficulties about the “One” of Anaximander discussed in Chap. I. § 15.

602. This accounts for Aristotle’s statement, which he makes once positively (Met. Β, 1. 996 a 7) and once very doubtfully (Met. Γ, 4. 1001 a 12), that Love was the substratum of the One in just the same sense as the Fire of Herakleitos, the Air of Anaximenes, or the Water of Thales. He thinks that all the elements become merged in Love, and so lose their identity. In this case, it is in Love he recognises his own “matter.”

603. For the phrase τοῦ περὶ τὸν ἀέρα πάγου cf. Περὶ διαίτης, i. 10, 1, πρὸς τὸν περιέχοντα πάγον. Et. M. s.v. βηλὸς ... τὸν ἀνωτάτω πάγον καὶ περιέχοντα τὸν πάντα ἀέρα. This probably comes ultimately from Anaximenes. Cf. Chap. I. p. 82, n. 162.

604. Aet. ii. 31, 4 (Dox. p. 363).

605. Aet. ii. 11, 2 (R. P. 170 c).

606. Arist. de Caelo, Β, 13. 295 a 16 (R. P. 170 b). The experiment with τὸ ἐν τοῖς κυάθοις ὕδωρ, which κύκλῳ τοῦ κυάθου φερομένου πολλάκις κάτω τοῦ χαλκοῦ γινόμενον ὅμως οὐ φέρεται κάτω, reminds us of the experiment with the klepsydra in fr. 100.

607. [Plut.] Strom. fr. 10 (Dox. p. 582, 11; R. P. 170 c).

608. Plut. de Pyth. Or. 400 b (R. P. 170 c). We must keep the MS. reading περὶ γῆν with Bernardakis and Diels. The reading περιαυγῆ in R. P. is a conjecture of Wyttenbach’s; but cf. Aet. ii. 20, 13, quoted in the next note.

609. Aet. ii. 20, 13 (Dox. p. 350), Ἐμπεδοκλῆς δύο ἡλίους· τὸν μὲν ἀρχέτυπον, πῦρ ὂν ἐν τῷ ἑτέρῳ ἡμισφαιρίῳ τοῦ κόσμου, πεπληρωκὸς τὸ ἡμισφαίριον, αἰεὶ κατ’ ἀντικρὺ τῇ ἀνταυγείᾳ ἑαυτοῦ τεταγμένον· τὸν δὲ φαινόμενον, ἀνταύγειαν ἐν τῷ ἑτέρῳ ἡμισφαιρίῳ τῷ τοῦ ἀέρος τοῦ θερμομιγοῦς πεπληρωμένῳ, ἀπὸ κυκλοτεροῦς τῆς γῆς κατ’ ἀνάκλασιν γιγνομένην εἰς τὸν ἥλιον τὸν κρυσταλλοειδῆ, συμπεριελκομένην δὲ τῇ κινήσει τοῦ πυρίνου. ὡς δὲ βραχέως εἰρῆσθαι συντεμόντα, ἀνταύγειαν εἶναι τοῦ περὶ τὴν γὴν πυρὸς τὸν ἥλιον.

610. Arist. de Sensu, 6. 446 a 28; de An. Β, 7. 418 b 20.

611. [Plut.] Strom. fr. 10 (Dox. p. 582, 12; R. P. 170 c); Diog. viii. 77; Aet. ii. 31, 1 (cf. Dox. p. 63).

612. Aet. ii. 13, 2 and 11 (Dox. pp. 341 sqq.).

613. Aet. iii. 3, 7; Arist. Meteor. Β, 9. 369 b 12, with Alexander’s commentary.

614. Arist. Meteor. Β, 3. 357 a 24; Aet. iii. 16, 3 (R. P. 170 b). Cf. the clear reference in Arist. Meteor. Β, 1. 353 b 11.

615. Seneca, Q. Nat. iii. 24: “facere solemus dracones et miliaria et complures formas in quibus aere tenui fistulas struimus per declive circumdatas, ut saepe eundem ignem ambiens aqua per tantum fluat spatii quantum efficiendo calori sat est. frigida itaque intrat, effluit calida. idem sub terra Empedocles existimat fieri.”

616. Arist. de An. Β, 4. 415 b 28.

617. Theophr. de causis plantarum, i. 12, 5.

618. [Arist.] de plantis, Α, 1. 815 a 15.

619. Alfred the Englishman translated the Arabic version into Latin in the reign of Henry III. It was retranslated from this version into Greek at the Renaissance by a Greek resident in Italy.

620. Α, 2. 817 b 35, “mundo ... diminuto et non perfecto in complemento suo” (Alfred).

621. Aet. v. 19, 5 (R. P. 173). Plato has made use of the idea of reversed evolution in the Politicus myth.

622. Arist. de Caelo, Γ, 2. 300 b 29 (R. P. 173 a). Cf. de Gen. An. Α, 17. 722 b 17, where fr. 57 is introduced by the words καθάπερ Ἐμπεδοκλῆς γεννᾷ ἐπὶ τῆς Φιλότητος. Simplicius, de Caelo, p. 587, 18, expresses the same thing by saying μουνομελῆ ἔτι τὰ γυῖα ἀπὸ τῆς τοῦ Νείκους διακρίσεως ὄντα ἐπλανᾶτο.

623. Arist. de An. Γ, 6. 430 a 30 (R. P. 173 a).

624. This is well put by Simplicius, de Caelo, p. 587, 20. It is ὅτε τοῦ Νείκους ἐπεκράτει λοιπὸν ἡ Φιλότης ... ἐπὶ τῆς Φιλότητος οὖν ὁ Ἐμπεδοκλῆς ἐκεῖνα εἶπεν, οὐχ ὡς ἐπικρατούσης ἤδη τῆς Φιλότητος, ἀλλ’ ὡς μελλούσης ἐπικρατεῖν. In Phys. p. 371, 33, he says the oxen with human heads were κατὰ τῆν τῆς Φιλίας ἀρχήν.

625. Cf. Plato, Symp. 189 e.

626. Arist. Phys. Β, 8. 198 b 29 (R. P. 173 a).

627. Arist. de Part. An. Α, 1. 640 a 19.

628. Aet. v. 10, 1; 11, 1; 12, 2; 14, 2. Cf. Fredrich, Hippokratische Untersuchungen, pp. 126 sqq.

629. Aet. v. 15, 3; 21, 1 (Dox. p. 190).

630. Aet. v. 25, 4 (Dox. p. 437).

631. Aet. v. 19, 5 (Dox. p. 431). Cf. Eth. Eud. Η, 1. 1235 a 11.

632. Arist. de Respir. 14. 477 a 32; Theophr. de causis plant. i. 21.

633. Nutrition, Aet. v. 27, 1; pleasure and pain, Aet. iv. 9, 15; v. 28, 1; tears and sweat, v. 22, 1.

634. That is, watery vapour, not the elemental air or αἰθήρ (§ 107). It is identical with the “water” mentioned below. It is unnecessary, therefore, to insert καὶ ὕδωρ after πῦρ with Karsten and Diels.

635. Beare, p. 96, n. 1.

636. Ibid. p. 133.

637. Aet. iv. 17, 2 (Dox. p. 407). Beare, p. 133.

638. Beare, pp. 161-3, 180-81.

639. Ibid. pp. 95 sqq.

640. Ibid. pp. 14 sqq.

641. Theophr. de sens. 26.

642. The definition is quoted from Gorgias in Plato, Men. 76 d 4. All our MSS. have ἀπορραοὶ σχημάτων, but Ven. T has in the margin γρ. χρημάτων, which may well be an old tradition. The Ionic for “things” is χρήματα. See Diels, Empedokles und Gorgias, p. 439.

643. See Beare, Elementary Cognition, p. 18.

644. Arist. de An. Γ, 3. 427 a 21.

645. R. P. 178 a. This was the characteristic doctrine of the Sicilian school, from whom it passed to Aristotle and the Stoics. Plato and Hippokrates, on the other hand, adopted the view of Alkmaion (§ 97) that the brain was the seat of consciousness. Kritias (Arist. de An. Α, 2. 405 b 6) probably got the Sicilian doctrine from Gorgias. At a later date, Philistion of Syracuse, Plato’s friend, substituted the ψυχικὸν πνεῦμα (“animal spirits”) which circulated along with the blood.

646. Beare, p. 253.