^{page 142 note 1} I can here deal only with the Works and Days: as for the other poems, ‘spatiis exclusus niquis, praetereo.’

^{page 142 note 2} For the evidence adduced in the notes I depend almost entirely on Rzach's magnificent edition (Teubner, 1902).

^{page 142 note 3} Crates rejected the Prelude on. the ground that it might be prefixed to any poem: the objections of Aristarchus are unknown.

^{page 142 note 4} See Proclus, , Praef. in Opp. Dimitrijević (Stud. Hes., p. 8)Google Scholar has pointed out that the copy seen by Praxiphanes may well be the same leaden copy which was shown to Pausanias on Mt. Helicon. The latter describes the document as ‘defaced by time,’ but does not say that the script was archaic. This copy need not, therefore, have been primitive or earlier than the fifth or sixth century.

^{page 142 note 5} We must assume that the –∪∪–– needed to complete this version was cut out by the conflator.

^{page 143 note 1} Both lines are quoted by Origen: the first alone is cited by the scholiast on Aratus 103.

^{page 143 note 2} I number the new lines of the Geneva Papyrus 169 αd, not (as Rzach) 169b–e.

^{page 143 note 3} The actual reading of the scholium is τοτον καί τòν ξς ϕληναϕώδεις ξοικἰζουσιτν 'Hσιóôου. Schoemann changed τóὐ into τοὐς; but Proclus may, in fact, have had in his text only 169, 169a, the additional lines 169b–d having already disappeared. For other testimonia relating to 169 see Rzach's larger edition.

^{page 143 note 4} Very possibly 174 sq., 504 sqq., are by the same hand.

^{page 144 note 1} The remainder of 1. 189 may well have been supplied by the conflator, but it may be retained as balancing κακν ῥεκτρα, etc. οὐκ σται (1. 193) is a similar addition: if this verb be removed αἰδώς will naturally be coupled with δκη, σοντγαι being understood.

^{page 144 note 2} There is surely a lacuna after 1. 239: very possibly 11. 244 sq. originally stood here. The lacuna would be filled, and the omission of these two lines by Aeschines when quoting 11. 240–247 would be explained. Their transference may be due to Alexandrine meddling.

^{page 145 note 1} Perhaps 11. 267–269 constitute a third. Otherwise we have (1) the ‘immortals,’ (2) the ‘immortal guardians of Zeus,’ and (3) Zeus himself, all engaged on the same task. If a third version be admitted, note that it is probably imitated by Sophocles, El. 174 sq., Ζες, δςϕορᾀ πντα.

^{page 145 note 2} Lehrs, (Quaest. Ep., p. 90)Google Scholar regards 393–404 as a ‘double recension.’ Steitz, (Werke u. Tage d. Hes., p. 121)Google Scholar would eject 396–404.

^{page 145 note 3} I regard οδ' πιμετρσω as a stop-gap, added by the conflator in fitting the two versions together. But the preceding line, though earlier than the date of conflation, is probably not original: like all the personal allusions Hesiod, it is dragged in.

^{page 146 note 1} LI. 502 sq. are both awkward and in a wrong context. Schoemann proposes to place them after 1. 497, but the neighbourhood of 1. 596 would provide them with a more congenial home.

^{page 146 note 2} I take it that the author of C used this word as equivalent to κυρτóν (cp. Shakespeare, , Tempest II. ii. 259Google Scholar), and by substituting τρἰποδι added a further detail— ‘bent over a stick.’

^{page 146 note 3} L. 648 may have been common to both A and C, and have been read after 1. 618, where an apodosis is needed. LI. 633–640, however early, are perhaps not original. Whether by the original poet or a later this personal reminiscence is dragged in.

^{page 146 note 4} LI. 649 sqq. or 654 sqq. may be more or less accretion.

^{page 147 note 1} I.e. Aristarchus (see Dimitrijević, , Stud. Hes., p. 139)Google Scholar.

^{page 147 note 2} So Hermann, who regards 682 sq. as alternative with 1. 678.

^{page 147 note 3} In 1. 68 translate μ δ', ‘But do not …’

^{page 147 note 4} Assuming that an original ‘cue’ ειαρινòςδ' ἂλλος (or . δ εἰ.) has been altered by the conflator. As the text stands εἰαρινòς δ' οδτος would suggest that a long passage had intervened, making resumption necessary.

^{page 147 note 5} So Hermann.

^{page 147 note 6} Hermann proposed to place 1. 687 after 1. 677.

^{page 147 note 7} The same order is followed in the fourth century Rainer papyrus (Rzach's A) as in the mediaeval MSS.

^{page 148 note 1} Poor or unaccommodating variants may have been omitted, but the nature of the variants preserved indicates that practically everything was retained.

^{page 148 note 2} If the editor meant only to produce a standard containing all permissible variants (between which rhapsodes might choose), his purpose must have almost from the first have been ignored. But the presence of stop-gaps is against such a supposition.

^{page 148 note 3} Cp. the scholium on Θ. 535–541 (ap. Sandys, , Hist, of Class. Schol. I. 142)Google Scholar, whence it appears that Aristarchus, Zenodotus, and others, recognized 535–537 and 538 sq., 541, as variants (εἰς γρ τν αὐτν γεγραμμνοι εισἰ διίνοιαν), and were at a stand to choose between them.

^{page 148 note 4} Plato's citations of 11. 122 sq. might be thought to indicate that the text was still unstereotyped; but since Plato's two quotations are not consistent, it is natural to conclude that he was quoting from memory.

^{page 148 note 5} stud. Hes., p. 10.

^{page 148 note 6} Ap. Plut. Thes. XX. ( = Rzach, Fr. 105). It may be noteworthy in this connexion that Theseus becomes more popular in Attica towards the close of the Peisistratean period (cp. Bury, , Hist. of Greece I. 213Google Scholar).

^{page 149 note 1} As though a foreigner should charge Lord Palmerston with having tampered with the text of Shakespeare.

^{page 149 note 2} It is useless to speculate as to the nature of the text on which Hereas relied.

^{page 149 note 3} I would not, if I could, try to reargue the case for the Homeric recension. So far as Homer is concerned I creep beneath the shield of Leaf (Iliad I., p. xix.): many of Dr. Leaf's arguments in support of the Homeric recension are equally in favour of a Hesiodic recension.

^{page 149 note 4} Does this attach new significance to the title ῥαΨΨδοί the ῥαπτν ων οιδοἰ?

^{page 149 note 5} Stud. Hes., p. II.

^{page 149 note 6} Hdt. I. 59; Ath. Pol. 15.

^{page 149 note 7} Ath. Pol. 16. Perhaps there is a special appropriateness, from our point of view, in the saying (ib. and cp. [Plato] Hipparchus, p. 229B) that the reign of Peisistratus was ⊙π κρóνου βἱος: cp. Works and Days, 11. III, 169.

^{page 150 note 1} I speak always of ‘Peisistratus,’ but should be well content to substitute Hippias or Hipparchus. In [Plato] Hipparchus, pp. 228c sqq., the latter is actually credited with attempts to educate country-folk as well as citizens by means of precepts which, though not Hesiod's, are Hesiodic. That Solon is also credited with the ordinance for Homeric recitation is natural enough: democracy does not stop at such trifles as the attribution to its hero of an enemy's achievement.

^{page 150 note 2} Stud. Hes., pp. 231 sqq.

^{page 150 note 3} For those who edited or in some measure studied the Hesiodic poems, see Rzach in Pauly-Wissowa, art. Hesiodos, cols. 1225 sq.

^{page 150 note 4} Our scholia relate only to the vulgate text, and naturally do not deal with verses which do not appear in that version: the note on 1. 169 is exceptional, being retained because its application was misunderstood.