^{page 25 note 1} The interpretation did not originate with Headlam but with Blomfield, who collected most of the illustrations. The label ‘Treacherous Hound’ comes from Verrall, J. Phil. X. (1882) 299.

^{page 26 note 1} Kock had also anticipated me in the series (wrongly claimed by me, p. 47) λε⋯ξασα, σ⋯νασα δ⋯ξεται. For this knowledge I am indebted to Wecklein's Addenda; I have not yet found K.'s n.

^{page 26 note 2} This use might or might not be low, and might be as old as Archilochus, see Bergk on fr. 185.

^{page 26 note 3} But renders unhappily; ‘loathsome bitch’ —both terms give quite wrong suggestions; ‘laid back a joyful ear’ —English cannot support so literal a translation of the hypallage.

^{page 26 note 4} I withdraw my objection to κλ⋯νασα φαιδρ⋯ν оὗς not so much for T.'s reason, as because in Plato Rep. 613c (cited by T.) the phrase τ⋯ ὦγα ⋯π⋯ τŵν ὣμων ἔχоντες, applied as it there is to human beings, includes the head with the ear, cf. Photius ap. Adam. I take it that during 914–929 Clyt. would have ‘inclined her ear’ in ironical deference.

^{page 27 note 1} I follow those who omit 360, both words being very objectionable here.

^{page 27 note 2} The purpose of 1330 is to emphasise this contrast and to show that the pathos of mortal destiny transcends even this. I follow Lawson for sense, but I read κоὐ for κα⋯ instead of his оὐ; that is better Greek as well as gentler surgery; and cf. Suppl. 296, Eur, . I. T. 901Google Scholar, I. A. 396, all certain. For my part I should add Soph. Tr. 1046.

^{page 28 note 1} I have not the least doubt, and hope soon to show, that Sept. 33 is another instance; and there is a detailed case behind Wecklein, on Sept. 263Google Scholar.

^{page 28 note 2} Pearson on Eur. Hclid. l.c.

^{page 28 note 3} They can translate correctly when they like; see them all emphasise the tense of μоλ⋯ντоς at 34.

^{page 28 note 4} γὐναι, σὺ … μ⋯νων κτλ. But (i) μ⋯νων cannot agree even with the most scornful γ⋯ναι—whose scorn it destroys); (ii) a woman cannot in this sense αἰσχὐνειν εὐν⋯ν ⋯νδρ⋯ς. γ⋯ναι σὺ was the metrical adjustment of σὺ, γ⋯ναι, and what the poet wrote was σὺ γ⋯ννι, … cf. Fr. 61. σὑ is more scornful as first word, and that it had that place is rendered probable by the corresponding taunt. 1617 σ⋯ κτλ.

^{page 29 note 1} φρ⋯νωμα, 830, is my correction of φρ⋯νημα; vid. 831–850, cf. 788–809. You cannot ‘agree with’ a φρ⋯νημα. For the dramatic point see C. Q. XXVI, p. 46 n. 1.

^{page 29 note 2} And here is yet more. At Cho. 446–7 the unexplained ⋯τоιμ⋯τερα γ⋯λωτоς after πоλνσιν⋯ς κνν⋯ς has caused Tucker much conscientious thought; it is simply a passing reference for point's sake to the proverbial ‘grin’ of the dangerous dog. At Cho. 621 κνν⋯φρων refers not to Scylla's shamelessness merely but to her treachery. Eur, . Andr. 630Google Scholar, πρоδ⋯τιν κ⋯να.

^{page 30 note 1} τо⋯ γ⋯ρ τελо⋯ντоς оὐ ξνν⋯κα μηχαν⋯ν 1253. And this is the line which editors emend; τоὺς γ⋯ρ τελо⋯ντας (Heimsoeth) Sidgwick, Wecklein, Wilamowitz, others. (‘The alteration is easy.’) γ⋯ρ means ‘for’; it gives a reason, does not repeat a statement; and where in Greek will yon parallel ξνν. τоὺς τελ. ‘I understand who the perpetrators are’? not from , S.Tr. 868–70Google Scholar.

^{page 31 note 1} The childish extravagance in the monstrosity of the six-headed and twelve-footed Homeric Scylla is exceptional. Regularly in literature as well as art (cf. Waser, O., Skylla u. Charybdis in d. Lit. u. Kunst der Gr. u. Rümer, Zürich, 1894Google Scholar; later reff. in Lübker Reallexikon⁸) she is a fishtailed woman with a girdle of dogs’ (or wolves') heads; and the conception of a cannibal mermaid is, as I shall presently argue, the Aeschylean. Echidna was a ravening snake-tailed woman. Each dwelt in a cave.

^{page 31 note 2} Presumably from caution, but cf. next n.

^{page 31 note 3} It is true that in Anaxil. 22.5 and (of course) in A. Cho. 994 edd. spell ἔχιδνα without a capital; but to suggest that any rigid distinction was thus made in antiquity would be, I think, misleading. The Echidna (half woman, half ⋯φις ὠμηστ⋯ς) was simply an echidna par excellence. Her associates in , E.Ion 1261–5Google Scholar, as in Anaxil. l.c., show that the later ἔχιδνα was conceived as something of a monster; and others besides Herodotus knew that she decapitated her mate.

^{page 31 note 4} = Sidg. 996–1004; normally I refer to S.'s numeration, but here I cannot, since in the location of 997–1004 I follow Weil (so also Blass).

^{page 31 note 5} To whom, however, I owe the ref. to Dryden.

^{page 32 note 1} But for a true estimate (and relevant here) see the second paragraph of Tucker's Introd. to Choeph.

^{page 32 note 2} At the same time, the word implies ‘deal an effective blow’ as at Sept. 399.

^{page 32 note 3} I also note that in relating this very act of the ἔχιδνα Nicander, Ther. 131Google Scholar says κ⋯ρην ⋯π⋯κоψεν ⋯ μ ε ⋯ ν о ν.

^{page 32 note 4} Sidgwick well compares Plat. Phaedr. 234c.

^{page 33 note 1} Order of lines as in Prinz-Wecklein, i.e. 1261–5, 69–74, 79–81, 75–8; but 1266–8 I place after 1319.

^{page 33 note 2} Αιδоν μ⋯τηρ =necis auctor, presented by Lawson as a novelty, was exploded long ago by Karsten; but even if it were itself a possibility it would still be liable to three objections here: (i) you cannot well apply μ⋯γηρ thus tropically to one who is actually a mother: (ii) ‘source death’ would here be intolerably flat: (iii) a mother who rages and blows will belong to Aristophanic parody, not to Aeschylus. Lawson naturally dares not translate his text; he has to throw in a ‘witch’ and a ‘whirlwind’! But apporpriate epithets do not create a substantive.

^{page 33 note 3} I compare πρ⋯σαι γ⋯ρ τὺ φνσ⋯σαι Schol., R, Ar. Wasps 36Google Scholar.

^{page 33 note 4} See L. and S. (8) s.v. πρἱστις; cf. Mair (trans. and n.) on Oppian Hal. I. 369 f. The ‘saw-fish’ is of course no less a reality than the ‘blowfish,’ but the names evidently became confused. pristis = pistrix represents a further confusion.

^{page 34 note 1} This of course is perfectly true. Suidas himself may merely have been muddling his a Aelian (Nat. Anim. ix. 49), but in Aelian l.c. she is at least in such company; and anyway there is my ref. to Oppian.

^{page 34 note 2} More common in Ireland; cf. J. M. Synge's plays passim.

^{Page 35 note 1} Yes I can; because Enger himself completely destroyed the point through punctuating ώσπερε⋯ μ⋯χης τροπᾑ δоκο⋯σα χα⋯ρειν.

^{page 35 note 2} See (not his opinion) Pollard, A. W.Odes from the Greek Dramatists p. viiGoogle Scholar.