^{page 45 note 2} While κκτείνασα comes from Canter and ϕαιδρòν ούς from H. L. Ahrens, it was Madvig (Adv. Crit. I. 200–1) who first established this blend, which, grotesque as it is, has been very generally adopted, e.g. by Wecklein, Sidgwick, Wilamowitz, and the latest English (American) editor Weir Smyth; though by none of these with Madvig's own δξεται. Ahrens himself (Philologus, suppl. 1860, pp. 630 sqq.) proposed κλίνασα ϕαιôρòν οȗς comparing Xen. Cyneg. VI. 15 ùπò χαρς…έπικλίνασαι τά ὦтα. My main reason for rejecting this emendation will be evident in due course; but there is also another. Mr. J. U. Powell drew my attention to a remark of Ahrens which he justly describes as naive: ‘Das Hauptmanöver der schmeichelnden Hunde, namlich ούρῇ σαίνειί konnte auf die Clytaemblend, nestra natürlich nicht übertragen werden.’ Quite so; and the same applies to κλἱνασα… οὖς. I understand, indeed, that some persons possess that accomplishment, but they must be few, and it is not queenly. λείξασα is a very different matter; Clytemnestra's flatteries had been applied with her tongue.

^{page 46 note 1} See his note on 829 sqq., and cf. on 800. Nothing could be further from Agamemnon's thoughts at that moment than any suspicion of Clytemnestra; nor does he ever have any; she sees to that. His complete satisfaction and self-confidence are the point throughout this speech, the tone of which is struck in the excellent θεοùς εμοì μεταιτίοʋs of the opening sentence. 838–840 is unmistakable dramatic irony, characteristically Greek and very good. ‘I can speak from the experience of disloyalty’ (nothing to the experience he will yet have this day!), ‘for indeed I know it thoroughly’ (he does not know it when he sees it now!). The following remarks, scribbled by the previous owner (a lady) of my Headlam against the first two lines of his note on 829 sqq., appear to me both apt and acute: ‘I wonder if he is really thinking of Clyt. at all. It seems more poignant and also more natural for him to have forgotten after ten years that she could have any grudge against him.’

^{page 46 note 2} Cf. 828, Eum. 106.

^{page 46 note 3} Sylva Critica, § 26.

^{page 46 note 4} My own main objection to it is that, while the tails of compounds are often purely ornamental, this one is like a scorpion's, suicidal ϕαιδρός in this kind of connexion refers invariably and inevitably to the outward appearance; the very last thing it could be referred to is the νοs.

^{page 47 note 1} Hom. Hymn. III. 76=221.

^{page 47 note 2} Compare, in Shakespeare's terrible presentation of impotent senile fury, ‘I will do such things!’ (Lear II. iv. 280).

^{page 47 note 3} For evidence of interpolation in this play see Headlam's note on 1277, last par., and to that list add 1626 αІσϰύνουσ' for -ѡν (not -ας). And there is a good deal more, as I shall yet show in some sequel article; e.g. 7,301. 871, 1226, are demonstrable concoctions.

^{page 47 note 4} E.g. our play. 735–6 (in light of preceding context from 681); Soph, . Trach. 893–5Google Scholar; and still more ibid. 1050–2, where Έρινύων άμϕίβληστρον is analogous to our Άτης δάκος-and δολπις is just before it.

^{page 48 note 1} Verrall, has some excellent remarks on this point (leading to a brilliant correction of Eur. Hec. 1162) in J. Phil. X. 302–5Google Scholar.

^{page 48 note 2} Headlam's note has this astonishing remark: ‘οἶα is understood with λέξασα as well as with ϒεύξεται as though it were οἷα λέξασα οἶα τεύξεται’ That is exactly as if one should say ‘this board is white, as though it were black and white.’ Those two οἶα's refer to things diametrically opposite; how can they then be telescoped into one?

^{page 48 note 3} Sidgwick, Headlam, Weir Smyth, all place it in their texts.

^{page 48 note 4} Wecklein's Appendix attributes it to Burges, whose punctuation, however, stultifies it. Blaydes, in recommending it in his Adv. in Aesch. (1895) attributes it to NaberGoogle Scholar. Paley approves it.

^{page 50 note 1} See note 3 below, and cf. line 1291 below, Αιδον π ύ λ α ς δ τάσδ'.

^{page 50 note 2} Cf. 1433 Άτην Έρινύν θ', αἶσι τόνδ' ἔσϕαξ'έϒώ.

^{page 50 note 3} The dog of the Hesiodic passage is Cerberus himself; and I may now point out that it is evidently Cerberus who, owing to the peculiar character of his infernal post and responsibilities (for which see Hesiod here and Lucian, , De Luctu 4Google Scholar, and note the paradox of Statius, , Theb. ii. 28 ‘saeuus et intranti populo’)Google Scholar, is the proto-type of this not wholly natural, proverbially insidious and treacherous dog. That is shown not merely by the general resemblance, but by the point of contact in Ar. Knights 1030 κύναΚέρβЄρоν; compare my above remarks on that passage. (It is for such reasons that I have called Clytemnestra'hell'-hound. To appreciate this kind of poetry fully one must be alive to the implications or associations; now what is given in Lucian loc. cit. is the popular form of this immemorial piece of Greek folk-lore; observethen how perfectly it fits the part of Clytem-nestra—τοùς μν έσαϕικνονμένους ϕІλιόντι καì εἱρηνικòν προσβλέπων.) I should add that to understand Horace, , Odes II. xix. 31Google Scholar ‘et rece-dentis trilingui ore pedes tetigitque crura’ one must see that the point there is exactly the reverse paradox to that of Statins as quoted above; but Kiessling explains wrongly by ‘noch vertraulicher,’ commentators generally ignore, Mitscherlich cites only the Hesiodic passage, and Wickham is even at pains to justify the ‘intentional dullness’ (!) of this stanza—which is in fact more like a climax. But nowhere at all have I seen it noted that this conception of the house-dog of the House of Hades is itself a typically witty Greek paradox; his functions being, in the nature of things, the exact opposite of those of the ordinary house-dog. Cerberus, by the way, will bite even an incomer if he walks in too slowly, Lucian, , Dial. Mort. 21, 1Google Scholar κуὡ ἔτι διαμέλλοντα αύτòν δακών.