^{page 53 note 1} Cf. e.g. Paley's edition, p. 8. Rolfe's paper will be discussed presently.

^{page 53 note 2} Paley points out that 980 ff. are an exception : . It is instructive to compare the dialectical development of the same thesis in Med. 1081 ff. He might have added 758-760, where the speaker makes the extraordinary admission that a glorious death may be painful to the dead man, though a source of renown to the survivors! This seems to be intended as a counterblast to Eur. Tro. 633, Soph. El. 1170, in which the freedom from pain enjoyed by the dead is advanced as a reason for preferring death to a wretched life.

^{page 54 note 1} p. viii.

^{page 54 note 2} The Cyclops must be left out of account for various reasons: the peculiarities of metre and diction which distinguish it have no parallel in the Rhesus.

^{page 54 note 3} Crates of Mallus (schol. Rhes. 524) excused an astronomical error on the ground that Euripides was a young man when he composed the play. The same assumption is the last stronghold of modern critics who defend its authenticity.

^{page 54 note 4} The same point is made bv F. Hagenbach (de Rheso tragedia, Basiliae, 1863, p. 30).

^{page 54 note 5} Dindorf made the same suggestion (ed. Oxon. p. 560 f.), but seems subsequently to have abandoned it (Poet. Seen. ⁵, p. 21).

^{page 54 note 6} Using Beck's index, which is notoriously untrustworthy, I counted 165 instances (as against Murray's 177). By a rough comparison with the Bacchae and the Hippolytus for the letter a alone, I found the proportion for the Bacchae to be much larger, and for the Hippolytus slightly smaller.

^{page 54 note 7} G. Hermann, Opuscula, III. 262 ff. ; L. Eysert, progr. Lips. 1891 ; J. C. Rolfe, Harvard Studies, IV. (1893), 61 ff.

^{page 55 note 1} I found 18 instances belonging to this category under the letter a alone. Professor Rolfe has most courteously informed me that he is now aware of this omission.

^{page 55 note 2} After the elimination of a fair proportion of the doubtful cases, my figures showed 90 approximations to Euripides, as compared with 30 to Sophocles and 25 to Aeschylus. I have incorporated a few additional examples from Rolfe.

^{page 55 note 3} , though Euripidean by preponderance, just occurs in Aeschylus and Sophocles. It may here be observed that the non-Euripidean quality of the style is largely due to the absence from the Rhesus of much of the characteristic vocabulary. This would, I believe, be a fruitful field for investigation. From a casual inspection of the letter a I note the following, taken almost at random : , .

^{page 56 note 1} The remark is applicable to many of the , which do not directly suggest Aeschylus: , , etc.

^{page 58 note 1} See Decharme, Euripides, tr. Loeb, pp. 270, 273. I do not mention the employment of four actors or the anapaestic opening, since the fact in the first case and the significance of the second are disputed.

^{page 58 note 2} This tendency was illustrated by Jebb on O.C. 554. See also my n. on Eur. Hel. 674

^{page 58 note 3} Of course some allowance must be made for the fact that words appropriate to the subject-matter are apt to be repeated (cf. the recurrence in the Ion of the rare word ); but that principle is insufficient to account for the repetitions of the Rhesus. A comparison of the occurrences of certain common military words in the Rhesus and in Aesch. Theb., , as Aristophanes calls it, yields the following results : and its derivatives, 65 Rhes., 14 Theb.; , 22 Rhes., 13 Theb.; , 24 Rhes., 8 Theb.

^{page 59 note 1} E.g. the opening scene and that in which Odysseus and Diomedes escape from the guards are well-contrived and impressive.

^{page 59 note 2} is Nauck's certain emendation of , being clearly indicated by the words . It should be remembered that the aim of Dicaearchus was not critical.

^{page 60 note 1} Some such conclusion is adopted by Dieterich in Pauly-Wissowa, VI. 1265.

^{page 60 note 2} Einleitung in die gr. Tragödie, p. 41. Dieterich also says that' in style and metre the imitation of Sophocles is self-evident. It is a pity that these critics were not more explicit.

^{page 60 note 3} Wilamowitz had forestalled Murray's criticism that there is no evidence of such a tendency in the scanty fragments which survive from the fourth-century drama. But it is curious that he should have selected as its representative the very man—Theodectes—whom Murray rejects as unsuitable. I note in passing that two of the words common to Sophocles and the Rhesus occur together in Theodect. fr. 17.

^{page 60 note 4} 277e.

^{page 60 note 5} Vit. Soph. 12.

^{page 60 note 6} Ib. 13.

^{page 60 note 7} Arist. poet. 3. 1448a 26; Diog. L. 4. 20; Suid. s.v. .

^{page 60 note 8} Greek Tragedy, pp. 89 f.

^{page 61 note 1} On the critical history of the Rhesus, see Wilamowitz, de Rhesi scholiis, Greifswald, 1877.

^{page 61 note 2} Aelian var. hist. 2. 8.