^{page 53 note 1} Recent detailed analyses by Dorrie, H. (Hermes lxxxiv [1956], 296319)Google Scholar and MacDowell, O. M. (CQ. N.S. ix [1959], 261–8).Google Scholar

But I feel that the solution to the problem of 1437–53 is to be sought in the repeat performance of the play, rather than in the transposition of lines and invention of lacunae in the present text.

^{page 53 note 2} Stanford, W. B., Aristophanes, The Frogs (1963), 192.Google Scholar

^{page 53 note 3} The battle of Notium, as a result of which public opinion in Athens once again swung against Alcibiades, took place sometime during the winter months of 407/6 B.C. It was certainly no later than March 406 and, on balance, the somewhat confusing evidence of Xenophon (Hellenica 1. 5. 16–1. 6. 1) points, I think, towards autumn 407.Google Scholar

The Parian marble puts Euripides' death in 407/6, Apollodorus (cited by Diodorus Siculus, , 13. 103) in 406/5. The latter date is less likely in view of the fact that the Frogs had been conceived, written, and performed by January 405. There is a statement in the Life of Euripides that on the news of his death Sophocles, who himself died late in 406, dressed himself in black and introduced his actors and chorus as a token of mourning. If there is any truth in this story it must refer to the City Dionysia in March 406, and the implication is that Euripides' death occurred early in that year. He died in Macedon, having left Athens a year or so previously. But he cannot be supposed to have forgotten in such a short time the vital preoccupations of his city.Google Scholar

^{page 54 note 1} Either 1431a or 1431b1 must go, and the choice is not a difficult one. It is essential to retain thewithout which the line is neither accurate nor apposite cf. Agamemnon 717–18 Google Scholar

The source of the variant which appears in most manuscripts (though not in V or A) is no doubt Plutarch's faulty quotation of the line (Alcibides 16. 2). It is hardly necessary to point out that such quotation was frequently done from memory and, therefore, often inaccurate. As for Valerius Maximus (7. 2. 7), he hardly inspires confidence in his accuracy when he attributes the remark to remissum ab inferis … Periclem! Further, whereas in his citation (non oportere in urbe nutriri leonem) the solitary laonem might suggest 1431b, the opening words non oportere equally strongly suggest 1431a.Google Scholar

^{page 54 note 2} It was probably the realization of this basic difficulty which caused the Scholiast to comment

^{page 55 note 1} The view of Coulon, Aristophane, tome iv (Budé, series, 1962), 152, that both poets express themselves against a return of Alcibiades, and that this represents Aristophanes' own opinion, is quite untenable. The two are contrary, and no doubt reflect the broad division of public opinion on the issue.Google Scholar

^{page 55 note 2} Cf. Thuc. 1. 68. 4 and see Denniston, J. D., The Greek Particles, 370–1Google Scholar. In any case one must not be over-literary on this point. As Dover, K. J. has pointed out (article on Greek comedy in Fifty Tears (and Twelve) of Classical Scholarship, 125), the gestures of comic actors were as important as their words.Google Scholar

^{page 55 note 3} See, e.g., Lowe, J. C. B., ‘The Manuscript Evidence for Changes of Speaker in Aristophanes’, BICS ix (1962), 27–42.Google Scholar