¹ See Duckworth, G.E., The Nature of Roman Comedy, Princeton, 1952, p. 121Google Scholar and n. Don. on Eun. 197, implying that the two men do not leave the stage, says:‘recte Thais nunc partem argumenti exsequitur tacitam apud Phaedriam propter praesen-tiam servi, quern poeta vult ita nescire, ut audeat ad vitiandam virginem subornare Chaeream.’ Drexler, , Hermes lxxiii (1938), 82Google Scholar, takes the opposite view and says of Phaedria and Parmeno: ‘196 gehen beide ins Haus.’

² Studies in Menander, 2nd edition, Manchester, 1960, p. 70.Google Scholar

³ At this juncture Thais admits no more than that it is a possibility: ‘arbitror; certum non scimu'’ (110-ii); cf. Don. on Eun. no. 2: ‘bene ‘arbitror’ et nihil certi: quomodo enim ausurus esset Parmeno ador-nare Chaeream ad vitiandam virginem, si praescisset civem esse ?’

¹ See, for example, the passage of Webster quoted, Duckworth, , op. cit., p. 121 n.Google Scholar, and Drexler, , Hermes lxxiii (1938), 82.Google Scholar

² I am assuming throughout that such a prologue did exist in Menander's Eunuchus. This cannot, of course, be proved, but such evidence as we have from other plays by the same author makes the assumption of its existence reasonable.

³ See Oppermann, , Hermes lxix (1934), 262–9.Google Scholar

⁴ Ad. 432 and Don.

¹ Eun. 341, 363; Phorm. 151, 458; Hec. 272; Ad. 247, 432.

² On Eun. 341, 363 (‘hoc dicere abeuntes solent’); on Hec. 272 (‘discedentis verbum’); on Ad. 247.

³ I am indebted to Mr. I. M. Barton for drawing my attention to this point.

⁴ On Eun. 198: ‘hic Terentius ostendit virtutis suae hoc esse, ut pervulgatas personas nove inducat et tamen a consuetudine non recedat, ut puta meretricem bonam cum facit, capiat tamen et delectet animum spectatoris.' Webster, by suggesting in the passage quoted that the innovation is a Menandrian one to be explained in a monologue, seems either to overlook Donatus' comment, or to regard his testimony as worthless; his view on this point appears even more clearly in his Studies in Later Greek Comedy, Manchester, 1953, pp. 117 and 124.Google Scholar

⁵ 79 may only be a slave's typical and prejudiced reaction to his master's lover, particularly if she is a meretrix, and need not be truly at variance with the portrait at 197. But there is a further indication of more ‘normal‘ characterization of Thais at 739 ff. On the other hand, Thais'; actions in the latter half of the play do seem to bear out the good intentions she stated at 197 ff.; therefore if we are to understand from Don. on Eun. 198 that not only the passage of kindly sentiment on which he is commenting but also all references to a Thais of good character are an innovation by Terence, it might mean that the Roman poet has considerably altered the last part of the play.

¹ See Don. on Eun. Praef.1.5*.

¹ See, for example, And. 236–66.

² Hermes lxxiii (1938), 75.Google Scholar

³ Ibid, lxxvi (1941), 75 ff.

⁴ Ibid. 252 ff. Knoche's views are accepted by Webster, , Studies in Menander, p. 72.Google Scholar

⁵ Térence, ed.Marouzeau, J. (Budé edn.), i, Paris, 1947, p. 259 n. 1Google Scholar: ‘depuis le moment où Thaīs a indiqué qu';elle attendait Chrémès chez elle (v. 205–206), il s'est écoulé assez de temps pour que, suivant les conventions scéniques, l'entrevue ait pu trouver place.’

¹ Eum. 500–3.