^{page 1 note 1} In chronological order: Paratore, E., Dioniso N.S. xv (1952), 19 f.Google Scholar, and Studi Funaioli (Rome, 1955), 339f.Google Scholar; Friedrich, W. H., Euripides und Diphilos, Zetemata v (1953), 1104Google Scholar; Giomini, R., Saggio sulla ‘Phaedra’ di Seneca (Rome, 1955)Google Scholar, and his edition of the play (Rome, 1955); Zintzen, C., Analytisches Hypomnema zu Senecas Phaedra (Meisenheim, 1960)Google Scholar; Grimal, P., ‘L'Originalité de Sénèque, dans la tragédie de Phèdre’, REL xli (1963), 297–314Google Scholar, and his edition, Sénèque, Phaedra (Paris, 1963)Google Scholar; cf., also, Snell, B., Scenes from Greek Tragedy (Berkeley, 1964)Google Scholar, chs. 2 and 3 (by far the least generous to Senecan originality); Barrett, W. S., Euripides: Hippolytus (Oxford, 1964)Google Scholar, Introduction, 16–45; Heldmann, K., ‘Seneca's Phaedra und ihre griechische Vorbilder’, Hermes xcvi (1968), 88–117Google Scholar; and Herter, H., ‘Phaidra in griechischer und römischer Gestalt’, RhM cxiv (1971), 44–77.Google Scholar

^{page 1 note 2} Austin, R. G., Virgil: Aeneid IV (Oxford, 1955), Introduction, pp. ix and x.Google Scholar

^{page 1 note 3} See Maguinness, W. S., ‘Seneca and the Poets’, Hermathena lxxxviii (1956), 81–98Google Scholar, esp. 93. Seneca quotes Aen. iv less frequently than other books, but his interests in prose are essentially in the masculine virtues, which are better illustrated by other books of the Aeneid. In Ep. Mor. I have found three quotations from Book iv: 3–4 in Ep. cii; 158–9Google Scholar in Ep. lxiv; 653Google Scholar in Ep. xii.Google Scholar

^{page 2 note 1} 437–49.

^{page 2 note 2} Aen. iv. 441–6Google Scholar; Phaedr. 580–2.Google Scholar

^{page 2 note 3} This is perhaps best shown by reference to Euripides' tragedies, the main dramatic influence on Seneca's writing which is known to us. Barlow, Shirley A., The Imagery of Euripides (London, 1971)Google Scholar, writes of his economy in the use of ‘epic similes’ to create ‘emphasis by artificial spacing’ (p. 102) and draws her examples from messenger speeches, choral lyric, and monody, where the function is pathetic and descriptive. On the messenger speeches she writes ‘even the similes … while in many ways resembling those of Homer, are (partly because they are shorter) more disciplined to the main action they describe’ (p. 73). The parenthesis is crucial. Turning to Hippolytus and Medea we find many vivid metaphors, but only six and four similes respectively, of which only one exceeds a line of verse: Hipp. 1201 and 1221Google Scholar, Med. 1213Google Scholar mark the moments of crisis in the messenger speeches; Hipp. 564 and 828Google Scholar, Med. 1279Google Scholar occur in lyric and monody (both Hipp. 828Google Scholar ὄρνις γὰρ ὤς τις ἔκ χερῶν ἄφαντος εί and Med. 1279 ὡς ἄρ' ἤσθα πἐτρος ἤ σίδα/ρσς mark the pathos of the heroine's downfall). In dialogue Med. 523–4 and Hipp. 429 illustrate psychological hypotheses, not the dramatic action; Hipp. 872–3 πρὸς γάρ τινος / οὶωνὸν ὥστε μάντις εἰσορῶ κακόν, excised by Barrett as an interpolation, is in any case trivial. Nearest to the Senecan descriptive simile are the words of the Nurse at Med. 28 ὡς δὲ πἐτρος ἤ θαλἀσσιος / κλύδων ἀκούει… φἰλων, where the image is essential to convey to the audience the stony dehumanization of the as yet unseen Medea. By analogy with this passage, the full descriptive content of the Senecan simile argues strongly for the belief that his dramas were written not for stage-presentation but for recitation, in which similes would be needed for the same function as in epic.

^{page 3 note 1} Hipp. 304–5.Google Scholar

^{page 3 note 2} Med. 28Google Scholar; Andr. 532.Google Scholar

^{page 3 note 3} Dial. v. 25.Google Scholar

^{page 3 note 4} Cf. Const. Sap. iii. 5Google Scholar; Vit. Beat. xxvii. 3.Google Scholar

^{page 4 note 1} Phaedr. 580.Google Scholar

^{page 4 note 2} Compare in particular de Ira (Dial. iv. 15)Google Scholar, ‘omnes istae feritate liberae gentes leonum luporumque ritu, ut servire non possunt, ita nec imperare; non enim humani vim ingenii sed feri et intractabilis habent.’ For this sense of tractare cf. de Clem. iii. 15Google Scholar, ‘nullum animal morosius est, nullum maiore arte tractandum quam homo’, and the whole preceding analogy between the control of men and animals.

^{page 4 note 3} On the significance of the wound image for Virgil's presentation of Dido, especially in its fullest form at Aen. iv. 66–73Google Scholar, see Pöschl, V., The Art of Vergil (Michigan, 1962), 79–81Google Scholar, and Quinn, K., Virgil's Aeneid, A Critical Description (London, 1968), 313.Google Scholar

^{page 5 note 1} Virgil's language here is a conscious echo of Cat. lxvi. 23Google Scholar, ‘quam penitus maestas exedit cura medullas’ (so also Cat. c. 7).

^{page 5 note 2} lxviii. 53.

^{page 5 note 3} Her. xv. 12Google Scholar: me calor Aetnaeo non minor igne tenet.

^{page 5 note 4} Met. xiii. 867 ff.Google Scholar: uror enim, laesusque exaestuat acrius ignis / cumque suis videor translatam viribus Aetnam / pectore ferre meo.

^{page 5 note 5} Aen. xii. 526Google Scholar: fluctuat ira intus.

^{page 5 note 6} Met. ix. 465Google Scholar: verum tamen aestuat intus.

^{page 5 note 7} A peculiarly perverse example is the description of Tereus' falling in love, Met. vi. 490 f.Google Scholar, ‘at rex Odrysius quamvis secessit in illa / aestuat et repetens faciem motusque manusque / qualia vult fingit quae nondum vidit, et ignes / ipse suos nutrit, cura removente soporem.’ The introduction to this portrait (from the feast and general sleep to the contrasting at rex), the adaptation of vultus verbaque, of alit, the destruction of sleep by cura, are all modelled on the transition from Aen. i to Aen. iv. Other examples are Met. iv. 64Google Scholar, ‘quoque magis tegitur tectus magis aestuat ignis’ (the secret passion of Pyramus and Thisbe, and a possible influence on Sen. Phaedr. 363–5)Google Scholar, Met. ix. 465Google Scholar above and 765, and xiv. 351–2, ‘flammaque per totas visa est errare medullas. / ut primum valido mentem collegit ab aestu…’

^{page 6 note 1} Cf. Phaedr. 134Google Scholar: qui blandiendo dulce nutrivit malum / sero recusat.

^{page 6 note 2} Phaedr. 282Google Scholar: vorat tectas penitus medullas.

^{page 6 note 3} Aen. iv. 23.Google Scholar

^{page 6 note 4} Cf. Zintzen, Snell, Barrett (above, p. 1 n. 1). It is argued that the second scene based on the surviving Hippolytus was grafted by Seneca on to his main source, alleged to be the lost Hippolytus Kalyptomenos, and the conflict between Phaedra's unresisting surrender in the first scene and the torment of her second appearance is due to this grafting.

^{page 6 note 5} Hipp. 131–40.Google Scholar

^{page 6 note 6} Ibid. 181–207.

^{page 6 note 7} For this transition see Giomini, , Saggio (above, p. 1 n. 1) 51Google Scholar, and Heldmann, (above, p. 1 n. 1), 103 n. 1.Google Scholar

^{page 7 note 1} See Merkelbach, , RhM c (1957), 99–100Google Scholar (and Snell, , p. 43)Google Scholar; he argues that the idea of Hippolytus as a second Theseus—focus of what I have called the vultus-motif—derives from Euripides' lost play, since it is also found in Hel. Aeth. i. 10. 2Google Scholar, which cannot derive from Seneca. His attitude is that of the Hellenist (‘Alles Wesentliche dieser grossartigen Szene ist euripideisch’), yet even with his emendation of Heliodorus the resemblance to Seneca is too slight to cast doubt on his independence in developing this passage; there is no evidence for Seneca's intense emphasis on physical beauty in Euripides' lost play.

^{page 7 note 2} See above, p. 2 n. 3.

^{page 8 note 1} Aen. iv. 16Google Scholar, vinclo sociare iugali; 59, vincla iugalia; 496, lectumque iugalem.

^{page 8 note 2} The central verb of the Virgilian image, fluctuare, is absent from the concordance of Barry, Deferrari, and Maguire; they quote no comparable imagery based on the noun fluctus. Instead Ovid develops the fire image of torment, as in the passages quoted above, p. 5 nn. 3–7.

^{page 8 note 3} Virgil is himself echoing the rhythm and vocabulary of Cat. lxiv. 62, ‘magnis curarum fluctuat undis’. The verb is active in form, with passive or reflexive meaning in Catullus, Lucretius, Virgil, and at times in Senecan prose; there is one definite instance in tragedy, HF 699.Google Scholar The deponent form introduced by Livy is common in Senecan prose, and in tragedy occurs at Agam. 109Google Scholar, Med. 943Google Scholar, below, and Tro. 657—Google Scholar all images of emotional turmoil. On Aen. iv. 532Google Scholar, Quinn, , op. cit. 146Google Scholar, argues convincingly for Dido, not amor as subject of the verb; the parallels cited below, p. 8 n. 4, mostly favour the person as subject, expressing the passion in the instrumental ablative (but ef. Aen. xii. 526Google Scholar, in which ira must be subject, representing the anger of both warriors; and the nominative pudor in xii. 666Google Scholar, with the parallel verb aestuat).

^{page 8 note 4} We can see how Virgil varied his expression of the basic theme when these metaphors recur in later passages of the Aeneid, applied to the anxious responsibility of Aeneas, or the turbulent anger of Turnus. Aen. viii. 19–21Google Scholar, for example, combines with virtually no change Aen. iv. 532 and 285–6Google Scholar: ‘magno curarum fluctuat aestu / atque animum nunc huc celerem nunc dividit illuc / in partisque rapit varias.’ Books ix–xii illustrate further variations in successive allusions to Turnus; cf. ix. 798, below; x. 680, ‘haec memorans animo nunc huc nunc fluctuat illuc’; xii. 486, ‘vario nequiquam fluctuat aestu’; 525 f., ‘nunc, nunc / fluctuat ira intus’; 666–8 below; 831, ‘irarum tantos volvis sub pectore fluctus’. xii. 666–8 is interesting as an embryonic psycho-machia, containing in its array of conflicting emotions the nucleus of an elaborate form like Agam. 133–8.Google Scholar

^{page 9 note 1} mens aestuat ira.

^{page 9 note 2} aestuat ingens / una in corde pudor, mixtoque insania luctu / et furiis agitatus amor et conscia virtus.

^{page 9 note 3} Perhaps an echo of Aen. iv. 531 f.Google Scholar, ‘rursusque resurgens / saevit amor’.