Two new manuscripts have come to my attention since I made a study of the MS. tradition of Seneca's Natural Questions. The first is Munich, Bayerische Staats-bibliothek, Clm 18961, part ii (foll. 25–46v), which I shall call Y, a late-ninth-century manuscript from Brittany or the lower Loire. It contains a short collection of theological and philosophical excerpts from a variety of authors, a collection emanating from the circle of Alcuin, probably from the generation after Alcuin himself. Included in the collection are three extracts from the preface of Book 1 of Seneca's Natural Questions. Since the earliest surviving manuscripts of the Seneca work hitherto known are from the twelfth century, it is of considerable interest to see how the text of the Natural Questions in these excerpts relates to the later MS. tradition.

The recent editors, Luck (Zürich, 1964), Hanslik (Leipzig, 1979), and Goold (Cambridge, Mass., 1990), allow into the text these emended instances of en

When Seneca comes to describe the appearance of the monstrous bull which appears out of the sea to kill Hippolytus in answer to his father's curse, he uses (among other things) a metaphor of birth: the sea's wave is said to be ‘heavy with burdened womb’ (1019f.: onerato sinu / gravis unda portat). If line 1016 is genuine – it was athetized by Leo – the sea is said to be ‘pregnant with a monster’ (tumidumque monstro). The metaphor has not passed unnoticed in modern commentaries but it has not been fully appreciated. I want to examine further linguistic aspects of the metaphor, and to consider its significance in its literary context. Seneca is a writer who likes – as he himself acknowledges – to re-work tradition: all the more significant, then, when he introduces a conspicuous metaphor into his text which does not appear to have literary antecedents.

E's ‘unam’ is the reading favoured by all modern editors. Either reading is possible in terms of sense and metre, and the choice is not an easy one. The attraction of ‘ unam’ is obvious and consists in the stress which it lays on Jocasta's isolation and vulnerability in the face of ‘omnis… iuuentus’. The appeal of ‘una’ is a little more subtle. It emphasises the coming together of the youth, whether Argive or Theban, in common cause against the aged Jocasta. It is ironical that the two fiercely opposed sides are urged to come together not against a mutual foe but against a defenceless old woman. And even more ironical is the notion of the brothers' uniting to kill their mother: from the beginning of the play they have been portrayed as uncompromisingly hostile to each other; now, bent on the impious act of fratricide, they are envisaged as acting together for the first time, but only to accomplish an equally great or even greater impiety – matricide. The reading of A is attractive not only because of the characteristic irony which it expresses, but also because it heralds the even stronger expression of the same notion in 445 with ‘ciuis atque hostis simul…’. One may note also that ‘omnis ruat / una’ appears to be an echo of Vergil's ‘una omnes mere’ (Aen. 8.689).

This line involves a variety of important points.

The late Sir Roger Mynors, in a letter to Sebastiano Timpanaro quoted in the latter's Contributi difilologia e di storia della lingua latina (Rome, 1978), p. 543 n. 15, states that he had wondered ‘whether it might be a habit of Latin writers, when they were putting only one or two “parolette” between two pieces of Greek’, to use Greek rather than Latin: he invents as an example ‘ἦθος κα πθος where logic demanded ἦθος et πθος’. The answer is that they sometimes did: the present paper will concentrate on the type instantiated by his imaginary example, the use of κα for et. I do not claim to have recorded every case, but those I have observed are the following.

Commentators on the Annals naturally observe that the famous first sentence of Tacitus' preface (‘Urbem Romam a principio reges habuere’) alludes to the preface of Sallust's Bellum Catilinae (6.1 ‘Urbem Romam, sicuti ego accepi, condidere atque habuere initio Troiani’). But it seems that none of them has observed a further allusion to Sallust's preface in the last sentence of Tacitus', which is almost equally famous (1.1.3)

Cicero's discussion of wit in the de oratore includes an entertaining story about Ennius and a certain Nasica (whom it is almost certainly wasted energy to attempt to identify) (2.275–6): ‘Valde haec ridentur et hercule omnia quae a prudentibus per simulationem subabsurde salseque dicuntur. Ex quo genere est etiam non videri intellegere quod intellegas… ut illud Nasicae, qui cum ad poetam Ennium venisset eique ab ostio quaerenti Ennium ancilla dixisset domi non esse, Nasica sensit illam domini iussu dixisse et ilium intus esse; paucis post diebus cum ad Nasicam venisset Ennius et eum ad ianuam quaereret, exclamat Nasica domi non esse, turn Ennius “quid? ego non cognosco vocem” inquit “tuam?” Hic Nasica “homo es impudens:ego cum te quaererem ancillae tuae credidi te domi non esse, tu mihi non credis ipsi?”’ This anecdote, devitalized by its divorce from a well–known name, finds a place in the Philogelos, a compendium of jokes compiled in late antiquity and ascribed to the otherwise unidentifiable Hierocles and Philagrius, (193): Δύσκολόν τις ζἠτει. δ πεκρίνατο∼ Ούκ εἰμì ὧδε. τοû δ γελσαντος καì εἰπόντος∼ Ψεύδῃ∼ τς γρ φωνς σου κούω – εἶπεν∼ ῏Ω κθαπμα, εἰ μέν δολός μου εἶπχες ἂν αὐτῷ πιστεσαι∼ γὼ δ σοι οὐ φαίνομαι ξιοπιστότερος κείνου εἶναι

The recent publication of a new edition of Dionysius Periegetes, the first since 1861, and the first ever to provide adequate information about the MS. tradition, has no doubt stimulated many of us to re-read this author, a rotten geographer but a competent versifier and recycler of Alexandrian flosculi. The new recension is a distinct improvement on Müller's, and the collection of citations and parallels is a valuable complement to the critical apparatus. Here are half a dozen suggestions for further amelioration of the text.

These notes continue the sequence begun in ‘Some Emendations in the Text of Maximus of Tyre, Dialexeis 1–21 (Hobein)’, published in CQ 41 (1991), 566–71. References to the text are by number, page and line in Hobein's Teubner edition; R is the principal MS., Parisinus graecus 1962, U is Vaticanus graecus 1390, I is Laurentianus Conventi Soppressi 4; U and I, being descendants of R (as are all other surviving MSS. of Maximus), offer conjectures not alternative readings. My thanks go again to Donald Russell and David Sedley for salutary comments on earlier and rasher drafts.

The extant MSS. of the Pervigilium Veneris, which all derive from a single archetype, are unanimous in their reading at line 74. Yet, as is widely agreed, this reading cannot be correct. The poet is describing the descendants of Venus

In L. D. Reynolds (ed.), Texts and Transmission: a Survey of the Latin Classics (Oxford, 1983), it is stated by J. G. F. Powell (p. 122) and by M. D. Reeve (p. 327 n. 4) that MS. Berlin, Deutsche Staatsbibliothek Lat. 4° 404 (s. ix), containing Cicero, De amicitia, and Publilius Syrus, Sententiae, is missing. Not so: it is currently held with numerous other Berlin MSS., including humanistic and musical autographs, at the Jagellonian Library in Cracow, where I saw it on 5 May 1992. Other classical MSS. from Berlin now held in Cracow are Lat. 4° 1 (Eutropius, Breviarium, and Victor Vitensis, Historia persecutions Africanae prouinciae, s. xi), 257 (Vergil, Bucolics 3.107–4.47, s. ix), 590 (Boethius, opuscula, s. xii), 939 (Boethius, Cons. Phil. with Remigius' commentary, s. x/xi; till 1935 Schloβ Maihingen, Fürstlich Öttingen-Wallensteinsche Bibliothek, I 2, lat. 4° 3), and 8° 102 (Statius, Achilleis, 1.375–513, 2.120–257, s. xiv).