Most of the work done on tracing persistent themes and images in the Oresteia has failed to take account of the associations of the theme or image for the original audience. Some of these associations are with certain highly emotional rituals. In evoking the ritual the poet evokes also some at least of the emotion which generally accompanies its performance. I will take here as an example the association of the manner of Agamemnon's death, the fatal bath and the fatal robe, with the ritual of the funeral. This will I hope help to enrich our own emotional reaction to Aeschylus' presentation of this event, as well as to shed light on certain problematic passages.

The line in question occurs towards the end of Orestes' final exchange with Clytemnestra, after her attempts at self-defence have all met with rebuttal.

The word Eὐμεν⋯δες occurs six times in our texts of Greek tragedy (four times in Eur. Or., twice in Soph. O.C.) and once as a play title (Aesch. Eum.). This may make ‘Eumenides in Greek tragedy’ sound like a restricted subject, but it is one that has seldom been discussed as a whole, and scholars have tended to consider each of the three plays in question in the light of unargued assumptions about the other two, and about the nature and affinities of Eumenides in general. I shall begin with some introductory remarks on cults, since not all the information is readily available in reference books, and then discuss the three plays in turn, starting for convenience with the Orestes.

A few years ago, as part of an investigation of the authenticity of Prometheus Bound, I published figures for the occurrence of non-Aeschylean words in that play, as compared with two undisputedly Aeschylean plays (Th., Pers.) and with one Sophoclean play (Aj.) The figures showed that Prom, contained a greater number of words not found elsewhere in the surviving plays of Aeschylus (Eigenwörter); and also that, like Soph. Aj., but unlike the six undisputed plays of Aeschylus, it contained a relatively large number of Eigenwörter that occur more than once in the play, some of them quite common and familiar words. The discrepancy between Prom, and the two Aeschylean plays chosen for comparison was quite marked; and the high rate of repeated Eigenwörter for Soph. Aj. seemed to confirm that this criterion might be a good one for distinguishing unAeschylean characteristics in a non-Aeschylean play. But the sample was very small, and I was conscious that the criterion (or the particular application of it) was somewhat arbitrary; I was therefore not very confident of the value of my findings.

More Than a quarter of a century has elapsed since the publication of the Oxyrhynchus papyrus which Lobel identified as a fragment of Sophocles’ Inachus, and though it has revolutionised our knowledge of the play, it has proved an excellent example of the papyrological commonplace that each new discovery creates more problems than it solves. What could with reasonable confidence be inferred about the Inachus from the comparatively numerous ancient quotations and allusions is well summarised in Pearson's introduction: Inachus, Hermes, Argus, and (almost certainly) Iris took part in the action, but beyond the fact that the play told of Io's transformation and of blessings bestowed on Argos as a result of Zeus' visitation (not necessarily in that order) the development of the plot could only be conjectured. A majority of those who have discussed the play have shared Hemsterhuys' view that it was satyric, and the publication in 1933 of a Tebtunis papyrus preserving part of a dramatic composition dealing with Zeus’ love for Io and strongly suggesting a satyr-play was welcomed by many as confirmation of this conjecture. But the ascription of the Tebtunis fragment to Sophocles is far from certain, and its testimony will not allay the doubts of those who are troubled by the lack of direct evidence for satyrs in the Inachus and by the feeling that the importance of the moral and theological issues raised makes its theme unusually serious for satyric drama.

Troades has often been thought to lack any coherent structure, and this has been variously attributed to its being the last play of the trilogy and to Euripides' overriding concern to impress the horrors of war upon his fellow Athenians. More recently, however, attention has been drawn to how the constant presence of Hecuba gives unity to the play and to how it is articulated by the striking entries of Cassandra, Andromache, and Helen. Cassandra and Andromache enter in mock triumph, Cassandra waving torches in her ironical wedding song and Andromache on a waggon, while Helen is dragged out by force and her scene marked off by Menelaus' second prologue. All three women emerge from the tent, argue with Hecuba but fail to convince her, and depart for marriage in Greece. Andromache's ideal marriage to Hector (645–56) contrasts with Cassandra's perverted ‘marriage’ to Agamemnon and with Helen's destructive marriage to Paris. The departure of the Trojan women for marriage in Greece balances Helen's earlier departure from Greece for marriage in Troy.

Aristophanes' last two surviving plays, Assemblywomen and Wealth, have long been regarded as something of an enigma. The changes in structure – the diminution in the role of the chorus, the disappearance of the parabasis, etc. –, as well as the shift of interest away from the immediacies of current politics towards broader social themes, can reasonably be interpreted as an early stage of the process that ultimately transformed Old Comedy into New, even if it is unlikely ever to be finally agreed whether Aristophanes was leading or following this trend. The decline in freshness, in verbal agility, in sparkle of wit, in theatrical inventiveness, which is perceptible in the earlier play and very marked in the later, may be put down to advancing years and diminishing inspiration. Such an explanation squares with the evidence of a marked decline in Aristophanes' productivity towards the end of his life. Whereas in the first seven years of his career (427–421) he seems to have produced, or had produced for him, not less than ten plays, and in the years 420–405 approximately another eighteen, the twenty years or so that followed Frogs yielded a further eleven at the very most unless some titles have been completely lost; and since it is not likely that after the outstanding success of Frogs, and the public recognition that followed it, Aristophanes would have experienced any difficulty in securing a chorus, the explanation can only be that he was writing less. But the truly puzzling feature of the two late plays we possess is the apparent sea-change in the author's social orientation. In his fifth-century plays, from Acharnians toFrogs, as has been shown (in my view conclusively) by de Ste Croix, Aristophanes reveals himself as one who instinctively speaks the language and thinks the thoughts of the well-to-do, even if at the same time he can laugh with the common man at ostentatious and useless wealth in the shape of Pyrilampes' peacocks, Leogoras' pheasants or the sultan-like garments of an Athenian imperial official – as one who was happy for the Demos to be sovereign so long as it was willing to be guided by the advice of its' betters', the καλоί тε κáγоθоί of (e.g.) Knights 738 or Frogs 727–9, and to leave them in the quiet enjoyment of their property. At first sight in Assemblywomen and Wealth this seems to have changed almost diametrically.

The reader of Demosthenes can hardly avoid the impression that there was something sadly awry with the Athenian naval system in the two decades prior to Chaeronea. The war in the north Aegean was essentially a naval war, and Demosthenes frequently enough blamedAthen's failure on her lack of preparation. ‘Why do you think, Athenians,… that all our expeditionary forces are too late for the critical moments?…In the business of the war and the preparation for it everything is in disorder, unreformed, undefined. So it is that some news arrives, we appoint trierarchs, we give them the opportunity legally to get others to act in their stead, we consider provision of money, then it is decided that the metics and those “living out” should go on the ships, then again you yourselves, then to allow substitutes, and during these delays^the cause for which we would have been sailing is lost before we have sailed] For we spend the time for action on preparations, but the critical moments for action do not last as long as our dilatoriness and cautious procedures’ (4. 35 ff.). This was spoken in 351 and it is much the same a decade later (8. 10 f.).

The view that Pausanias 9. 13. 1–15. 6 is a simple epitome of Plutarch's lost Epaminondas, first advanced by Wilamowitz in 1874 and later elaborated by Wilamowitz himself and by Ludwig Peper, has commonly been accepted, with little or no further discussion, by students of Plutarch, Pausanias and fourth-century history. In a recent general reaffirmation of the thesis John Buckler does note that what Pausanias says about Mantinea is hard to square with Plutarchan evidence and he therefore admits some contamination of the hypothetical source with non-Plutarchan material. But Buckler's discussion of the matter within the framework of a book on the Theban hegemony is necessarily brief and somewhat unsystematic, and may be thought over-ready to assume Plutarchan connections where none can be demonstrated. The purpose of this paper is to show that the Wilamowitz/Peper hypothesis in its pure form (as accepted apparently by such authorities as Ziegler and Sandbach) cannot reasonably be sustained, that at the very least some degree of contamination must be presumed, and that the existence of a direct connection between Pausanias and Plutarch can hardly be regarded as established beyond reasonable doubt.

The discussion of mathematical knowledge and its relation to the construction of an appropriate diagram in Aristotle's Metaphysics Θ 9. 1051 a21—33 is an important, if compressed, account of Aristotle's most mature thoughts on mathematical knowledge. The discussion of what sort of previous knowledge one must have for understanding a theorem recalls the discussion at An. Post. A 1. 71 a 17–21, where the epistemological point is similar and the examples the same. The first example, that the interior angles of a triangle equal two right angles, appears no less than thirty times in the corpus (inter alia, An. Post A 5. 74a 16, 23.84 b 6–9, 24.86 a 22–30). The example of the angle inscribed in a semicircle being a right angle also occurs at An. Post. B 11. 94a 27–34, but in a very different context from its two companions. Illustrations of both theorems provided clear stock examples for Aristotle.

Scholarly opinion about Ptolemy Soter's history of Alexander has been far from unanimous. Not long ago Ptolemy was held to stand in the first rank of ancient historians. His history was described as brilliant, rational, straightforward, and exhaustive, while he himself was proclaimed a ‘second Thucydides’. In recent years, however, Ptolemy's reputation has seriously declined. His shortcomings, acknowledged also by his admirers, have been stressed and extensively analysed. Fritz Schachermeyr clearly reflected current opinion when he equated a ‘version from the Hauptquartier's circles’ with a lie, a fraud, and an intentional omission. The purpose of this paper is to examine the recent reassessment of the nature and the aims of Ptolemy's work.

The purpose of this paper is to assess the validity of some methods currently being used to interpret the demographic evidence from the ancient world. For example, it has been claimed that during the Hellenistic and Roman eras, birth rates were 40/1,000/year, death rates 36/1,000/year, and that 10% of healthy infants were killed, raising the death rate to 40/1,000/year; the claims rest on comparative material and anecdotes from literary sources. This paper will question (I) the use of comparative material from modern primitive societies as ‘evidence’ for Greece and Rome, and (II) the value of anecdotes for elucidating ancient demographic structure and population policies.

I offer some notes on the text of Propertius. In the apparatus to individual passages Ω is employed to indicate the archetype, i.e. the consensus of N and A. Only two quires of A are extant, and after 2. 1. 63 its place is taken by descendants: F, L (2. 21. 3 ff.) and P. These derive more immediately from a manuscript of Petrarch (here called Π), copied from A in the Sorbonne in 1333, and now lost. The delta mss. (DVVo.) I discard; they are too interpolated and contaminated to aid in the reconstruction of Π. They may contain some elements of truth independent of N and A/Π, but J. L. P. Butrica has shown that another group of mss. headed by Vat. Lat. 3273 has a stronger claim to be cited in the apparatus for these readings, as it antedates Δ. However, as the matter is controversial and not of vital importance for my discussions, the group is here ignored. In the absence of a satisfactory arrangement of the humanistic conjectures into groups akin to those introduced by Mynors in his Catullus, I use the age-old catch-all ς for unauthoritative readings found in fifteenth-century mss.

In his discussion of Virgilian imitations of Lucretian phraseology Cyril Bailey examines the phenomenon of what he terms the ‘doublet’, that is, the procedure whereby Virgil imitates separate elements of a Lucretian phrase at different points in his own work. Take, for example, De Rerum Natura 1. 210–12:

esse videlicet in terris primordia rerum,

quae nos fecundas vertentes vomere glebas

terraique solum subigentes cimus ad ortus.

In April 1741 there appeared a slim volume entitled An Apology for the Life of Mrs Shamela Andrews by a certain Mr Conny Keyber, whose name is generally supposed to conceal that of the novelist Henry Fielding. Shamela, to give the book its more familiar title, was a parody of Samuel Richardson's epistolary novel Pamela: or Virtue Rewarded, which had been published to great acclaim the previous year. In a series of letters purportedly sent to each other by the main characters, the story unfolds of the honest servant-girl Pamela, her efforts to avoid seduction by her master Mr B., and her eventual marriage to him. Fielding's chief target was the morality of the book (Pamela's virtue contains a disturbingly large element of self-interest), but in passing he drew cruel attention to some of the pitfalls of the epistolary form as a vehicle for narrative. One passage in particular deserves quotation, from Letter VI, which Shamela writes to her mother at (so we are duly informed at the top of the letter) twelve o'clock on Thursday night:

Mrs Jervis and I are just in Bed, and the Door unlocked; if my Master should come – Odsbods! I hear him just coming in at the Door. You see I write in the present Tense, as Parson Williams says. Well, he is in Bed between us, we both shamming a Sleep, he steals his Hand into my Bosom, which I, as if in my Sleep, press close to me with mine….

Discussions of the constitution of the Principate are usually focused on the powers of the emperor, and relatively little attention is given to the role of the senate; by exception much has been written on its jurisdiction, with which I shall not be concerned. Despite his theory of a dyarchy of emperor and senate, which I do not wish to revive, Mommsen, partly because he devoted separate volumes to each, did I not exhibit the extent to which Augustus and Tiberius at least worked through the senate, and on occasions attributed to them legal powers to act by their own authority, when in reality (as I shall argue) they caused the senate to take action as the only proper means of effecting their wishes. More recently, F. de Martino in his admirable account of the Principate dedicates only one out of twenty-seven chapters to the composition, functions and procedure of the senate. For Syme it was simply an ‘organ that advertised or confirmed the decisions of the government’. This description does not bring out the truth that it was performing a role essential to Augustus' design. For though in effect he founded a monarchy, he commonly thought it expedient on necessary obtain for his measures senatorial approval. That this was the practice of Tiberius in his early years is clearly attested, and some suppose that he behaved in an entirely different manner from Augustus. This view seems to be mistaken, and the mistake is of some consequence.

Velleius text rests on a single manuscript, found at Murbach in 1515, which long ago disappeared. We know of it from three reports: (a) the editio princeps (P) by its discoverer Rhenanus, published in 1520, apparently from an inaccurate copy taken by an anonymous friend;1 (b) notes on the manuscript taken by Rhenanus' secretary Burer (B), who compared it with proofs of the edition; and (c) a copy (A), probably from the same source as P, taken in 1516 by B. Amerbach and discovered by Orelli in 1834. Of this text one editor has written: 'While modern scholarship has made progress in solving its enigmas, the text of Velleius, unless some long-hidden manuscript shall unexpectedly come to light, will always continue to be one of the most corrupt among the surviving texts of classical authors.

The text of each passage commented on is that of Housman except where otherwise stated. The following editions of Lucan or other works concerned with him are indicated by the scholar's name only: (i) Text: A. E. Housman (Oxford, 1926). (ii) Text with commentary: F. Oudendorp (Leiden, 1728); P. Burman (Leiden, 1740); C. H. Weise (Quedlinburg and Leipzig, 1835); C. E. Haskins (London, 1887); R. J. Getty, Book 1 (Cambridge, 1940).

In the period between Constantine's reunification of the Empire in 324 and the deposition of Romulus Augustulus in 476 M. Valerius Probus enjoyed a large reputation as master of all areas of the ars grammatica. The commentary on Terence attributed to Donatus and the commentary of Servius on Virgil cite him more often than they do any other ancient authority. His fame persisted through the Dark Ages. Eugenius of Toledo set him with Varius and Tucca against Aristarchus, the greatest of the Alexandrian students of Homer. Modern writers on the history of Roman scholarship have estimated in widely different ways his quality as a textual critic, the level of his reputation during the century after his death and the influence which his activities had on the transmission of the Latin classics. That he ‘annotated’ at least some of these in the manner of an Aristarchus is not in dispute, but everything about the nature of his ‘annotation’ is. This paper will treat afresh a famous statement about Probus in Suetonius' De grammaticis (24. 3), two lists of notae associated with Probus’ name in a late eighth-century manuscript from Monte Cassino, cod. Paris, Bibl. Nat. lat. 7530 (CLA v 569), two references to such notae which have been detected in Virgilian scholia (Serv. Aen. 10. 444 and Serv. Dan. Aen. 1. 21) and a number of statements in these scholia which appear to give Probus’ reasons for affixing notae. The results of my study are largely negative but may help to control general discussion of the history of a number of Latin texts.

Chapter 57 of the Periplus Marts Erythraei is our sole source for a nugget of vital information: how westerners learned to sail over the open sea to India. Unfortunately, emendations distort the Greek in all editions1 and misconceptions the rendering in all translations.