The late Sir George Hill in the first volume of his monumental History of Cyprus remarks that the Cypriot syllabary is found in use until the third century B.C. This, it may be noted, is the traditional opinion which for some sixty years has stood the test of time. I read therefore with interest on p. 330 of the same volume, among the addenda, that ‘pottery with incised inscriptions, discovered in 1939 in an excavation four miles from Nicosia, shows that the syllabary continued to be used as late as the first century B.C.’ Sir George Hill in effect is here accepting the claim of Mrs. E. H. Dohan and Professor R. G. Kent, which he has hitherto ignored, that the syllabary survived until 50 B.C. Now this is an important claim, partly because it would add from two to three centuries to the life-span of a script already in possession of a long and reputable lineage; more particularly because, if true, it will convict the ancient Cypriot, admittedly a conservative individual, of a degree of conservatism with which I find it hard to credit even him. That there should still have been men in the hinterland of the island under the governorship of Cicero so little affected by the impact of the whole Hellenistic age that they were prepared to write Greek in a manner so uncouth and preserve, incidentally, certain elements of their old Arcadian tongue, is to my thinking highly improbable. A fresh examination of the evidence on which Mrs. Dohan and Professor Kent based their claim is clearly called for.

In re-opening the case of the authorship of the Hellenica of Oxyrhynchus I am afraid I must state at once that you will find that there is not much in it, at least nothing new. The few pages, recently published by Vittorio Bartoletti, from a papyrus book which evidently contained the same work as P. Oxy. 842 (whether they belong to the same papyrus is therefore a secondary question) have not changed the state of the problem. They confirm (which is important, though not surprising) the two primary facts known about the author in question: (1) that he is a continuator of Thucydides—a fact which was inferred at once from the very exceptional use of Thucydides' war-year, and which, incidentally, does not allow of the further inference that he intended merely to complement Thucydides, ending his work with the fall of Athens, as it is often assumed that Xenophon originally did; (2) that he was the main source of Ephoros for the period for which Thucydides was no longer available. But, unfortunately, they do not contain a title-page or subscription; and it does not help that in the remains of col. ii the anonymous author seems to quote Thucydides. Quotations of prose writers by name are extremely rare in the fifth and even in the fourth century, but they do occur: Herodotus quotes Hekataios for a special point, and Thucydides quotes Hellanikos for a special period, though it is worth while mentioning that both quotations are polemical.

The text of Martial is fairly good, though not as good as many editors seem to think. We have three separate sources from which our manuscripts are derived, which are called by Lindsay in the O.C.T. AA, BA and O.1 All three sources seem to have suffered from editing, AA less than the others. In A.D. 401 one Torquatus Gennadius seems to have emended the text (feliciter as the ‘puff’ in some of the BA MSS. puts it). We do not know whether all three sources were affected by this recension: all we know is that the BA family are the only manuscripts that mention it. But there are many signs of recension discernible in the readings of the other families too, e.g. pas-sages which have evidently had grammar or syntax thrust upon them—and in these I do not include the now obscure allusions and the dead jokes past recall.

In C.Q. xliii (1949), p. 126, Messrs. Klos and Minio-Paluello write: ‘Burnet's and Robin's collations of W… differ for the text of the Phaedo in about 130 readings of a more than orthographical interest. A new inspection of the manuscript has shown that Robin very often corrected Burnet, but added some twenty mistakes.’ As this may give a false impression of Burnet as a collator, it will be well to recall Burnet's own statement in C.Q. xiv (1920), p. 132: ‘He [Wilamowitz] says (p. 333, n. 1) that my collations of W are inadequate, and have been shown to be so by Schoene for the Symposium, and by Hensel for the Theaetetus and Politicus. As he refers his readers to my Prefaces (p. 332, n. 3), he has presumably read them, and ought to have known that I never collated W at all or even saw it.’

I. The present paper was read as one of ‘Three Notes of Doubt and One of Despair’ at the Oxford Philological Society in June 1948. Generally speaking, there is nothing to be said in favour of publishing an article on a text one is admittedly unable to under-stand. But in this case it has seemed advisable to start a discussion. I am therefore grateful to my friend P. Fraser for having consented to communicate his thoughts on the matter in an appendix to my note. We both hope that other scholars will be able to provide the satisfactory solution that has escaped us. Meanwhile we thank P. Maas, C. H. Roberts, and H. M. Last for their great help.

The Thesaurus lists a number of texts where demolior appears as a passive. To these should be added Cicero, Topica 4.22: ‘Ab efficientibus rebus hoc modo. Omnibus est ius parietem directum ad parietem communem adiungere vel solidum vel fornicatum. Sed qui in pariete communi demoliendo damni infecti promiserit, non debebit prae-stare quod fornix viti fecerit. Non enim eius vitio qui demolitus est damnum factum est, sed eius operis vitio quod ita aedificatum est ut suspendi non posset.’

There are three outstanding events in the internal history of Sparta during the sixth century. First, there is the constitutional settlement denning the functions of the Crown, the Senate, and the Assembly: this is now generally admitted to have taken place about 600 B.C. Secondly, there is the increase in the importance of the ephorate, a pseudo-democratic development associated with the ephor Chilon and the year 556. Thirdly, there is the decline in Spartan material culture; this process begins shortly after the turn of the century with the abandonment of gold and ivory work, and culminates shortly after the middle of the century in an almost total collapse of artistic and cultural standards.

It has usually been assumed that the Actian Games at Nicopolis were founded in 28 B.C. (cf. Reisch, P.-W. 1. i, s.v. ‘Aktia’). In Mélanges d'Arch. et d'Hist., 1936, pp. 94 ff., J. Gagé argues also for 28 B.C., his principal grounds being as follows: ‘Comme Auguste leur conféra en même temps le rang isolympique et que le calcul du temps par Actiades fut admis çá et lá á remplacer celui des Olympiades, il est logique de penser que ces deux computs coïncidaient. Or, la première célébration olympique après Actium est celle de l'an 28 av. J-C.; c'est done presque sûrement en 28 que commence en fait la première Actiade.’

This is partly a verbal question, depending on the meaning of the word ‘caste’. I propose to assume that if we say that a State is a caste State we imply at least two things:

(1) that its members are divided into mutually exclusive endogamous classes, and (2) that no one may be transferred from one class to another—unless possibly to a lower class. The State which Plato describes in the Republic satisfies the first of these conditions. Dr. Popper, who believes that it is a caste State, maintains that it also satisfies the second. In addition he contends that the original basis of the class division is racial. My object in the notes which follow is to argue that both these contentions are false.

See Lobel, Proc. British Academy xxxv. 1 ff., June 1950.

I. col. ii. 1. γυ[ ]… εῖδον [ο]ὐκ εἴκαсμ⋯ τι

εἰκ⋯сματι = εἰκαсμ⋯ι, ‘by guessing’, seems at least as likely as εἴκαсμ⋯ τι = εἴδωλ⋯ν τι ‘[At first I could not make out who or what it was: but when] I saw Gyges [clearly,] not by guesswork, I was afraid of a plot for murder.’ For example:

ΓΓ[γην сαϕ⋯][εἰ]εῖδον, [ο]ὐκ εἰκ⋯сματι, κτλ.

For the relation of εἴκαсμα to εἰκ⋯ζω, cf. κ⋯λαсμα, κ⋯μπαсμα, ὕβριсμα, etc.

Two firm points in the chronology of Antiphon's speeches are VI περ⋯ τοὖ χορευτοὖ in 419/81 and the Defence (περ⋯ τ⋯σ μεταστ⋯σεωσ) in 411/02. Speech V περ⋯ τ⋯σ 'Hρῷδου is now generally dated between these two; only the vaguest attempts have been made to date I κατ⋯ τ⋯σ μητρυ⋯ασ; there is no general agreement on either the date or the authorship of the Tetralogies. The main purpose of this paper is to adduce linguistic as well as external evidence for the dating of V in the neighbourhood of 414 and to show grounds for dating I between VI and V. Its secondary purpose is to apply some of the results gained in this inquiry to the fragments and the Tetralogies.

This papyrus was acquired, with some others, for the Egypt Exploration Society in 1914 at Medînet-el-Faiyûm by Dr. John Johnson. It consists of five large pieces; of these four join and together measure 37 × 24·5 cm.; they contain remains of three adjacent columns. Another piece measures 5 × 13·5 cm., and a small unplaced scrap (Fr. A) 4·5 × 3·7 cm. The writing of the recto, which runs along the fibres, is large, regular, rounded, and clear, but not elegant; I would assign it to the second century B.C. The verso was written across the fibres; the hand may be the same, but seems more cursive; the surface is very damaged and no whole word is legible. A certain irregularity in the spacing of letters and lines suggests that the papyrus was used again for a document.

The text of Euripides' Troades depends mainly upon the versions preserved in two manuscripts, viz. Palatinus 287 (P) and Vaticanus 909 (V). Both these manuscripts are well known and have been several times collated. Their importance in relation to one another and to the other existing manuscripts for the study of the text of Euripides in general has been frequently discussed and is no longer a subject of major controversy. The notes which follow are designed (a) to put on record a number of corrections to the readings ascribed to P and V for the Troades in the apparatus criticus of Prinz-Wecklein and later editors; and (b) to correct and amplify the general descriptions of P and to call attention to certain difficulties in the current views of its relationship to Laurentianus xxxii. (L).

What kind of Theory of Music and Theory of Metric was taught to the young Pindar or the young Sophocles? So far are we from an answer to this question that we do not even know how far extra study was necessary, or usual, for the professional poet as compared with the ordinary educated Greek citizen. The interdependence of music and metric in lyric poetry gave complexity to the word-rhythms but kept the study of music, the subordinate partner, theoretically simple. Doubtless by the time the young poet had learnt by heart the words of past masters of lyric verse, sung or accompanied them on the lyre, and danced them in choir or κ⋯μος, he had absorbed by practice and somewhat rule-of-thumb methods of training a great deal of the τ⋯χενη which his calling would require. He would have to learn, in common with the χοροδιδ⋯σκαλος, how to read and write a simple score, and so he must know the symbols of pitch in such scales as were then in use. The notation of arsis and thesis may have been required too, and if we knew that this was so, and how he was taught to apply it, we might be in a much better position to assess the kind of theory he learnt. But of one thing we may be sure: genius apart, it was his practical training, as performer and spectator, rather than theoretical teaching which gave his ear its cunning and technical skill to compose new rhythms.

(i) In Epictetus‘ chapter on Providence (1. 16), the eighteenth section contains an evident corruption: ταντα επ’ εκαοτον επνειν εδει και και νον μεγιστον υμνονεπνμνειν…

The duplication of cannot be genuine. It is not an iteration of the kind which heightens the effect of a passage: it just falls flat. Wilamowitz, in his Lesebuch, printed in the first place. This is an improvement, but I doubt whether this conjecture really settles the point. The duplication is stressed, and not eased, by the noun and Epictetus is hardly likely to have followed the verb by the homophonous synonym At one of the two places, then, the scribe missed the original verb by duplicating the other. I believe that consideration of the context can show what the original was and from which place it was ousted.

It is proposed to account here for a number of words with closely allied forms but different meanings: περιοτερα ‘pigeon’ περωτερις ‘vervain’ and ‘a woman's ornament’ ‘vervain’, ‘a woman's ornament’, and diminutive of περωτερις There are several other by-forms, as may be seen by reference to LSJ, but they add nothing to the well-established partition of meaning in this stem between the three senses ‘pigeon’, ‘vervain’, and ‘a woman's ornament’. It is my belief that we must explain the three senses together, and this I shall attempt to do.

During the excavations of 1924–5 at Karanis a papyrus (Michigan 2745) of the second or early third century A.D. was discovered, and subsequently published by J. G. Winter (‘A New Fragment on the Life of Homer’, Trans. Am. Philol. Ass., lvi, 1925, 120 ff. and pl. A), which under its single column has a subscribed title which should almost certainly be restored as ‘Alcidamas, On Homer’. The first fourteen lines of the papyrus give most of the story of Homer's death and the riddle that caused it, which is common to all the extant Lives of Homer; the remainder is a general eulogy of Homer and a profession of transmitting his works to posterity. The interest of the discovery lies in the knowledge that it gives of a hitherto unrecorded work by Alcidamas, the rhetorician and contemporary of Isocrates, and the new fuel that it provides for an old controversy about the origins of the work known as the Certamen. The first part of this article aims at both re-examining the value of the papyrus and reopening some of the old questions on the Certamen.