This article is intended not only to record additions and corrections to Island Gems (published by the Hellenic Society in 1963 as its tenth Supplementary Volume; here abbreviated IGems), but also to discuss a class of engraved stones which was not properly distinguished in that book. The main series of Island Gems belongs to the seventh and early sixth centuries B.C. Most of the stones and a few of their devices copy Bronze Age forms, and the material is generally a distinctive, often translucent, green serpentine (‘steatite’). This series ends in the first half of the sixth century, but it was possible to identify, by their material and technique, some later gems from Island workshops which correspond in style and shape with the contemporary Greek scarabs in harder materials. The comparatively soft material had meant that the intaglio devices on Island gems could be cut without recourse to the drill or cutting wheel, and this technique was retained for the later scarabs. It is evident, however, that for a while already in the sixth century Island artists had experimented with the harder materials then being employed for engraved seals in Greece; and with some shapes which seem to compromise between the old and the new. These stones seem to belong to the middle and second half of the sixth century. A list follows:

(i) Oxford 1925.129. plate I. Green steatite tapered scaraboid with convex face. L. 20 (mm.). A lion. IGems no. 349, fig. 6, pl. 13.

(ii) Athens, from Sunium. plate I. Green steatite. Shape as the last. L. 20. Contorted bull-headed man. IGems no. 350, pl. 13.

(iii) Paris, Bibl. Nat., ex Louvre C 8514. Rock crystal plump lentoid. W. 18. Summary representation of a winged horse.

(iv) Boston 27.678 (once Bruschi, Warren), plate I. Chalcedony lentoid with domed back and shallow convex face. W. 21. Facing head of a satyr with fillet ends behind the ears and an arrow marking at the centre of the forehead.

Many studies of the campaign and the battle of Marathon have suffered from conflicts between scholars in the field of hypothesis. The article by J. Kromayer illustrates this admirably. He concerned himself mainly with the rival hypotheses of Curtius and Delbrück, the former maintaining that the Persians embarked most of their force and all the cavalry just before the battle, and the latter that the Persians did not do so at all but delivered a full-scale but unsuccessful attack up the Vraná valley, which resulted in a counter-attack by the Greeks. Neither hypothesis rests upon the ancient evidence. Yet Curtius has been followed by Munro, Grundy and others, and recently by Gomme, Pritchett and Burn; and Delbrück has been followed by Meyer, De Sanctis and others. Other hypotheses are made about the duration of the engagement. They vary from Munro's matter of ‘minutes’ to Delbrück's three phases of hard-fought action, although they are both in conflict with the evidence of Herodotus. Again, hypotheses have been advanced in an attempt to dispense with the topographical evidence, for instance of the Mound at Marathon, e.g. the hypothesis that it existed before the battle. In a paper delivered in 1920 and published in this Journal in 1964, Whatley expressed his doubts about the value of such hypotheses; but he himself became involved in drawing analogies between the massive, complicated and many-fronted First World War and the one-day battle of Marathon—analogies which are quite misleading. In this paper I propose to be as economical as possible in making hypotheses and to keep to the ancient evidence first. This leads to a different order of exposition; for most scholars have begun with the campaign, formed their theory of the aims of the Persians and of the Greeks, and tried to make the battle conform with the theory, but I shall begin with the battle itself, for which we have much evidence, and treat the campaign afterwards.

The editor having invited me to reply to Professor Bernard Ashmole's ‘New Interpretation of the Portland Vase’, the reader will, I hope, forgive me for returning to a topic which he may think in danger of being worn threadbare. Although my own interpretation has been generously characterised by Professor Ashmole as ‘persuasively argued’, its failure to convince not only him but Professor Hans Möbius and Professor F. L. Bastet shows that it was not persuasive enough; and the aim of the following pages is to re-state the case for it and to fortify it, as I hope, with additional arguments which would have overloaded the popular publication in which it first appeared.

I begin, as before, with the young man who advances to the right in front of a small structure consisting of two square pillars raised on a continuous base and supporting a Doricising entablature (Plate IV). According to Möbius and Ashmole this structure is a gateway and the young man has just passed through it. ‘Ask yourself’, Ashmole bids the reader, ‘where he can have come from if not through the gateway, and you will see that this must be the correct solution.’ But must it? If the young man had come through the supposed gateway, surely the foot he first set to the ground—his left—would be more or less on the axis of the opening and not, as it actually is, with the heel almost touching the corner of the base nearest the spectator: an artist in whose work ‘no detail is negligible’ would hardly have put a foot as wrong as that.

Thucydides' account of the events at Melos in 416–15 B.C. falls into two parts, the famous dialogue in which representatives of Melos and of Athens discuss the submission of Melos, and a series of notes about the siege of the city culminating in the account of its destruction. But as I shall try to show, the two sections form part of a single whole. The discussion between the negotiators centres on two topics. In the first half of the dialogue the speakers discuss the expediency of forcing Melos into the Athenian Empire, in the second they discuss the likelihood of the Melians resisting successfully. But since the Melians are offered no alternative to becoming subjects except complete destruction, and since they are clearly not ready to choose the safe but dishonouring alternative, even though they have no chance of defending their city successfully, the inevitable destruction of Melos casts its shadow over the whole of the negotiations.

Many of the arguments used in the discussion are equally relevant to the destruction and to the subjection of Melos. This is partly a result of the Athenian aim: to impress their island-subjects with their power. This they can achieve by forcing Melos to become a subject—but equally well by destroying it. On the level of expediency the Athenian argument would be equally applicable to either course, and the tactlessness of the Athenians suggests that they are not much concerned which of the two they will adopt. The Melians on the other hand appear to anticipate their own rejection of the ultimatum and to include the consequences of this inevitable rejection within the scope of their arguments. So a debate about the expediency of forcing an independent state to forgo its freedom is at the same time a debate about the expediency of destroying an independent city that refuses to become a subject.

Widely differing opinions have been expressed on Plato's place in the history of Greek science, as on many other features of his work. The view that Plato's attitude and influence were nothing short of disastrous for science has been widespread. Platt, for instance, put it that ‘Plato, being first and foremost a metaphysician with a sort of religious system, would not have us study anything but metaphysics and a kind of mystic religion’. Dampier-Whetham believed that ‘Plato was a great philosopher, but in the history of experimental science he must be counted a disaster’. In his History of Ancient Geography J. O. Thomson spoke of Plato's ‘positive contempt for observation, upon which natural science rests’, and in his influential book on Greek Science Farrington put it that ‘from the scientific point of view the Timaeus is an aberration’.

On the other side Plato has found almost as many defenders from among those who were primarily philosophers, or philosophers of science, or even practising physicists, as from among the specialist Greek scholars themselves. In such books as Science and the Modern World (1926) and Adventures of Ideas (1933) Whitehead granted that Plato was responsible for diverting interest from the observation of particular facts in physical science, but suggested that ‘Plato had another message … An intense belief that a knowledge of mathematical relations would prove the key to unlock the mysteries of the relatedness within Nature was ever at the back of Plato's cosmological speculations’. There is a sense, then, in which ‘Plato and Pythagoras stand nearer to modern physical science than does Aristotle’. In 1927 Shorey quoted Whitehead and substantially developed this line of interpretation in his paper ‘Platonism and the History of Science’, still probably the most outspoken vindication of Plato's reputation as a scientist.

In the earlier years of the nineteenth century a few scholars questioned whether there was any direct relation between Anaxagoras and Empedocles. Since then these doubts have been voiced but rarely. Instead, opinion has been divided principally on which thinker was dependent on the other. Most scholars, following Zeller and Burnet, have thought that Anaxagoras was dependent on Empedocles. A minority, including Tannery, has held the opposite view.

It is essential to distinguish two questions. Who wrote first? Did he, whichever of the two it was, influence the other? Both questions are answered by a piece of external evidence, which since Zeller's second edition has been largely neglected: Alcidamas is reported by Diogenes to have said that Empedocles ‘heard’ Anaxagoras. This should imply both that Anaxagoras wrote earlier than Empedocles, and that he influenced him. Examination of passages in Aristotle and in some later writers, notably Plutarch, confirms the priority. Internal evidence may be applied to the question of influence. I shall seek to argue in particular that Anaxagoras' and Empedocles' theories of vision are such that they imply the influence of whichever is the earlier philosopher on the later.

In a study earlier in this volume, ‘The Relation of Anaxagoras and Empedocles’, pp. 93–113, I listed the ancient evidence to the effect that Anaxagoras first gave the correct explanation of an eclipse, and that he was followed in this by Empedocles. A more extensive examination of the evidence raises certain difficulties. For what are, or might appear to be, Anaxagoras' theories are attributed elsewhere to earlier thinkers.

There are two principal elements in this contradiction, the one direct and the other indirect.

1. There is a direct contradiction when Thales, Anaximenes and some Pythagoreans are said to have given the correct explanation of an eclipse, at least if we suppose the Pythagoreans in question to have been earlier than Anaxagoras.

2. There has been thought to be an indirect contradiction when several thinkers before Anaxagoras are said to have derived the moon's light from the sun. For a theory of derived light for the moon has been thought, whether rightly or wrongly, to entail the correct explanation of an eclipse.

In what follows I shall attempt to solve these, and some other incidental difficulties.

The speeches of Critias and Theramenes form the central tableau of Xenophon's account of the reign of terror of the Thirty at Athens in 404–3 B.C. The purpose of the present paper is to examine the style and content of these speeches, in order to discover to what extent they may have been based on the texts of the original speeches. The vividness and pathos of the narrative of the siege of Athens, her surrender, the establishment of the Thirty and their expulsion have seemed to some commentators to have the quality of personal reminiscence. Recently the case for Xenophon's having been a witness of some of the events of this period has been argued anew, and in the present writer's view convincingly, by E. Delebecque, who refers to several passages which seem to convey personal impressions of events. Delebecque also argues, to the satisfaction of Sir Frank Adcock, that Xenophon composed this part of the Hellenica before setting out on the Cyreian Expedition. The acceptance of both these theses would add some weight to the present argument; but the possibilities here envisaged do not depend upon their support.

According to Xenophon's account, both speeches were delivered before the βουλή. The members of this body had been chosen by the Thirty, so that it seems certain that they were men who were known to favour the oligarchy. But whatever were Xenophon's political views at the time, it is unlikely that he was old enough to be a member in 404 B.C. Hence it is improbable that he heard either speech. Four possibilities may therefore be considered:

(1) That the substance of the speeches was reported to him from memory by a member, who perhaps took special pains to remember some of the purple passages.

(2) That his informant took the speeches down in shorthand.

(3) That Xenophon obtained complete copies of the speeches, and condensed them. (Both speeches as they stand are too short to be historical.)

(4) That the speeches in the Hellenica are entirely the product of his imagination.

In a review (JHS lxxxvi [1966] 205 f.) of P. A. Stadter's Plutarch's historical methods: an analysis of the Mulierum virtutes (Cambridge, Mass.: Harvard University Press, 1965), Mr T. F. Carney, who was at that time Fulbright Visiting Scholar in the Department of Political Science of the Massachusetts Institute of Technology, claimed that the book ‘analyses somewhat dated problems with only partially up-dated tools’ and that the ‘actual findings of the book replicate what is already known’. Although Mr Carney takes it for granted that his readers know what the ‘up-dated tools’ are, at the end of his review he drops the key words ‘content analysis’, in which, according to him, Stadter had engaged unwittingly, but obviously without success.

Inasmuch as the book in question grew out of a dissertation suggested and directed by this writer, it may perhaps be allowed to reply to the rather serious charges raised in this review, charges which are unprecedented in the history of this Journal and which, if they were justified, could have far-reaching implications. This writer must confess that he was first baffled both by the degree of hostility displayed and by the criticisms, which, though specific, are often murky; but he was baffled even more by the general accusation of ‘only partially up-dated tools’ and inappropriate use of ‘content analysis’, a term he had never previously encountered. But this writer soon found out that content analysis is ‘a research tool in mass communication’ (R. W. Butt and R. K. Thorp, An Introduction to Content Analysis [A Publication of the University of Iowa School of Journalism, Iowa City, Iowa, 1963] 1), ‘a research technique for the objective, systematic, and quantitative description of the manifest content of communication’ (Bernard Berelson, Content Analysis in Communication Research [The Free Press, Glencoe, Illinois, 1952] 18 [still the chief book in the field]); and it became clear that Mr Carney wore the social scientist's hat when he wrote his review.

Mr Bloch's note contained other surprises for me beside that of discovering that I'd been in the Political Science Department of M.I.T. There does seem to be a real issue, however—propriety of use of a technique like content analysis—so reply seems indicated, unaccustomed as I am to this form of public speaking.

Firstly, logic. My reference to content analysis was specifically limited to Mr Stadter's chapter four (third paragraph of my review). Mr Bloch generalises this, apparently (see his paragraph one), to refer to the whole review—certainly he does not deal specifically with my criticisms in paragraph two, though they involve specific, substantive points. My paragraph two dealt with semantics, a separate, if related, issue. Let's not muddle them up here. Some inconsistency seems involved in arguing that ‘content analysis has nothing to contribute’ (Mr Bloch's paragraph three) while yet ‘students of literature and historians have for generations pursued “content analysis” as a major aim of their professions' (paragraph four). I note also that Mr Bloch was able to define the use of content analysis as ‘inappropriate’ even when it was ‘a term he had never previously encountered’ (paragraph two).

In modern books on Greek Athletics the Greek word τριαστής is often used of a triple victor at a meeting, but it is not to be found in LSJ. An enquiry into its authenticity proved interesting. The authors of this century who use it are strangely coy about giving any authority for it. Jüthner simply calk the three victories of Phanas of Pellene in one day at Olympia ‘Eine Leistung, die den Ehrentitel “τριαστής” eintrug’. Moretti, usually meticulous in giving references, writes, ‘E noto che chi conseguiva questo risultato era chiamato τριαστής’. In his later Olympionikai he applies the title to Phanas, to Astylus of Croton and Syracuse, to Leonidas of Rhodes and to Hecatomnus of Mitylene, but it is not made clear to which of the many authorities for these athletes whom he cites he owes the word. E. N. Gardiner seems to have been the first modern author to use the Greek word. In his Greek Athletic Sports and Festivals (1910) we read of ‘Phanas of Pellene, who, by winning three races at Olympia in one day, won the tide of triple victor (τριαστής)’. Later in the same book he states, ‘A still finer record is that of Leonidas of Rhodes, who won all four foot races in three successive Olympiads, 164–156 B.C., thus three times earning the title of τριαστής or triple victor, given to those who won the stade race, diaulos and dolichos'. [Gardiner has slipped up in his facts here. Leonidas did not win four events at three Olympiads but three events at four Olympiads.

The Curator of the Museum of Greek Archaeology at Reading University, Mrs A. D. Ure, recently drew our attention to an apparendy unnoticed Etruscan inscription on a fragment of an Attic red-figure stemless cup datable to the third quarter of the fifth century B.C. Mrs Ure further referred us to the classification of the cup in J. D. Beazley, Attic Red-figure Vase-painters2 (Oxford, 1963) 1297 (no. 7), the Painter of London E 122.

The tondo inside the cup (plate VIIa) is practically complete (diam. 12.5 cm.), but all the external decoration is lost except small parts of the bottom of the floral pattern under each handle. Inside the tondo are the figures of two women, one seated in a chair and holding a sceptre, the odier standing frontal, looking to the left towards the seated woman and holding out a box in her direction, though she does not seem to be actually offering it to her. The underside of the base has painted rings and a moulded black circular band 1 cm. wide (Plate VIIb).

An inscription in Etruscan letters is scratched on the underside of the base (plate VIIc). The letters run from right to left, as is usual in Etruscan inscriptions; we transliterate cipisvinas (cf. the tables of Etruscan alphabets in G. Buonamici, Epigrafia etrusca, 1932, 122–3). There will be no doubt about the genuine Etruscan character of the word. One notices the vowel harmony, an often attested feature (M. Pallottino, Etruscologia,2 1947, 287).

In Volume lxix of this Journal I drew attention to three little vases in Reading University belonging to the group published by Wide in AM xxvi (1901) 143 f., at that time regarded as Boeotian and connected by Wide with Mykalessos. They are now known to be Corinthian, since vases and fragments in the same style have been found in the Potters' Quarter at Corinth and are now displayed in Corinth Museum. The general appearance of the vases from this workshop is far removed from that of normal Corinthian, which in the fifth century was decorated largely with floral or linear patterns. When compared with Corinthian red-figure and other vases with outline drawing the childish aspect of many of the figures, their long heavy eyelashes, dimples, chubby limbs, feet with toes on the side facing the spectator, all combine to set them in a field apart. Some of the subjects too are unusual. Among the divinities there is Demeter enthroned with torch, corn and poppies on a plate in Athens; bearded Herakles with club and bow inside cups in Reading and Athens; a youthful Herakles, weary and thirsty, leaning on his club as he fills his cup at a fountain, on a pyxis in the British Museum; Dionysos as Liknites, Winnower, horned and wearing a fawn-skin, holding fork and shovel, on a pyxis in Reading. Among the mortals we find a slinger, a girl playing kottabos and a centaur watching a tortoise inside cups in Athens, London and Leningrad. An unpublished cup in Athens, formerly in the Empedocles collection, shows a warrior advancing with his spear at the ready; another unpublished in Oxford, on loan to the Ashmolean Museum, has Oedipus and the sphinx. The only floral subject that I know is seen in the Reading cup with a rosebud between sprays. There is a replica of this (but with the bud black instead of red) in a cup with a tall foot in Corinth Museum.