The appearance of a further instalment of these ‘Studies’ has been regrettably delayed, owing to several reasons. The present article is planned on different lines from its predecessor (JHS li (1931), 139–63), for, instead of offering a comprehensive study of one reconstructed stele, it attempts to review the post-Euclidean Hekatompedon-lists down to 390/89 as a group, and to establish their chronological sequence, in the light of fresh discoveries. It is proposed, also, to deal in subsequent articles with the corresponding lists for the Parthenon and the Opisthodomos in this period, and to continue the study of the later lists both of the Treasures of Athena and of those of the ‘Other Gods’ into the second half of the fourth century.

The opportunity of working continuously in the Epigraphical Museum for several weeks during the summer of 1931 enabled me to recognise that in many instances two or more fragments of these Traditiones which had been published separately could be assigned to one stele, and also to identify a few still unpublished pieces belonging to this series.

It has long been known from the extant ancient Greek musical documents that some composers correlated melodic contour with word accents. Up to now, the evidence of this compositional technique has been judged impressionistically. In this article a statistical method of interpretation through computer simulation is set forth and applied to the musical texts, focusing on the convention of correlating a word's accent with the highest pitch level in the melody for that word: the Pitch Height Rule. The results provide a sounder basis for judging evidence for the operation of this convention in specific pieces and a sharper delineation of its use in the history of ancient Greek music. The ‘rule’ was used by at least some composers from the late second century BC through the second century AD, but there is no certainty that it was used before or after this period. In some cases where previous scholars have discovered the rule's operation, statistical analysis casts doubt. Of special interest is the showing that one piece long judged as offering no evidence of the use of the rule probably displays an inversion or parody of the rule for rhetorical-musical effect.

In discussions of Aeschylus' theology one of the passages most often quoted is the so-called ‘hymn to Zeus’ in the first chorus of the Agamemnon (Ag. 160–83). Fraenkel in his commentary goes so far as to call it ‘the corner-stone not only of this play but of the whole trilogy’. The passage concludes with two lines which in all modern editions are read as a statement, though our oldest manuscript, the Medicean, writes them as a question. Textually the difference is merely one of accent, but the difference of accent carries with it a reversal of meaning. As a statement the lines mean that the gods are something to be grateful for, that there is some χάρις or kindness associated with them. Taken as a question they deny this. Clearly then it is of great importance for the interpretation of Aeschylus to decide which is the correct reading.

The lines in question, written without accents, are

Our oldest manuscript, M, as I have said, writes ποῦ with an accent. So does our next oldest, the manuscript 468 of the Biblioteca di San Marco, generally known as V. If this reading stems uncorrupted from the time when accents were first applied to the text of Aeschylus and if at that time the oral tradition of the poet's words was not yet dead, then it will not be destitute of authority. But the thread is far too tenuous to bear any weight of proof.

Equally there can be no argument from authority on the side of reading the lines as a statement. For though Triclinius and the closely associated manuscript F write που without an accent as an enclitic, this is as likely as not to be due to simple conjecture.

The three-sided Boston relief (Plate XIIIa), which is to be dated in the second quarter of the fifth century, has been the object of a long controversy with regard both to its subject matter and to its authenticity, which has been doubted by some scholars. The authenticity of the monument will be taken for granted here, since the work of Jucker, and especially the recent exhaustive stylistic and scientific study by Ashmole and Young leave no possible room for doubting it. Another aspect of the relief which I will take for granted in this paper is the artistic milieu which created it, since it has been convincingly shown that it is of South Italian, and more specifically Epizephyrian Locrian, origin. The object of the present paper is to discuss the iconography of the monument, especially with reference to the cult and religious environment of the city in which it was produced.

The interpretation of the central scene and the two side-panels of the Boston relief is still a matter of controversy, although many hypotheses have been put forward since the monument first appeared in the antiquities market. Discussions of the iconography of this relief tend more often than not to connect the problem, in some way or other, with the subject matter of the Ludovisi throne (Plate XIIIb), another three-sided relief belonging to the artistic environment of Locri Epizephyrii, but of a much higher artistic quality. The interpretation of the scenes on the Ludovisi throne has not provoked the same amount of controversy, and it would, I think, be a fair statement that the interpretation of the central representation as the birth of Aphrodite is now generally accepted—more accurately, it is the new-born Aphrodite being assisted out of the sea, and to the shore, by the Moirai or the Horai. On each of the side-panels a female figure is shown, a naked pipe-player on one, a heavily draped young matron burning incense in a thymiaterion on the other. They have been interpreted as hetaira and young bride or wife, two contrasting figures associated with Aphrodite's Locrian cult.

The join is between the right end of 1008 and the left end of 1010, (plate Ia, b). The reason why it has escaped notice is that the Knights of St John, when building the reliefs into the walls of the Castle of St Peter at Bodrum, hacked the slab in two clumsily, destroying several inches of the front surface in doing so. Once the two pieces had been set up in the British Museum, with slab 1009 interposed, and catalogued in that order, the join was unlikely to be recognised, for the subjects do not at first sight appear to connect, and the join even now is not obvious from the front, because of the rough way in which the block was split in two; but when the fractured surfaces further back in the slab are brought together, the correspondence of the contours leaves no doubt whatever that they fit, and thus produce a single, very large, slab, nearly nine and a half feet long, as against the seven feet odd of the next longest survivor, 1022.

The new join is of interest partly because it prolongs a continuous run of the frieze (1007 plus 1008) by several feet (plate Ib), giving a sequence of eleven figures, if we include the two legs at the right edge; but chiefly because the composition thus produced regains a significance which the separation of its parts had obscured. The figure on 1008 who wields a club and wears a lion's skin must surely be Herakles: the addition of 1010 on the right shows that this important figure is balanced by another Greek who ought to be equally important, and who is engaged with an Amazon on horseback. Would this not be Theseus? On Athenian vases of a century earlier, a composition not unlike this, of the duel between Theseus and an Amazon, each using a spear, was already established (plate Ic): its prototype may have been in some famous wall-painting, and the designer of the Mausoleum frieze may have drawn his inspiration directly from that or from some version of it in another work of art now also lost to us. The identity of the two Amazons is a separate problem.

In the second and third centuries AD, many of the cities in the eastern Mediterranean could boast about having their own athletic games. In the fourth century, however, these games quickly declined. In recent years, the traditional explanations for the end of athletic games, most prominently the supposed ban by Theodosius, have been proven unfounded. This paper proposes an alternative explanation: institutional and financial changes hindered the successful organization of athletic contests by the cities in the fourth and fifth centuries. In order to show the effect of these changes, this paper first offers a detailed analysis of how athletic contests were founded and funded in the early imperial period. It then examines how and to what extent these procedures and funds were affected by changes in late antiquity. The decline was not caused by a general financial crisis - in fact the estates (partially) funding the games remained a stable form of financing. Instead the shift of power to a centralized bureaucracy limited the cities in their administration of the games: they could no longer independently meet deficits in the agonistic budget from the city treasury and had to rely increasingly on elite sponsors, whose ambitions focused mainly on the provincial capitals and who gradually lost their interest in athletics.

The scientific treatment of the Geography of Ptolemy (Γεωγραφικὴὑφήγησις) had made considerable progress during the last century, so that it seemed as if this work had been brought at least to a provisory issue. An edition arranged according to the demands of science and, as was to be desired, an edition that could be called final had not yet been produced, but there was reason to believe that the edition undertaken by the well-known editor C. Mueller in the great Bibliotheca Scriptorum Graecorum, published by Firmin Didot in Paris, would come up to these expectations. However, owing to his death in 1893 it has remained unfinished. After Part I. had appeared in 1883, C. Th. Fischer, to whom the continuation of the work was entrusted, was able in 1901 to publish Part II., which had been found almost ready for the press among the literary remains of the deceased.

Amongst the legendary heroes who appear in leading roles in the surviving plays of Sophocles, it is noteworthy that Oedipus, Ajax and Heracles all received some form of divine worship in Attica, not to mention localities more readily associated with each of them. Sophocles is not unaware of this aspect of each of these figures, but where the future prospect of their cult is alluded to in the plays, such allusions are not always prominent or explicit; though the future cult of the Oedipus of the Oedipus Coloneus is crucial for the play, it is only directly mentioned in a few passages, while the Ajax of Ajax is seen as a future receiver of cult only in a single unusual scene of supplicating his dead body, and it is unclear in the Trachiniae whether the audience is intended to supply a future cult on Oeta and apotheosis for Heracles. I should here like to argue that in Philoctetes Sophocles again consciously employs a hero destined to receive worship after his death, and that this is subtly suggested at the end of the play.

The problem of the date of the surrender of the Messenians blockaded by the Spartans at Ithome has been subjected to considerable discussion in recent years. From this discussion two main attitudes have emerged:

(a) The majority hold the view that the surrender took place round about 460, and that the phrase ‘δεκάτῳ ἔτει’ in Thucydides' account (i 103.1) requires emendation, though they are not agreed on what the text should be.

(b) A minority view, put forward by Scharf, and independently and in greater detail by Hammond, agrees in putting the surrender early, round about 460, but keeps ‘δεκάτῳ’ in the text of Thucydides, and supposes that his calculation was made not from the revolt that followed hard upon the famous earthquake (which is the usual view), but from an alleged earlier rising dated to 469/8. In a rather different form this view is also held by Sealey.

In Book viii ch. 6 Polybius describes the structure and use of a siege-engine, the σαμβύκη, so called from its resemblance to a musical instrument of that name. The purpose of this article is to review the evidence for the shape of the musical instrument, and for the structure of the siege-engine, and for possible points of similarity between them.

References to the musical instrument are not uncommon in literature from the fifth century B.C. onwards; there are a number of points on which they agree. The σαμβύκη was a stringed instrument, and was generally thought to be of non-Greek, near-Eastern origin. The alternative view, that it was invented by Ibycus the lyric poet of Rhegium, was probably based on a fanciful etymology, and can reasonably be disregarded. The derivation of ἰαμβύκη (probably just a variant spelling) from ἰαμβός is also fanciful.

As regards chronology, the σαμβύκη seems to have been introduced into Greece in the late fifth century B.C., and was at that time regarded as something of a novelty by the comic poets. Some later scholars, however, saw fit to correct this impression, and point out that it was in fact a very ancient instrument. Its main use was apparently to provide a musical accompaniment for debauchery: in fact, so disreputable was the instrument that the word for a female player of it, σαμβυκιστρία, was virtually synonymous with πορνή.

The site map of Cyzicus reproduced in this number (Pl. VI) represents the results of Mr. A. E. Henderson's survey in the summers of 1902—3. The first season's work, of which alone I can speak at first hand, accounted for the coast line, prominent remains (including the city walls), and main roads: the limits of the marsh land on the isthmus were also ascertained with considerable exactness with a view to the recent discussion of the original nature of the Cyzicene Peninsula: these limits, however, are subject to a certain amount of variation with the season, and cultivation is yearly encroaching upon them. The maze of walls within the enceinte—some of them modern vineyard boundaries, others doubtless resting on ancient foundations—was added by Mr. Henderson and Mr. Peet in 1903: contouring was found impossible in the time at their disposal, and indeed the general levels of the site are fairly adequately shown on the excellent Admiralty Chart of Artaki Bay (of which a section is here reproduced in miniature in Fig. 1) and on the sketch maps of Perrot and De Rustafjaell.

Describing the approach toward Sardis in the spring of 1699, the British antiquary Edmund Chishull wrote: ‘We continue our journey through a spatious and fertile plain, curiously beset on each side the road with [a] variety of round hillocks, which from their number, figure, and situation, in so level a campaign, appear plainly to be artificial’. The tumuli or burial mounds so described by Chishull still dominate the Hermus river plain opposite Sardis. Their modern name is Bin Tepe, or ‘Thousand Mounds’.

Cle omedes, De motu circulari corporum caelestium. II. 75 (p. 136 Ziegler). Ἐλέγχεται δὲ καὶ διὰ τῶν ὑδρολογίων τὸ εὔηθες τοῦ λόγου [viz. ὅτι ποδιαῖός ἐστιν ὁ ἥλιος]. Δείκνυται γὰρ δι' αὐτῶν, ὅτι, ἂν ᾖ ποδιαῖος ὁ ἥλιος, δεήσει τὸν μέγιστον τοῦ οὐρανοῦ κύκλον ἑπτακοσίων πεντήκοντα ποδῶν εἶναι. Διὰ γὰρ τῶν ὑδρολογίων καταμετρούμενος εὑρίσκεται μέρος ἑπτακοσιοστὸν καὶ πεντηκοστὸν τοὑ οἰκείου κύκλου. Ἐὰν γὰρ, ἐν ᾧ αὐτὸς ἀνέρχεται πᾶς ἐκ τοῦ ὁρίƷοντος ὁ ἥλιος, κύαθος, φέρε εἰπεῖν, ῥεύσῃ, τὸ ὅδωρ ἀφεθὲν ὅλῃ τῇ ἡμέρᾳ καὶ νυκτὶ ῥεῖν εὑρίσκεται κυάθους ἔχον ἑπτακοσίους καὶ πεντήκοντα. Λέγεται δὲ ἡ τοιαύτη ἔφοδος ὑπὸ πρώτων τῶν Αἰγυπτίων ἐπινοηθῆναι.

This essay presents a new interpretation of what happened in the opening of the Pylos campaign; one that rests mainly on an attempt to discover, in Thucydides' text, Demosthenes' original plan for the campaign and the Spartan reaction to it. Thucydides says little about plans for Pylos; indeed, he seems to describe this part of the campaign as unusually haphazard and fortuitous. Unfortunately, his account of it (iv 2–16), which is our only real source, is so compressed, obscure, and elliptical that no interpretation of these events can now approach proof or certainty. All scenarios and explanations are therefore speculative. They can be evaluated only by their degree of consistency with all the data, unsatisfactory as these may be, and by their relative plausibility—a subjective and fallible measure at best since human affairs so often turn out less plausibly than we expect. Despite these limitations and caveats, rigorous analysis is possible, and can reveal enlightening and previously unnoticed connections both between events within the Pylos campaign, and between these difficult chapters and other, clearer parts of Thucydides' text.

Before coming to the discussion of the three unpublished vase-paintings which illustrate this article, and of the questions which they suggest (Plates I., II., III.), it will be proper to give some account of the Centaurs in general, as figured on the painted vases of the Greeks. The passages or episodes of the Centaur myth habitually illustrated in this form of art are five in number, viz.:—

1. The battle of the Centaurs and Lapithae at the wedding feast of Peirithoos and Hippodameia, or Deidameia, on Mount Pelion; when the Centaurs, being present as guests, maddened themselves with wine, and one of them seized the bride; whereupon a general conflict ensued, ending in the rout of the monsters and their expulsion from Thessaly.

This battle is said by Aelian to have been made the subject of a separate poem by an early epic writer, Melisandros of Miletus; but neither of Melisandros nor his work have we any other record. In our extant writings, allusion is made to the battle twice in the Iliad: once where Nestor extols the Lapith warriors, whom he had known in his youth, as having been the mightiest of earthly heroes, and having quelled the mightiest foes, to wit the Centaurs; and again in the catalogue of ships, where the Thessalian leader Polypoites is commemorated as the son begotten of Hippodameia by the Lapith king Peirithoos on the day when he chastised the monsters and drove them from Pelion.

On a fall night before Issus in 333 BC, we are told by Curtius Rufus, Alexander ascended by torchlight to the summit of a lofty mountain, and in accordance with ancestral custom performed sacrifices to the chief guardian deities of the place: an intriguing historical remnant, the more challenging because Curtius provides no other details, and no other Alexander account gives reliable information about this particular display of piety. Consequently the gods in question remain unknown. The purpose of this study is to establish the likelihood that Alexander did in fact perform these rites, and to determine the probable identities of the deities in the light of what is known about religious cults at Issus before he arrived.

Among the sherds found at Gela in what must have been the sanctuary of the hero Antiphamos—the Rhodian co-founder, with the Cretan Entimos, of the city itself—was a fragment of a large Attic skyphos representing the Assassination of Hipparchos. It was described, though not figured, by Orsi in Notizie degli Scavi 1900 p. 276. He compared it with the well-known picture of the same subject on the stamnos by the Copenhagen Painter in Würzburg (515: AZ. 1883 pl. 12 and pp. 215–8, Boehlau; Klio 20 pll. 1–4; Langlotz pll. 182–3 and 210: ARV. p. 193 no. 5). Petersen noticed the fragment in R.M. 16 (1901) p. 103, and there are one or two bare mentions elsewhere. In 1923 it was still in private hands at Gela, where I saw it, but not long afterwards it passed to the Villa Giulia, and was figured, with a hasty description, by Cultrera in Bollettino d'Arte 7 (1927) p. 319, above, 2. This publication seems to have escaped general notice, at least it is not quoted in recent studies of the Tyrant-slayers. The fragment has such exceptional interest of subject that I may be excused for returning to it.