So far as authority goes, Kaerst founded his theory of Alexander's worldkingdom on two passages in Diodorus and on nothing else. The first, 17, 93, 4, alludes to Ammon having conceded to Alexander the power over the whole world, τὴν ἁπάσης τῆς γῆς γῆς ἐξουσίαν the reference is to 17, 51, 2, where Alexander says to the priest of Ammon, εἶπέ μοι εἴ μοι δίδως τὴνἁπάσης <τῆς> γῆς ἄρχην and the priest replies that the god grants this. The second passage is 18, 4, 4, the story of Alexander's supposed plan to conquer Carthage, etc., and go to the Pillars, from his alleged ὑπομνήματα Every one will agree with Kaerst when he says that the political information in the Arrian tradition is imperfect, and that it is very desirable to supplement it; but the real question, which has to be faced, is, are we in a position to supplement it? It is no good using unsound material as a supplement; it is better to say we do not know, if it comes to that. My object here is to examine the Diodorus passages and see what kind of material they offer.

Some of our extant authorities, as Justin and Appian, state or assume that Alexander had two sons, Roxane's and Barsine's. Others, as Diodorus in the events prior to 309, and Curtius in parts, state or assume that he had only one, Roxane's. Now it makes a considerable difference in our view of the events of 309 whether the lad called Heracles, who appeared in that year as a reputed son of Alexander and Barsine, were really Alexander's son or an ordinary pretender. No modern historian has even noticed that there is a conflict of authority; for though Beloch saw that Heracles' age was wrong he did not follow it up, but altered the age. Before coming to the events of 309, the source of the evidence for Alexander having one son only must be considered.

In past numbers of the Annual of the British School at Athens and elsewhere I have tried to deal with some of the questions connected with Byzantine Music, and, having brought to a close my studies of the Round or Later Mediaeval System, I am unwilling to leave the subject without giving my views on the abstruse and difficult problem of the older notation.

The later forms of the Linear or Neume System have a visible likeness to the earlier forms of the Round System already familiar, and hence all investigators seem to have started with the idea that the general principles of decipherment could be transferred from the later to the earlier stage, or, in other words, that the task simply consisted in the interpretation of certain interval-signs possessing fixed value. But of the two scholars who have published their researches in this field, Gastoué and Riemann, neither has been able to carry this principle through, and their proposed solutions fail to give us such a chain of interval-signs as we are tempted to expect.

In my last Bibliography (J.H.S. xxxix. 209 ff.) I attempted to cover the three and a half years from July 1915 to December 1918 inclusive, though I was only too well aware that, under the conditions of the period of war and armistice, I could not claim completeness for my record. In the present article I deal primarily with the years 1919 and 1920, but I have inserted references to a number of books and articles which actually appeared earlier though they did not become accessible to me until the years under review. Excavation has not yet been renewed on anything like the pre-war scale and the number of Greek inscriptions published for the first time is correspondingly small, but gratifying progress has been made in many directions in the restoration of mutilated texts and the fuller interpretation and utilisation of documents already known. The reader who glances even cursorily through the following pages will, I hope, be struck, despite the compression necessitated by considerations of space, by the vitality and interest of the study to which they relate, and by the many-sided contribution it has made to the understanding of Hellenic language, literature, religion and history.

The question when, and by whom, our constellations were invented, will probably never lose its fascination, because it is never likely to find its solution. For those who have allowed themselves to be brought under its spell the name of Cleostratus has a special interest. If we could by any means learn more about the man who is said to have been in some sort the deviser of our zodiac, we might obtain a light upon the history of the celestial globe which at present seems likely to be for ever withheld, unless some Egyptian papyrus should reveal some part of the lost History of Astronomy by Eudemus.

By his careful collection—in the December number of this Journal, 1919—of all the notices that we have of Cleostratus, Dr. W. K. Fotheringham therefore deserves a gratitude which I am the more anxious to express because I cannot at all agree with the theory of Babylonian influence which he deduces from them, nor with the interpretation of Greek and Latin passages which he puts forward in support of that theory. The latter point I could willingly leave to the criticism of scholars abler than myself, whom I cannot think likely to be convinced by Dr. Fotheringham that the passages bear the sense which he has endeavoured to extract from them.

The bronze statuette reproduced for the first time on Pl I. and Fig. 1 has for many years past formed part of the national collections. The earliest date to which it has so far been traced is 1885, when it was included in the category of ‘unclassified or suspect bronzes.’ Beyond 1885 it enjoys at present the happiness of having no history; but as in that year it bore no mark of registration, the inference may be drawn that it entered the Museum with the ‘old collections,’ perhaps a hundred or a hundred and fifty years ago. It remained in retirement until the early years of the present century, when attention was called to its affinities with the newly discovered art of prehistoric Crete; and the publication, in 1912, of the Tylissos praying figure (Fig. 2) supplied a parallel sufficiently close to establish beyond doubt that the British Museum bronze was a work of the same school and period.

It has occurred to me more than once that there was yet some work to be done on this topic, even after the meritorious and very accurate labours of Steele, the notes and indices of a series of editors, notably Ernesti, Orelli, and Tyrrell and Purser, and the dissertations of Bolzenthal, Font, and Laurand. Of these, the editors are concerned chiefly with establishing a correct text, and explaining the meanings of the words, which last task has for the most part been satisfactorily performed (see Tyrrell and Purser, passim, also Boot's excellent edition of the Letters to Atticus). Laurand mentions the matter only incidentally, and gives a list, not very reliable, of the words used in the rhetorical works; Font's chief interest is not lexicographical, but rather an attempt to answer the question why Cicero should ever use a Greek word at all when a Latin one was available. Bolzenthal I have not been able to consult, but gather from Font's synopsis of his work, pp. 3, 28 sqq., that it is largely superseded by Steele. Steele sets out to study the whole vocabulary of the letters, including quotations, but omitting the Greek words in the other works; and his chief interest, apart from tracing the quotations to their sources, is in a grammatical analysis of the words used by Cicero and his correspondents, with a list of those words which occur only or for the first time in the letters.

In Vols. XVIII. (1898) and XXXI. (1911) of the Journal I gave some account of black-figured vases acquired by the British Museum subsequently to the appearance of Vol. II. of the Catalogue of Vases in 1893. On page 1 of the latter volume a promise was made that another paper should follow, describing red-figured vases similarly acquired; but its appearance has been delayed by the war and other circumstances, with the result that the number of vases now included amounts to nearly fifty. Seventeen other vases acquired during the period 1895–1920 are omitted here as having been already published elsewhere, but a list is appended on page 150. The total number of red-figured vases added to the collection since 1894 is thus over sixty.