The story of Orpheus and Eurydice has in recent years received attention from Heurgon, Norden, Guthrie, Linforth, and Ziegler, who have in different ways supplemented the admirable article by Gruppe in Roscher's Lexikon published fifty years ago. Unless new texts or new monuments are found, it does not seem likely that fresh evidence will be forthcoming to solve old problems, and our task is rather to make a constructive use of what evidence we have. This paper attempts to consider only one part of the whole question, the versions of the story which lie behind the familiar form. Much said here is not new, but perhaps the attempt to sort out some points is worth making, especially as the experts disagree on several important matters.

Anyone who reads the opening chapters of Thucydides’ history consecutively will soon find it difficult to follow the thread of the argument. If he turns to a summary of the subjects chapter by chapter, he will not be greatly enlightened. In this paper the question is asked: why did Thucydides arrange his subjects as he did? In Part I the conclusion is reached that in the arrangement of his subject-matter he was following a clear-cut system. In Part II the implications of this conclusion are considered.

An investigation into the prosopography of Libanius shows a confusion of identity in the following cases. The textual references are to Foerster's Teubner edition, of which the index compiled by Richtsteig (vol. xii) is to be handled with caution.

There is in the British Museum a twelfth-century manuscript, Egerton 3055, of Suetonius' De Vita Caesarum which has not previously been noticed by editors or in the reports of manuscripts of Suetonius made by Professor C. L. Smith and Professor A. A. Howard, though summarily classified by E. G. Millar.

I Assume that in Latin there was a stress accent which, in the time of Lucretius, was governed by the well-known ‘law of the penultimate’; also that in Latin poetry, although the metre is determined by the quantity of the syllable, nevertheless the stress accent must not be ignored. In fact, the inter-relation of the ictus of the quantitative metre with the stress accent is a most important factor in determining the rhythm of the verse. It is well known that in the Latin hexameter special importance was attached to the coincidence of ictus and accent in the fifth and sixth feet, so as to secure a characteristic and definite ending to the line. But the coincidence or clash of ictus and accent has its effect in every foot, although to a less extent. Particularly in the fourth foot is it important. There is an obvious difference between a line in which stress accent reinforces the ictus in the fourth, fifth, and sixth feet, and a line in which conflict occurs between ictus and accent in the fourth foot. W. F. J . Knight refers to these lines as homodyned and heterodyned. Each foot of the hexameter could be examined for the effect of homodyne and heterodyne; I am limiting this discussion to the fourth foot. In classifying fourth feet as homodyned or heterodyned, I have described as homodyned a foot in which a stress accent coincides with the ictus; all other feet are heterodyned. It must be realized, however, that there are possible varieties of homodyned fourth feet. They may be spondaic or dactylic; they may be caused by the overlap of the last two (or three) syllables of a polysyllabic word, or they may consist of a single word or two words. There is a great difference of metrical effect. Heterodyned lines may again be differentiated between those which contain a stress accent clashing with the ictus, and those in which there is no stress accent at all.

The present study of the manuscripts of the Knights arose out of the preparation of a text of the scholia for a forthcoming edition. The completion of a collation of all the manuscripts for the scholia seemed a suitable occasion for extending the inquiry and re-examining our manuscript tradition in both text and scholia, especially as the scholia in a manuscript, provided they come from the same source as the text, can often reveal facts that might escape an investigator who confined himself to the text. The scholia are far greater in bulk than the text, and scribes are often less careful and so more self-revealing in scholia; so that both from the quantitative and qualitative points of view scholia are important for the student of the tradition.

For horribilesque we need something better than Haupt's horribile aequor (so Postgate, Merrill, Kroll, Schuster); and Mr. E. L. B. Meurig Davies (Class. Quart, xliv, 1950, p. 31) comes near the truth, I think, with his proposal horribilem niue. A noun in the ablative indicating cold to define horribilem is just what we require. That noun does not seem to me, however, likely to be niue. Read rather horribilem gelu; cf. Luc. 2. 570 ‘(Caesar) Rheni gelidis … fugit ab undis’, Claud. Rapt. 3. 321 ‘non Rheni glacies, non me Rhipaea tenebunt’, Paneg. 7(6). 6 ‘duratus gelu Rhenus’.

Mr. G. S. Kirk's interesting and illuminating treatment of this curious fragment in CQ,. xliv (1950), pp. 149–57, prompts me to offer some further conjectures (i) on the reading and interpretation of lines 15–23, (ii) on the character of the fragment as a whole. I am indebted for valuable help both to Mr. Kirk himself and to Mr. C. H. Roberts.