The construction of ἄν with the future has been hotly denied as impossible, so far as Attic Greek and indeed post-Homeric Greek generally are concerned. The opponents of the construction have had among their number such scholars as Dawes and Cobet; and of late, it seems, editors of texts generally. The view of Cobet is given on p. 469 of his Miscellanea Critica, with reference to Demosth. 9. 70 (128) πάλαι τις ⋯δέως ἂν ἴσως ⋯ρωτήσων κάθηται. Cobet, who has been followed by later editors, altered ⋯ρωτήσων (the universal reading of the codd.) to ⋯ρωτήσας, commenting ‘ubi semel Constiterit ⋯δέως ἂν ⋯ρωτήσω, ⋯δέως ἂν πεύσομαɩ aut simile quid pro ⋯ρωτήσαιμι vel πυθοίμην recte dici, turn demum librorum lectioni acquiescemus. quod equidem nunquam futurum esse satis scio.’ This view, which must of course be understood as excepting Homer from its scope, is nothing more than a blank denial of the possibility of the construction. We see more of an argument in Kühner-Gerth (Griech. Gramm. ii. 1. 209), where it is remarked that the construction, frequent in Homer, was later given up because ἃν with the optative was sufficient to express a future possibility; and that possibly emenders have done right to alter passages in Attic which contain it. This is not expressed with any great certainty. Nevertheless it has become to such an extent the prevailing view among editors that in modern texts it is extremely rare to find the construction allowed to remain.

The transition from the Athenian Peripatos of Aristotle to the Alexandrian Museion of Callimachus has often attracted notice. So closely akin was the organization of scholarship in the two centres of learning, so definite was the personal connexion between the two, that it seemed possible to trace an uninterrupted line of succession from the older to the younger school. That Callimachus the scholar worked in the Aristotelian tradition appeared obvious: ‘he might be called a Peripatetic in the same sense as his pupils, but with more justification’ (Wilamowitz, Hellenist. Dicht. i. 214). Regarding Callimachus the poet and literary critic the picture seemed less clear. The non-classical character of his poetry had often been emphasized, and cannot be overlooked. But quite recently an attempt has been made to turn Callimachus into an Aristotelian even by virtue of his literary criticism. The alleged harmony of this development rouses suspicion and calls for inquiry. In the following pages I have tried to collect and discuss what material I could find for the relation of Callimachus to Aristotle and his school. I hope to show that such influences as there were are of a different and more subtle nature, and that this poet, if any, was his own Longinus.

Of all the dialogues that may be said to be in Plato's normal style, the Republic seems to be the richest in imagery. The Phaedrus may contain more of such figurative language, but its whole atmosphere and style are so artificial as to place it outside comparison. The Republic stands, in this respect as in philosophic content, between the relative plainness of the earlier works and the didactic heaviness of the Laws, which is relieved by proverbial, rather than by imaginative, illustration.

It is, I think, generally believed that the last oracle delivered at Delphi was that given to Oreibasios announcing the inability of Apollo to prophesy there again. This oracle begins with the line: εἴπατε τῷ βασιλ⋯ϊ· χαμα⋯ π⋯σε δα⋯δαλος αὐλ⋯ and has been translated by Swinburne as The Last Oracle. Of it Myers wrote: ‘(It is) the last fragment of Greek poetry which has moved the hearts of men, the last Greek hexameters which retain the ancient cadence, the majestic melancholy flow.’ But there is evidence which suggests that this was, in fact, not the last prophetic utterance of the Pythian god, and no ancient authority implies that it was.

βρυχεδαν⋯ς (of a locust's greed, cf. βρο⋯κος, βρο⋯χος): πολυφ⋯γος, οἰ δ⋯ μακρ⋯ς. For μακρ⋯ς read μ⋯ργος.

ζ⋯γγος· ⋯ τ⋯ν μελισσ⋯ν ἦχος, κα⋯ τ⋯ν ⋯μο⋯ων. L. and S. translate literally, ‘humming of bees, etc.’; but to buzz or hum is not a common property of insects, it is peculiar to a few. For τ⋯ν ⋯μο⋯ων I suggest τ⋯ν μυι⋯ν. ζ⋯γγος refers especially to the buzz, or ‘ping’, of a mosquito (μυῖα), LL. zanzara; cf. Cassiodorus (cit. Forcellini, ed. De Wit) ‘Ciniphes genus est culicum, fixis aculeis permolestum, quas vulgus consuevit vocare zinzalas’; and in the Appendix, De Wit adds: ‘zinzala; parva musca, id est culex; unde zinzalarium, conopeum ad eas arcendas, Gloss. Ugut. (Hinc vulgo apud nostrates la zenzaliera)’—i.e. a mosquito-net. Culex, by the way, is an interesting and a homely word; we lack mosquitoes in Scotland, but we know the angry bite of a ‘horse-fly’ and call it a ‘cleg’.

κάτοπτρον, which is in all the manuscripts, was emended by Canter to κάτοπτον, and this emendation, or Headlam's κατόπτην, has been received by subsequent editors. Those who read κάτοπτον have been in the habit of taking the word to mean here ‘looking down upon’, and in support of this interpretation they sometimes adduce a scholium in M, κατόψιον. This does seem to prove that the scholar, whose note is copied in our scholium, found κάτοπτον in his text. Presumably he took Σαρωνικο⋯ πορθμο⋯ κάτοπτον to signify ‘descried from the Saronic gulf’, having very possibly in his mind the phrase in Euripides, Hippolytus 30, where the temple of Cypris is said to be κατόψιον γ⋯ς τ⋯σδε, visible from Troezen. At least we ought not without proof to lay to his account what appears to be the solecism of regarding κάτοπτον as if it meant ‘looking down upon’. Headlam's κατόπτην has at least an active force, but I cannot find that it means ‘one who looks down upon’, if that is the sense desired; it is a ‘spy’ or an ‘inspector’. And, at the end of all discussion, when one asks what in fact is this peak or headland that ‘looks down upon the Saronic crossing’, then commentators take refuge in silence or tend to contradict one another. An emendation that nobody can be sure he understands can hardly be regarded as satisfactory. It may be feared that what has happened is this. In order to get rid of κάτοπτρον, which appeared to them devoid of meaning, editors have introduced into the text a word which does not mean what they want it to mean, and which, if it had that meaning, would still present us with a picture which no man can recognize.

In my first study of the borrowings from Athenian sacred treasure to finance the Archidamian War I assumed, in common with others, certain irregularities in the stoichedon order of IG. i. 324. The text has subsequently been amplified and improved by Tod, notably with the addition of one amount of interest due to Athena (line 101) and of the total amounts of principal credited to the Other Gods (line 119) and to all the gods (line 122). This further expansion, however, has introduced additional irregularities, the restoration [τάδε τοῖς ἂλλοις θεοῖς ⋯ϕέλοσι τόκο ⋯]ν ἒνδεκα ἒτεα[ιν:…] being too long by one letter in line 120 and the restoration [τάδε ἂπασι τοῖς θεοῖς ⋯ϕέλοσι τόκο ⋯ν] ἒνδεκα ἒτεσ[ιν:…] being likewise too long in line 123.