¹ It is even suggested, in Schwyzer, , Griech. Gramm, i, p. 38Google Scholar, that this phrase may be an example of the taboo-expression, designed to avoid saying the proper name! This view would surely have astonished Homer, as much as it does me.

² K.Z. xl, p. 359 note.Google Scholar

³ K.Z. Ixiv, p. 63, note 2.Google Scholar

¹ Chantraine, P., ‘Sur le vocab. maritime des Grecs’, Étrennes E. Benveniste, pp. 1–27.Google Scholar

² Meillet, So, Aperçu d'une hist. de la langue grecque, p. 12.Google Scholar On the other hand, Buck, , Classical Studies presented to Edward Capps, pp. 42–5Google Scholar, treats of θάλαττα and tries to show that the formation may be genuinely Greek. So it may; but the matter is very doubtful, and I prefer the general view of it as Aegean. In the other part of his paper Buck argues that even if the word is non-Greek, it is not thereby proved that the Greeks had not had acquaintance at an earlier stage with the sea. His summing up of the history of the word Lat. mare in Celtic, Germanic, and Balto-Slavonic (p. 44) is a little misleading. He speaks of ‘loss or subordination of the old word among the Norsemen, the Anglo-Saxons, the Irish, and the Letts’. But the word has been lost only by Lettic, and survives in allied Lithuanian; the subordination is true of the other peoples, but is a very different thing from loss, and deserves greater prominence. In Greek the word simply does not exist. Of course one would not lay much stress on one word, such as θάλαττα, in the effort to show the former continental habitat of the Greeks: but I think it is clear from what I say above that the example is by no means isolated.

³ Due to Cowley, in Allen, , J.H.S. xxx, p. 317Google Scholar; v. T. Allen, W., The Homeric Catalogue of Ships, pp. 166–7.Google Scholar

⁴ Allen, , Hom. Cat. p. 162.Google Scholar

¹ Revue Belge de Philologie, et d'Histoire, iii (1924), pp. 315–17.Google Scholar

² Ionic shows -ειν- for original -εF-. More fully, -ειν- appears in East Ionic, Central Ionic (Paros, Thasos, etc.), and also East Doric (Crete, Rhodes, Thera, etc.), v. Schwyzer, , Griech. Gramm. i, p. 228Google Scholar; but there is no gain in regarding this particular form εὔξεινος as due to East Doric.

³ Trist. iv. 4. 55–6:Google Scholar

frigida me cohibent Euxini litora Ponti:

dictus ab antiquis Axenus ille fuit.

⁴ The forms found in the MSS. of Eur. I.T. are somewhat confused. The relevant passages are I.T. 94, 125, 218, 253, 341, 395, 438, 1388, of which 94 does not refer to the sea. LP give εὐξειν- at 125, 395, 1388: the Oxford text (G. Murray) follows Markland in changing 125 to ἀξειν-, the other two to ἀξεν-. This move preserves the metre at 125 and 395: at 1388 it is indifferent whether we read -ειν- or -εν-. LP are of course corrupt in this matter, as is shown by 395 and also by 94 (their ἄξεινον has to be changed to -ενον for the metre: this was proposed by Musurus). The net result is that the Oxford text has in all the passages either ἀξεν- or ἀξειν-. This is one way out of the difficulty, but I think there is a better. My solution is: read εὐξειν- in the passages (125, 438) which require -ειν-, and ἀξεν- in the others. This would (I) produce the common distinction between ἀξεν- and εὐξειν-, and (2) explain better the sad confusion of LP. But this scheme should be modified to have ἀξείνον at 218, where the negative epithet is more suited to the context.

¹ I cannot agree here. ‘Pacific Ocean’ may be mentioned.

¹ Leaf, , Troy, p. 299.Google Scholar

² Meillet, , Aperçu, pp. 7–8.Google Scholar

³ Ullman, B. L., ‘Early Greek Alphabets’, Classical Studies presented to Ed. Capps, pp. 333–42.Google Scholar