^{page 10 note 1} The Antiatticista (Bekk. 108. 22) attributes to Sappho theword μ⋯ῤῥα. which occurs in v. 30 of this poem.

^{page 10 note 2} Neue Jahrb. XXXIII, 1914, p. 229 sqGoogle Scholar. Cf.Lobel, , ‘Αλκαίου μήλη p. xviiGoogle Scholar, ‘β2… is almost certainly supposititious’; Bowra, , New Chapters in Greek Literature III, p. 13Google Scholar, ‘The poem, then, is not Sappho's’.

^{page 10 note 3} Atene e Roma, XVII, 1914, pp. 224 sqq, esp. p. 247 n. 2Google Scholar, ‘I nuovi frammenti di Sappho’.

^{page 10 note 4} Among minor objections to Sappho's authorship, there is one other curious circumstance.— In P. Oxy, 1232, our poem begins at the head of col. ii; now at the foot of col. i there was apparently a blank space equivalent to several lines. Enough papyrus remains on the righthand side of the foot of col. i to assure us that, if lines of β2 had been there, we should have been able to see the ends of them. It is suggested(1) that this spaee contained a note referring to the Echtheihproblem of β2 (this cannot be disproved; but it is unlikely, since fragments of it should have been visible), (2) that thespace was purposely left to separate the ‘spurious’ β2 at the end of the Book from the ‘genuine’ poems which preceded it. This also cannot be disproved; but it is mere speculation.

^{page 11 note 1} Theocr. XXVIII. 7 leptu, in an ‘Aeolic’ poem: but he has too many strange notions about the dialect.

^{page 12 note 1} The spelling is curious here and elsewhere. In v. 33 π has ỏνκαλ⋯ονΤες: we should have expected ⋯ν in both places (because one abnormality would naturally introduce another). ⋯νκλήογΤες and ⋯ν⋯ρουσε are an odd mixture. Cf. the varities -μείφνυΤο, μίγνυΤο within a lew lines of each other.

^{page 12 note 2} Whatever explanation is accepted, let it not be Jurenka's: Sappho wrote φίλοις dative ‘weil das lesbische nicht in den Vers ging’.

^{page 13 note 1} The reading in β2 is not really uncertain: πορφύρ[a] κάΤ ⋯ύΤ[με]να it must have been, though I do not pretend to know what it means. I cannot find the word ⋯υΤμήν elsewhere in classical Greek, except once in the Iliad and once in the Odyssey.

^{page 13 note 2} It is by no means obvious that the presence of an Attic form in an Abnormal Poem apparently by Sappho is sufficient reason for abjudieating the poem from her. For (1) poetry was written in Attica long before Sappho's lifetime, cf. the famous hexameter δς ν⋯ν δρφησΤ⋯ν πάνΤων ⋯ΤαλώΤαΤα παίζει (the continuation Το⋯Το δεκ⋯ν μιν is—at least for the last six letters—mere speculation); the inscription is incised, but is probably not later than the end of the eighth century (seeFurtwaengler, , Mitth. d. Ath. Jnst. VI 106 sq.)Google Scholar. One may be pardoned for thinking that a good deal of poetry was composed before house holders, however gracefully they danced, began to scratch hexameters on undistinguished vases to prove their ownership. (2) If it be said that the dialect of this inscription is not necessarily Attic, it must at least be confessed that Solon, who went to Asia Minor in his youth, admittedsome Attic forms in his poetry (e.g. παρούσας, 3. 9 Diehl, μονάρφον 10. 9 Diehl). It is not at all improbable that his songs were sung in Lesbos during Sappho's lifetime. Of the contact of Lesbos with Attica in the Sigean War and of the Stele of Phanodikos pur I say nothing. (3) Does ’ΑΤθίς= ‘the girl from Attica’? If so, here is a ready explanation orf Sappho's familiarity (if she was familiar) with Attic forms in poetry.—She learntthem from Atthis. Γογγύλα is ‘Dumpling’; Μίκα is ‘Tiny’; Γυρίννω is ‘Tadpole’; Gorgo, of whom we only know that someone ‘had had quite enough of her, is, naturally, ‘The Gorgon’, and appropriately belonged (I suppose) to Andromeda's rival establishment. So did Gello, ‘Baby-Snatcher’ (Hesych. s.v. δαίμων ⋯ν γυναῖκες Τ⋯ νεογνά παιδίαφασίίν άρπ⋯ζειν). We must at least allow that (a) some dactylicpoetry admitted a very sparse admixture of Attic forms during Sappho's lifetime, (b) Sappho could easily have been acquainted with some of that poetry. These considerations, together with our profound ignorance of the beginnings of Attic poetry, should give us pause. At the same time I have no reply if asked whether Attic forms in Ionic may justify Attic forms in Lesbian.

^{page 14 note 1} The a is missing, but there is only room for one letter, and that letter must have been a, not 17, because a ‘long’ stroke is clearly visible above the gap.

^{page 14 note 2} X 441.

^{page 14 note 3} This argument may not be as weak as it seems at first sight: we know too little about this side of Greek Stilgefühl: I have to admit frankly that it is possible that Sappho might have acquiesced in ēā from adjectives in ης and nouns in -ος (3rd decl.), and yet have recoiled in horrorfrom -ēā from a 2nd decl. adj. in -εος.

^{page 15 note 1} I am indebted to Mr. Lobel for several important criticisms and suggestions: but my acknowledgment must not be taken to imply that he agrees with my conclusions. I am also grateful to Mr. Bowra, who, besides first suggesting that πορφύρα might be an Atticized version of an Ionic form, generously assisted my discussion of the view which he expressed in New Chapters. He has modified his own viewin his forthcoming publication on ‘Greek Lyric Poetry from Alcraan to Simonides’.