218 Journal of Philology. Boissonade's text is : Zpiraur AxtXXvs xw *a&. Bergk adopts Dissen's emendations. And Rauchenstein (ZeiUchrift f. d. Alter' thumsvnssenschaft, 1844, no. 51. p. 407) gives us the following arrangement of the text: /3apv 8e a(j>t 6tjk velnos Kafifias a}iat ap Ax*Xevs a<j> appaTav. Or : /ea/3/3as x a H-^ ls p" AxXw. As no one of these proposals has been generally received, it is not surprising that I cannot acquiesce altogether in any of them, or that I think the true reading still undiscovered. It appears to me that it still lurks in the traces of the old text, which is, -a(pi vcIkos cpTvca AxiXXevs X a F a * xcififias, Ka$3ay, or in one MS. Kaphas. The first step is to consider the meaning of the context. The scholiast, from whose fiapelav Kal iiraxOrj pdxyv Suit (pikoveuclav avrots <-WSetei> Dissen has borrowed his $* velicos, obvi- ously did not understand the passage. Pindar says that the fame of the jEacidae had spread afar by land and sea, and had leapt at one bound even to the ^Ethiopians. Then follows our text, which ought surely to signify that Achilles, by slaying Memnon, had caused them, far removed as they were, heavy sorrow or pain. There is nothing in the context or in the mytho- logy of the ^Eacidae to justify the term vukos here. Pindar must have written /Sapv nevOos (as in 01. n. 25), or, what would be the same thing in his style, Papi> ZXkos (see Pyth. u. 91). The name AxtXXcvs written with two X's, which is the less common ortho- graphy in Pindar, and transferred from its proper place before a(p* appAravy to the end of the previous line, contains the last traces of cKkos, which ought to be restored to the text without its gloss axos* ; and I am the more convinced of this because the pre- ceding word fin(r seems to me to have sprung from Kappas, a faulty reading for Kafids, which must have preceded Ax'Xffa in some copies. If then we remove (pirfa-' AxXXct? from the end of the line and leave them to be represented by Kafias, or, as in the Augustinus codex, by Kaphas AxiXevr, and retain only ZXkos from Ax-iXXeus, we shall find the necessary verb cvtitcv in the substan- tive v(7kos, and changing bco-cpivtiKos into 8' t?$ o-<p' cveuccv, we shall have a reading required by the context, and not very far removed from the traces of the MSS. If however we take the
- Ax-tXXevf=dxo$-?/co.
