[611] For σάωσαν Ar. read “σαώσαι”, which is defensible on much the same grounds as “πεφεύγοι” above. The use of the sing, verb, even though πόδες is one of the nominatives, is sufficiently supported by 2.339, 3.327, Od. 14.291, cf. 17.387. Bekker has ingeniously but needlessly conjectured that both readings spring from an older “σαώσαιν” = “σαώσειαν”, of which he finds another instance by reading “ἄλφοιν” for “α<*>´λφοι” in Od. 20.383. This form of the 3rd pl. opt. is sufficiently established by inscriptions from Delphi for the thematic tense (-“οιν”), and we should therefore be justified in assuming its existence in the aor. form, if there were any need. See G. Meyer Gr. § 589, H. G. § 83.
