In the former part of this paper I started from the fact that the use in Attic prose of forms in -οσυνος, -οσυνη, is limited by certain rules as to meaning and etymology, which do not apply to Ionic literature; and I observed that the usage of the Attic tragedians followed in these respects the practice of the Ionians. This being so, I advanced the hypothesis that the tragedians and their audiences must have been aware of the peculiar character of these words, and that upon investigation the passages of tragedy in which they occurred would be found to have in other ways also an Ionic colour or a special connection with Ionic language and tradition, which colour or connection the reader must perceive if he would fully appreciate the tone, and in some cases the meaning, of the author. So far as the limit of my space extended this anticipation was, I think, fulfilled. I showed that in some cases certainly (p. 272), in others almost certainly (p. 279 foll.), these words were actually inflected according to the Ionic rule.