^{page 68 note 1} The inscriptional material, carefully collected and conveniently put together in a different sense and for a different purpose, is contained in Th. Eckinger's pamphlet die Orthographic Lateinischer Wörter in Griechischen Inschriften (München, [1892]), to which booklet (141 pp.) I refer once for all in order save time and space and to avoid needless and disturbing repetition by constant reference to the same source.

^{page 68 note 2} Griechische Urkunden zu Berlin.

^{page 70 note 1} I remark against Prof. Blass (The Pron. Of Ancient Greek, English trans, p. 35) that if linteum were faithfully rendered by λ⋯ντε∘ν this form would require contraction to λιντ∘⋯ (like κ⋯νε∘ν καν∘⋯ν), and thus become almost irrecognizable.

^{page 70 note 2} For the converse process, that is to say for the change of Creek u and o during Graeco-Roman times, compare: (K. Brugmann, Gr. Gram.³ 51).

^{page 71 note 1} For the converse process of changing Greek υ(= V) into ου during Graeco-Roman times compare: for for Eὔτυχ∘σ (Blass. Pron. 75), ἱκετε∘νσ⋯σης(Brugmann Gr. Grant. 50).

^{page 72 note 1} Compare Schol. in Dion. Thr. 346, 23(Hilgard): ; also Quint. Inst. I. 4. IO.: nisi quis putat etiam ex tribus vocalibus syllabam fieri.