Sanakirja.org

Esimerkit

• These are of a sort found especially on Athos: an example is four epsila, which refer to the Holy Cross and stand for Ἐκ θεοῦ Ἐδόθη Εὔρημα Ἑλένη. Byzant. Zeitschr., xxvii.
• …what is the source of k’- in Arm. k’san “twenty?” The explanation, I believe, lies, like that for the initial epsila of East Greek ἑϜίϰοσι and the pan-Greek ἑϰατόν, in the syntax of old numeral phrases.
• Round, normal-size epsila (l. 3) contrast with a large, ungainly epsilon (l. 7).
• Finally, for (b), ll.17 and 20, with the epsila indented from the names below, Bradeen suggested ἐ[ν] or ἐ[πί] respectively, followed by a geographic heading.
• To the right of → 6–8 the few remaining fibers are devoid of writing indicating that this might be a remnant of the original margin. Unfortunately, at the corresponding place on the other side, the vertical fibers are missing thus making any definitive judgment on the matter impossible. However, as Roy Kotansky observed, the epsila of ↓ 7–9 are vertically aligned so neatly under each other that they too probably represent the first letters of each line.
• In the subscription one letter, the final epsilon at the end of line 28, is written in exactly the same way as the epsila in the main body of the text, quite unlike the other epsila in the subscription.
• Epsila in the Getty recto lines 3 and 7 and PFayum 4.3, 4, 6, 7, 9, and 12 receive similar treatment. In all examples a tendency toward angularity is modified by smooth curves where horizontal members meet the vertical shaft. Isolated examples, however, in Pfayum 4.10 and 11, are the clearly squared epsila pointed out as a “remarkable palaeographical feature” by Grenfell and Hunt.²⁵ Compare epsila in the Getty recto text lines 1 and 3, where a tendency toward angularity can be observed.
• Inconsistent letter forms: e.g. upsila in 1.2; kappas in 1.7; long rho in πρωτον vs. short rho in ημετερα; epsila in επελθω.