¹ Studies in the History and Art of the Eastern Provinces, pp. 305 to 378. The inscriptions in that article are quoted as Q 1 etc. (Q = Quatercentenary Publication, Aberdeen).

² Waggons can go to Saghir empty, but not with any load.

³ [Κυρί]ου Σεβ[αστοῦ] Κυρίων Αύτοκρατόρων Δεῒ Κυρίῳ The Ormelian priests were of native families (ἐκ γένους) succeeding by some unknown rule.

⁴ These publicani under the Empire were of totally different character from those of Republican times; and all comparisons between them ought to disappear from commentaries and works on New Testament times:· their true character has been shown by Rostowzew, Studien z. Gesch. d. röm. Staatspacht and after him by Ramsay, in Hastings' Dict. Bib. v. p. 394Google Scholar b.

⁵ In Studies, p. 309, I say that ΚΤΙСΑΝ 3 is uncertain. These letters are quite clear, yet give a hopeless reading: Calder notes that all six letters are certain.

⁶ 1905 corrects ΤΗС to ΤΟС: as the letters became blurred and broad, С was evolved out of Υ.

⁷ On the contrary, Ziebarth, , Griech. Vereinswesen, p. 67Google Scholar, regards Anagrapheus and Brabeutai as officers of the Association.

⁸ Luke uses also κατακλίνεσθαι All four Gospels and Septuagint use also ἀναπίπτειν ἀνά has the distributive sense in these compounds.

⁹ I put this in a rough fashion, implying no definite opinion as to local usage. The term ἀνακλίνεσθαι has not yet been found in Egyptian papyri; but perhaps the idea does not occur.

¹⁰ Ziebarth, , Griech. Vereinswesen, pp. 41, 65Google Scholar.

¹¹ Euseb, . Hist. Eccles. viii., ix. 3–9Google Scholar; Lactantius, , M.P. 36Google Scholar, 37. Ramsay, , Pauline and Other Studies, Art. iv., quotes many illustrartions from inscriptions: see also Cit. and Bish. of Phrygia, ii. p. 567Google Scholar.

¹² Sacraments, at any rate baptism, were Mithraic.

¹³ This view that the rite was performed with twice-fired bread, δπτύρψ, has been proposed by Mr.Reinach, A. J.Google Scholar (not observing my suggestion of it as possible on p. 349, though neither of us has made a restoration in accordance with this idea). His excellent paper is used in the sequel.

¹⁴ Παπᾶς either bad grammar (like δόντοσ with nominative nouns, and other solecisms), or due to remembrance of a Phrygian genitive. τυ[ιτηνοῦ καὶ] with a second name is too long. Yet τῦ for τοῦ is a unique misspelling.

¹⁵ Reinach, in Revue Celtique, 1907, pp. 225Google Scholar f. The thonght of διπύ[ρῳ] occurred to me too late for the text p. 319, when that sheet was already on the machine; I could only add the reference in the note to p. 349, where I have mentioned this possibility, quoting some evidence that ordinary bread was avoided in the Phrygian ritual, but confessing inability ‘to see how the sign could be exhibited by means of the twice-fired bread.’ It is, however, now easy to see how well this adapts itself to the newly discovered Protanaklites.

¹⁶ The Christian authorities say that the priests ate no bread.

¹⁷ A second oase is now known: Miss Hardie's paper, No. 2.

¹⁸ Sterrett prints in his epigraphic copy ΠΑΠΑΜΑ. My notebook of 1886 gives the text correctly (as in 1911); but presumably I accidentally omitted the ⊏ in the copy which I sent him; and thus Ποπᾶ appears in his text and hence in Q 9.

¹⁹ The inflexion of nouns in -ϵύς troubled the composer seriously: he uses -ϵος and -ϵως in nom., -ϵύς in gen.

²⁰ For nouns in -ϵύς see note 19.

²¹ As Miss Hardie suggested.

²² Misspelt Kranasana here.

²³ Ethnie before father's name, as in Q 15, 11; Q 2, 29.

²⁴ On the flat top of the cippus are three small circular bosses.

²⁵ Keller, in Berl. Phil. Woch. 1896, p. 118Google Scholar and Lewy, Semit. Fremdwörter in Griech. (Berlin 1895Google Scholar) holds that Semitic ch has been dropped in various Greek words, ἄβρα = Chabrā (Keller, , Volkselymol. p. 196Google Scholar), Εὔα Eve = Chawwa (Vulg. Heva), ἄριζοσ = Charis, ἀπήνη and καπήνη = Chāpap or Chābā, εὐνοῦχοσ Chānūk (approved on trial), ἀβαλαι (i.e. Φεῦ) = Chabāl. De Cara takes ᾿´ Γδη= Κύδη and quotes Muséon, Apr. 1891 on ὐ= κυ in Carian, Lydian, etc. city-names. Lightfoot, , Philip, p. 51Google Scholar explains the name Gangites or Angites at Philippi (Appian, iv. p. 106, Herod. vii. 113), modern Anghita, on the theory that the initial was ‘a guttural sound like Semitic ayin, sometimes omitted, sometimes represented by γ—[as in Gaza and Aza, alternative renderings of G.F.H.]

²⁶ Zulueta, in Oxford, Studies, i. 2, p. 60Google Scholar; od. Theod. xi. 24, 6, 7; B.G.U. 12, 10–11.

²⁷ The epigraphic copies of 26, 27 B, D, and 20 A, are by Miss Hardie, who intended to do the present paper, but had to leave for Athens too soon.