^{page 1 note 1} xxiv. 12–13; PA 2946.

^{page 1 note 2} IG ii.² 1421; in his original publication of this stone (AJA xviii (1914) 1–17) A. C. Johnson mistakenly restored the heading so as to date the Secretaryship of Glauketes to 371/70 B.C.; the correct restoration, giving his date as 374/3. was published by Kolbe, , Philologus xxxviii (1928) 263Google Scholar; cf. IG ii.² pt. ii, Add. p. 799.

^{page 3 note 1} For Theophemos, cf. PA 7091–8; for Theodemos, ibid. 6746 (only); for Thoudemos, ibid. 7247–8.

^{page 3 note 2} It is much to be hoped that Professor Louis Robert will in due course shed further light on this text. Having seen the note in which he stated his intention of returning to it (Ét. épigr. et philol., p. 62, 6), I ascertained that he had no objection to my publishing these notes on the subject, and I wish to record my warm appreciation of his approval.

^{page 3 note 3} RA xviii (1941) 215 f.

^{page 4 note 1} Attische Urkunden i (1911) 8 f.

^{page 4 note 2} In the Budé edition (edd. Mathieu and Haussoullier) we have χαὶ τοὺτων αὺδεμὶα ὲν τῶ αὺτὼ ὲγγὶγνε[ται. Νῦν]δὲ πρὸσκειται [καὶ ῾Η]φαὶσ[τια]ὲπὶ κηφισοφῶντος ὰρχοντος whilst Blass⁴ (Teubner, 1903) had suggested τοὺτων οὺδεμιᾶ ὲν τῶ αὺτῶ ὲν[ιαυτῶ] γὶγνετ[αι. ῾Εχτη] δὲ πρὸσκειται κτλ Neither reading leaves any doubt that penteteric games were in existence in Eleusis before the archonship of Kephisophon (329/8 B.C.).

^{page 4 note 3} The number of vacant spaces after μὲδιμνοι at the end of line 260, shown by Kirchner as eight, cannot be right, since the total in line 262, namely ΗΗΗΗ . . . . is made up of four payments, as follows: 244 medimnoi, given κατὰ τὰ πὰτρια to the priests and priestesses over the past four years (lines 255–8), plus two payments of 70 medimnoi each, to which is to be added the amount shown as occupying eight spaces; but assuming that the total of 400 plus four symbols (for which there are sixteen choices available between 404 and 480 inclusive) is correct, there is no sum of eight symbols which, added to 384, will give a total of ΗΗΗΗ . . . . If we reduce the eight vacant spaces to seven, we find that there are seven alternative amounts, of which the most likely in this context would be either 69, 78,87, or 96, which when added to 384 would give us either 453 or 462 or 471 or 480, in each case exactly filling the eight spaces in line 262.

^{page 5 note 1} Cf. Wilhelm's discussion of such expressions as οὶ μὲχρι Θερμοπυλῶν Ελληνες (Thuc. ii. 101, 2) in Att. Urk. i. 10–11.

^{page 5 note 2} ᾿Αθ. πολ 60, 1.

^{page 6 note 1} There is no need to cite parallels for this common phrase, of which a good example is quoted in note 3, p. 4 above. Cf. for many other examples the index to SIG ³ s.v.

^{page 6 note 2} Cf. Meisterhans, GAI ³ III; for fifth-century examples, IG. i.² 16. 25 (restored); 39. 59; 82. 16; 108. 34; 115. 7; IG ii.² 1.27 (restored). This usage seems to have been much less common in the fourth century, cf. IG ii.² 27. 10 (c. 394–387 B.C.); 141.14 (not later than 360 B.C.), which all strengthen the case against Schweigert's restoration here.

^{page 6 note 3} Cf. GAI ³ 45, n. 302; the use of -εια for -εα appears to be most commonly found in the period 350–300 B.C. For γραμματὲια cf. IG ii.² 226. 19; 530. 3; 2943. 15–16.

^{page 7 note 1} The decree honouring Nikon of Abydos (IG ii. ²493 + 518)Google Scholar is quoted below, p. 8.

^{page 7 note 2} Meritt kindly informs me that Wilhelm, in a letter, had suggested for the end of line 14 the name φ[ιλοὶτ]ιον and would restore φ[ιλοὶ]τιος in Hesp. xv. 178, no. 25, line 6.