¹ I am indebted to him for permission to publish his discovery and for generous assistance in the work, and also to Professor Wade-Gery, Dr. M. N. Tod, and Professor A. Raubitschek for reading through the manuscript and making valuable corrections.

^1a For the bibliography of IG II² 10, see M. N. Tod, Greek Historical Inscriptions II, no. 100.

^1b I am indebted to him and to Mr. Kotzias, who, during my residence at the British School in 1947–48, was ephor of Attica, for permission to publish them, also to Mr. R. M. Cook for photographing them and to the Greek Archaeological Council for having them moved to the Epigraphical Museum, where they are now numbered 13103a and b.

² The term ‘chequer-unit’ is not used because it could not apply to the columns of names, which are not strictly stoichedon, or even to the decree, which is stoichedon but had not been ruled out in squares, as can be seen from note 3.

³ The distance from the centre of the alpha in αὶ in line 8 to that of the alpha in προστατ- is between 0·303 and 0·305 m.

⁴ Two columns are just possible, but not likely.

⁵ AM XXV 34.

⁶ RivFil LI 287–308, especially 303.

⁷ ÖJh XXI 167.

⁸ The breadths of the preserved columns on Face B are: Aigina fragments, first column, 0·144 m. at line 67, 0·147 m. at 79, 0·151 at 97. EM 8147, first column, 0·116 m., second 0·114 m. The left edge of the first on Face B of EM 8147 corresponds with the left side of the tau of προστα - - in line 8 on Face A, that of the following column to the space between alpha and upsilon in αὐτοϊς in line 5, and that of the following to the left-hand side of the alpha in Ἀθηναίοις On Face A in IG II² 2403, the maximum breadth of that fragment is 0·23 m., the distance from the first letter in the last line to the edge is over 0–165 m. and under 0·;179 m and the column would have been at least 0·17 m. across if Dr. Peek's restoration of line 13 in a letter to me is right, Ἀπολλωνὶ̣δ̣η̣ς σαν[δαλ(σποιός )]. On Face A of the fragments from Aigina the distance from the left edge of the inscribed surface to the right edge of the fragment is 0·178 m. These measurements fit best with columns of about 0·2 m. across, possibly a little more. That, and the measurements on Face B, suggest that the breadth of the stele was rather more than 0·8 m.

The disposition of the columns on Face A in Fig. 4 is not intended to be more than approximate.

⁹ The heading, of course, might not have been symmetrical. The centres of other headings of decrees, which were inset as if meant to be symmetrical, were sometimes not exactly in the centre of the stone, such as IG II² 1 (Kirchner, Imagines pl. 19) and 111. Sometimes it is over the central letter of the decree but not in the centre of the stone, as in I² 76 and 145, headings which are, however, not inset. In IG I² 115 and IG II² 1742 measurements from the squeezes suggest that it was in the centre. Gross asymmetry was usually avoided, which argues against Ziebarth, (AM XXIII 28)Google Scholar, who suggested that the secretary's demotic appeared in the heading. A demotic to the left of ἐγραμμάτευε would make the line too long for the formula, while one to the right would be unusual and make the heading grossly unsymmetrical.

¹⁰ If we were to place the name of Hippothontis on Face B of the Aigina fragments after the Aigeis of E.M. 8147, and thus assume fewer than seven columns and a narrower stele, then the edge of IG II² 2403. might have been original, because we could assume four columns on Face A. Such an arrangement on Face B, is, however, unlikely, for reasons discussed on p. 111, and also because it would be improbable if there were an equal number of people in each tribe. It might be possible to squeeze it in if we assume five columns on Face A and eight on Face B, but we have seen that this is unlikely.

¹¹ ÖJh XXI 164.

¹² One way to avoid either of these alternatives would be to assume that the decree began with ἔδοξεν τῆι βολῆι only, as do IG II² 6 (c. 404/3, see p. 117), 13 (399/8), 16, 17, and 18 (394/3), or ἔδοξεν τῆι βολῆι as in IG II² 26 (394–387), and possibly 3 (ca. 403/2). This would mean that there were six columns on Face B and the breadth of the stele would be about 0·72 m. There would then be three columns on Face A. These, and the heading, would be more unsymmetrical than is really likely.

¹³ See, for example, REG XL 88, note 2.

¹⁴ Stelai were not usually as broad as that if they were 0·135 m. thick. IG II² 1700 is 1·57 m. high, 0·9 m. broad, and 0·1 m. thick; and 1713 is 0·155 m. thick and over a metre broad. But if there were an equal number of people in the tribes, to postulate eight columns would involve the unlikely assumption that the sides were inscribed as well.

¹⁵ There may not, of course, have been an equal number of people in each tribe. They may have been assigned to the tribes in which they had been fighting in the war (cf. Lysias XIII 79), or have been distributed in some other way which we do not know. In that case we cannot tell how they were distributed on the stone, but an arrangement which implies either more or fewer than seven columns on Face B is, for reasons discussed on p. 104, unlikely.

¹⁶ If there were six columns on Face B and three on Face A, we should have to assume two lists if there were an equal number in each tribe. It would be a further argument that the list on Face A was continued on Face B if the tribal name [ΑΚΑΜΑΝΤΙΔ]ΟΣ were restored in line 2 of the Aigina fragment of Face A, which is likely if the letters of that line were stoichedon ([ΙΠΠΟθΩΝΤΙΔ]ΟΣ being just possible), but the letters of tribal names were sometimes spaced more widely than those of the lines above and below (see Fig. 1, Face B) in which case it could have been [ΑΝΤΙΟΧΙΔ]ΟΣ The surface is destroyed where the delta should be.

¹⁷ AM XXV 38.

¹⁸ Greek Historical Inscriptions 80.

¹⁹ Hesperia X 286.

²⁰ See, e.g., Koerte, , AM XXV 393.Google Scholar

²¹ Cf. Wilhelm, , ÖJh XXI 161.Google Scholar A rather unlikely explanation of this term will be found in REG XL 91.

²² Xenophon, , Hellenica II 4, 25Google Scholar; AM XXIII 36; ÖJh XXI 162. The formulae for grants of isoteleia will be found in the Index to IG II², but if there were seven columns of names on Face B the formula used was probably shorter than the average.

²³ Vit. X Or. 846a, 835e and ƒ. The two measures are usually identified, but it will be seen from the above references that the decree about Lysias was probably an honorary decree for a single man beginning ἐπειδὴ . . . followed by a statement of what he had done. The manuscript reading ἐπεί in 835ƒ has, however, been emended to ἔπειτε or ἐκεῖ. If there were two decrees Archinos quashed them both (836a and b).

²⁴ Hellenica II 4, 25.

²⁵ Plut. de Glor. Ath. 349ƒ, Vit. X Or. 835ƒ, 836a; Ἀθπολ. 39 and 41. For the events see especially Xen. Hell. II 4, 26–;42, and we must also consider the date of Lysias XII and of the decree of Phormisios opposed in Lysias XXXIV (p. 691 in the Loeb edition).

We might apply surgery and suggest that Thrasyboulos' decree, Phormisios' decree, and the proposal of isoteleia in Xenophon were all proposed on the same day, and identify the most likely one with IG II² 10, or that the Ἀθπολ. was mistaken about either the contents, the procedure, or the fate of the proposal of Thrasyboulos.

²⁶ Pythodoros, if we accept the identification of PA 12389 with 12412, was an oligarch and had probably gone to Eleusis. His name was later expunged from the list of archons, and if it was engraved on a decree would offer a good excuse to anybody who wanted the decree declared illegal. But the name seems to have been used in later inscriptions (IG II² 1371, and 1498 line 21, both restored).

²⁷ Athenaeus XIII 577b. Cf. REG XL 90, note 2, 99.

²⁸ Names for trades are discussed by A. Calderini, La manomissione e la condizione dei liberti in Grecia, 350–356, M. Clerc, Les Métèques athéniens, 390–;418, and Tod, M. N., BSA VIII 203–211Google Scholar, hereinafter referred to as Calderini, Clerc, and Tod. See also Tod, , ‘Epigraphical Notes on Freedmen's Professions’, in Epigraphica XII, 18–22.Google Scholar The references for the use of the words here discussed will be found in LS⁹ unless otherwise stated.

²⁹ Buck and Petersen, Reverse Index 90.

³⁰ SIG ³ 22.

³¹ Tod, 205, cf. Plato, Tht. 178d. For slaves on the land see Glover, T. R., The Challenge of the Greek 66–7.Google Scholar

³² Cf. Tod, 210, and IG II² 1673, line 4.

³³ Calderini, 353. The word might, of course, be μισθωτής, which could refer to a metic, cf. IG II² 1672, line 180.

³⁴ Diels, Fragmente der Vorsokratiker B70.

³⁵ BSA VIII 210–211.

³⁶ Clerc, 411.

³⁷ Loc. cit. 205–6, 209.

³⁸ Loc. cit. 397–403. See also Knorringa, H., Emporos (Amsterdam, 1926)Google Scholar, passim.

³⁹ Buck and Petersen, 627–628.

⁴⁰ Cf. Menander, Sam. 75.

⁴¹ The Athenian Expounders of the Sacred and Ancestral Law (Baltimore, 1950), Chap. I. On the Κῆρυξ, if that is the reading in line 17, I have no conjecture.

⁴² Loc. cit., 204.

⁴³ Loc. cit., 209, note 6.

⁴⁴ Strabo IV 6, 2, p. 202.

⁴⁵ IG II² 7864.

⁴⁶ IG XI (4). 598 and 1135.

⁴⁷ Names derived from ethnics and ending in —άδης are listed by Bechtel in HP 547, and we may compare Ναχσιάδες in IG I² 943, line 79. This last was probably given because of the Athenian capture of Naxos, cf. Tod, , GHI I ²112, p. 102.Google Scholar

^47a Cf. Πολεμων IV (1949), 1–2, p. 79, note 8.

⁴⁸ Athenaeus IV 173b.

^48a OCT reads, with one MS, Krethon.

⁴⁹ Polyaenus VIII 46. For the date see RE XVII 2, 2085.

⁵⁰ IG XII (9). 249A, line 6. For the date see Hesperia XVI 116–118.

⁵¹ Fouilles de Delphes III i, 111.

⁵² Paus. II 31, 6.

⁵³ Suidas s.v. Ἑρμώνιος χάρις, Adler I 416.

⁵⁴ Thuc. IV 58. Plut., Apophth. Lac. 222e, Alc. 25, Thuc. VIII 92. Cf. Vit. X Or. 835ƒ.

⁵⁵ Hdt. VI 126; FGrH (Jacoby) 105, F2, but cf. Hdt. VI 126.

⁵⁶ IG I² 374, line 62.

⁵⁷ IG II² 10660, between 390 and 365 B.C., and IG II² Addenda Nova 7328a.

⁵⁸ Dedications 408, no. 381.

⁵⁹ IG II² 9202, of a Lesbian, and 12194.

⁶⁰ IG II² 1533.

⁶¹ IG II² Addenda 7292A.

⁶² IG I² 801.

⁶³ Aristophanes, Equ. 1256, Vesp. 1220.

⁶⁴ IG II² 1534, lines 82 and 210.

⁶⁵ Suidas, s.v. Δημήτριος, Adler II 40.

⁶⁶ IG II² 1939, line 6.

⁶⁷ Dem. XXIX 23, IG II² 4353, 6557.

⁶⁸ IG XII (9). 245B, line 208, 247, line 6, IG VII 261; cf. the index to IG XII (9).

⁶⁹ IG XI (2). 144A, line 22, and very frequently during the third and second centuries.

⁷⁰ Athenaeus IV 161c.

⁷¹ I 9, 16.

⁷² Hesperia XV 173.

⁷³ IG II² 1951, lines 177 and 394.

⁷⁴ Cf. Koerte, , PhW LII (1932), 83–8.Google Scholar

⁷⁵ IG XII (8). 275 and 279.

⁷⁶ Inscr. Cret. I 21, p. 137.

⁷⁷ IG XI (4). 1040, line 3, Fouilles de Delphes III i, 359.

⁷⁸ HP 485ƒ.

⁷⁹ IG II² 1256.

⁸⁰ IG II² 1534, line 215. Without a demotic.

⁸¹ IG II² 1958, line 68.

⁸² IG XII (9). 1005.

⁸³ IG IV 731.

⁸⁴ Inscr. Cret. I p. 31; it also occurs on pp. 74 and 279, and II p. 205.

⁸⁵ IG XI (2). 114, line 17 (the first occurrence of the name of that person), and, another Dexios, , IG XI (4). 767 and elsewhere.Google Scholar

⁸⁶ Celsus V 18, 36, and, for a client of Lysias in the fourth century, Pape-Benseler s.v. ‘Δέξιος’ (= Bekker, Anecdota 129).

⁸⁷ Robinson and Fluck, Greek Love-names 112–113, but this is not likely to be the archer of IG I² 949. The name may also occur as that of a slave in IG I² 333.

⁸⁸ Diogenes Laertius IX 2, 11 but see Lucian, Macr. 20.

⁸⁹ Aristotle, , Rhet. III 17, 1418⁶ 31.Google Scholar

⁹⁰ FGrH IIIa, no. 262.

⁹¹ Schol. ad Aristoph. Pacem 680.

⁹² IG II² 5453.

⁹³ Plut. Apophth. Lac. 233e.

⁹⁴ Xen. Hell. V 4, 3.

⁹⁵ IG VII 3160.

⁹⁶ FdD III i, 161.

⁹⁷ Inscr. Délos 370, line 29. The reading is doubtful.

⁹⁸ IG II² 2280.

⁹⁹ RE III 2180.

¹⁰⁰ Suidas s.v.

¹⁰¹ See Pape-Benseler s.v.

¹⁰² References in Pape-Benseler, s.v. ‘Χαῖρις’, and RE III 2031; Suidas s.v. Χαῖρις (Adler IV 305), who refers to a second fifth-century harp player.

¹⁰³ Hesperia VII 88.

¹⁰⁴ IG IV² (1). 102, line 87.

¹⁰⁵ IG IV² (1). 71, line 35, between 242/1 and 238/7, in one of the years in which Aratos was not general of the Achaean League.

¹⁰⁶ References in Pape-Benseler, s.v. ‘Χαῖρις’.

¹⁰⁷ Greek Love-names, 90.

¹⁰⁸ Ibid., 2.

¹⁰⁹ I could find no parallel, but cf. Psen, a name which, like Κνίψ, means ‘gall-insect’, in IG XII (3). 591, an archaic inscription from Thera; Knopon, IG IV² (1). 94b, line 26; Konops, a Phrygian, IG II² 10487, and Kniphon, IG I² 943 line 66.

¹¹⁰ IG II² 6105.

¹¹¹ Strabo XIV 633.

¹¹² Empedocles, Diels, Vorsokratiker B123.

¹¹³ IG II² 1675; but we do not know the status of the deceased in IG II² 5329 and 11336.

¹¹⁴ IG IV 552, in the epichoric script. The inscription was copied by Fourmont.

¹¹⁵ TAM III 1, 897, reading doubtful; cf. 904. Cf. also Onialles, TAM II 3, 1207.

¹¹⁶ TAM III 1, 872, p. 265.

¹¹⁷ Diog. Laert. I 1, 22.

¹¹⁸ Hesperia, Suppl. VIII 362.

¹¹⁹ Ovid, , Met. V 203.Google Scholar

¹²⁰ Suidas, s.v. Αστυάγης, Adler I 393; RE II 1865. In the second century there was an Ἀστυάγου πάγος at Ephesus, (Forschungen in Ephesos I 17).Google ScholarCf. also Alyattes, in Hesperia XV 171Google Scholar and Tantalos, in IG II ²2392, line 16.Google Scholar

¹²¹ Wuthnow, Die semitischen Menschennamen 8. Cf. BSA XLVI (1951), 225–6 for Ἀβδόκως.

¹²² Dem. XXIX 5.

¹²³ Hdt. I 103.

¹²⁴ TAM II 3, 1101.

¹²⁵ TAM II 1, 168a, line 45, IG II² 9277, Inscr. Délos 384. Professor Bonfante suggests that the ending -yes is ultimately of Illyrian origin.

¹²⁶ Aristotle, , Pol. V 9, 22, 1315^b 26.Google Scholar

¹²⁷ Xen. Ephes. III 11.

¹²⁸ IG IV² (1). 264, IG II² 8818, Inscr. Cret. II p. 28, Athenaeus XIII 586e, IG IV 732 IV line 16.

¹²⁹ IG XII Suppl. 493, 460, 504, CIG 2034, IG III 3619 (not in IG II²), IG II² 9223. An exception is IG XII Suppl. 585, from Eretria.

¹³⁰ Diller, Race Mixture 143.

¹³¹ Stevens and Paton, The Erechtheum 384 no. XIV, 392 no. XVII, and 396 no XVII, a Gerys who was employed in channelling the columns. Ibid., 330 no. IX and 340 no. XI, another Gerys who sawed timber.

¹³² Which Wilhelm, , Att. Urkunden V 41Google Scholar, associates with IG II² 6, and dates about 402 B.C.

¹³³ No.

¹³⁴ IG II² 10995.

¹³⁵ See also Keil, in ÖJh XXVIII Beibl. 124.Google Scholar

¹³⁶ Cf. Fouilles de Delphes III i, 41.

¹³⁷ Hdt. I 34.

¹³⁸ In Pape-Benseler, s.v. ‘Ἄτυς’ They give an Atyanas of Adramyttion, Olympic victor in 72 B.C., and Attas occurs in Delos (Inscr. Délos 1603), Attes in Herakleia in the fourth century (IG II² 8699), Atteous as a genitive in TAM III 1, 144, 784 and 792, and further examples of similar names will be found in Pape-Benseler.

¹³⁹ The word φανερῶς may even suggest that the author saw the stone, and either read a mention of slaves in the part that is now missing or read the names and drew the same inference as is drawn here. The unusual substitution of τεχνικά for patronymics may be a device to conceal the servile origin of some of the beneficiaries. Some of the slaves, if this is correct, such as the χωρὶς οἰκοῦντες (for whom see Diller, Race Mixture 146), would find it easy enough to give their trade, but others may have found it more difficult, and that may be why Egersis, for example, has no trade recorded after the name.