¹ I am most grateful to Mrs. Dina Peppas-Delmouzou, Director of the National Epigraphical Museum, and her able staff for facilitating my study of the three inscriptions discussed in sections 1 and 2. I thank also D. M. Lewis, R. G. Osborne, P. J. Rhodes, and J. S. Traill, who kindly read a draft of this article and made criticisms and suggestions which have enabled me to improve it. The following special abbreviations are used:

APF Davies, J. K., Athenian Propertied Families 600–300 B.C. (Oxford 1971)Google Scholar

Eliot Eliot, C. W. J., Coastal Demes of Attika: A Study of the Policy of Kleisthenes [Phoenix Suppl., 5] (Toronto 1962)Google Scholar

Lauter Lauter, H., ‘Zwei Horos-Inschriften bei Vari: Zu Grenzziehung und Demenlokalisierung in Süidost-Attika’, AA (1982) 299–315Google Scholar

Petropoulakou-Pentazos Petropoulakou, M. and Pentazos, E., Ἁττική. Οίκιστικά στοιχεῖα—πρώτη ἔκθεση [Ancient Greek Cities, 21] (Athens 1973)Google Scholar

Siewert, Trittyen Siewert, P., Die Trittyen Attikas und die Heeresreform des Kleisthenes [Vestigia, 33] (München 1982)Google Scholar

Traill, POA Traill, J. S., The Political Organization of Attica: A Study of the Demes Trittyes, and Phylai, and their Representation in the Athenian Council [Hesperia Suppl., 14] (Princeton 1975)Google Scholar

Traill, Studies Traill, J. S., ‘An Interpretation of Six Rock-Cut Inscriptions in the Attic Demes of Lamptrai’, in Studies in Attic Epigraphy, History and Topography Presented to Eugene Vanderpool [Hesperia Suppl., 19] (Princeton 1982) 162–71Google Scholar

² See the commentary of Meritt, B. D. and Traill, J. S., The Athenian Agora, XV: Inscriptions: The Athenian Councillors (Princeton 1974) 54.Google Scholar In general my readings are closer to those of J. Kirchner in IG ii² 1750 than to those of Meritt and Traill: for example, rather than in line 68. There is definitely a chip out of the upper block which makes Kirchner's reading in line 2 preferable to in Agora XV, let alone in Siewert, Trittyen 17. The line-numbering in Agora XV is used in this article.

³ The new join is to be announced by MrsDelmouzou, in ADelt 32 (1977)Google Scholar

⁴ Thompson, W. E., Mnemosyne 22 (1969) 151 n. 1.CrossRefGoogle Scholar

⁵ EM 10598, not 10589 as in IG ii² 1750 and Agora XV, PP. 54. 484.

⁶ Siewert, Trittyen 19 n. 82, referring to Bicknell, P. J., Historia 27 (1978) 373–4.Google Scholar

⁷ For discussion see Kaloyeropoulou, A. G. and Vanderpool, E., ADelt 25 (1970) A' 204–16, especially 215–16Google Scholar; Bicknell, P. J., Historia 27 (1978) 369–74Google Scholar, Siewert, Trittyen 21 n. 90, 102–3 and 92 and my article ‘The Tribal Reform of Kleisthenes the Alkmeonid’, forthcoming in Chiron 14 (1984), under ‘X. Antiokhis’.

⁸ Siewert (Trittyen 18–19) accepts that the demes are grouped according to regional trittyes in this inscription. See also Löper, R., AM 17 (1892) 426Google Scholar, J. Kirchner on IG ii² 1750, Meritt, B. D. and Traill, J. S. on Agora XV 44.Google Scholar

⁹ (mason ) in column ii line 45 is aligned with the last councillor from Atene in column i and the heading in column iii (all deme headings on the inscription are in slightly larger letters). In fact the top 21 lines across the inscription (lines 4–24 in column i, lines 25–45 in column ii and lines 49–69 in column iii) are perfectly aligned, apart from the use of the interlinear space for Diogenes Diogeitonos (line 51).

¹⁰ Originally published by Kirchner, J., AM 29 (1904) 244–53Google Scholar; later IG ii² 1700, Agora XV 43. The officials in lines 227–34 are discussed by Meritt, B. D., Hesperia 10 (1941) 44–6.Google Scholar Meritt and Traill silently correct Kirchner's misalignment between columns vii and viii: the last patronymic from Kekropis, Dexixenou (line 175), should be level with Thrason (line 193), as in Agora XV, not with the patronymic Thrasymedou (line 194).

¹¹ Cf. Meritt and Traill on Agora XV 43 (p. 54).

¹² This is seen in lines 52, 54 bis (for consistency's sake it would have been better to assign a number to the line after 54 at the foot of column ii), 131 and 133.

¹³ So J. S. Traill, POA 126 s.v. Lamptrai. There (s.v. Pergase) Traill concludes on prosopographical grounds that the Pergase in line 11 was Lower Pergase and the other Upper Pergase.

¹⁴ This is also Traill's conclusion (POA 14 and 78 with 6).

¹⁵ Agora XV 17, on which see Traill's comments (POA 9, 78).

¹⁶ POA 78 (cf. 5).

¹⁷ As suggested (p. 54) in the commentary on Agora XV 43.

¹⁸ POA 78.

¹⁹ POA 6, 11 –12, 78. This is preferable to believing that two denies failed to send representatives in 335/4 B.C. (commentary on Agora XV 43, p. 54).

²⁰ As recognized in POA 9–10, 78. The foot of p. 53 in Agora XV should be emended to read, under ‘lacuna of 94 lines including 8 demotics and 1 line uninscribed’. Since three demotics appear in the preserved part of the roster, there cannot be nine demotics missing.

²¹ Commentary on Agora XV 43, p. 54.

²² POA 16 n. 20.

²³ POA 78; cf. 13–14.

²⁴ Agora XV, p. 55.

²⁵ Cf. Traill, POA 11.

²⁶ Sypalettos is located by the deme decree IG i³ 245, found south-west of Koukouvaounes. The general location of Daidalidai is indicated by the poletai inscription edited by Crosby, M., Hesperia 10 (1941) 14–27CrossRefGoogle Scholar (especially lines 10–12 and pp. 20–1). The location of Pithos, listed between Sypalettos and Daidalidai on Agora XV 43, is uncertain.

²⁷ Dow, S., Prytaneis: A Study of the Inscriptions Honoring the Athenian Councillors [Hesperia Suppl., 1] (Athens 1937)Google Scholar no. 116. 45 = Agora XV 293. 45.

²⁸ Walbank, M. B., Hesperia 51 (1982) 43–5.Google Scholar

²⁹ APF 99, cf. 103, 114.

³⁰ W. Peek, op. cit. 28. For the verbs I have suggested cf. (Klearchos of Soloi, fr. 37 Wehrli), (Philochoros, FGH 328 F 73), (Polemon, fr. 78 Preller), all cited by Athenaios 6. 234D–235E.

³¹ APF 191, 290 (cf. 389), 409, 470, 511.

³² APF 409, 470, 511. Euthydemos: APF 191.

³³ APF 389.

³⁴ APF 409, referring to SDAW (1934) 1023 no. 1A line 70. The demotic is certain (eight names from Kydantidai occur together), but the lack of a patronymic must allow the possibility that the subject of the curse is a relative and not the Nikokles on this inscription.

³⁵ APF 511–12.

³⁶ Davies, J. K., Historia 18 (1969) 309–33.Google Scholar

³⁷ Cawkwell, G. L., Historia 22 (1973) 759–61.Google Scholar Cawkwell also asked if the business on which the triremes sailed out (lines 88–9) was less exciting than trying to establish a kleruchy.

³⁸ Lewis, D. M., BSA 50 (1955) 34–5Google Scholar collects evidence in support of a high average age for the hieropoioi for the Pythais of (?) 326/5 B.C. (SIG ³ 296 = FdD iii 1. 3 [1929] no. 511) and for the commissioners for the Amphiareia of 329 B.C. (SIG ³ 298 = IG vii 4254).

³⁹ Siewert, Trittyen 82 n. 256, 118 and n. 164.

⁴⁰ Lykourgos, Klearchos, Demades, Glauketes, Kleochares, Nikeratos, Neoptolemos (SIG ³ 296): see APF 83, 89, 99–101, 351, 397, 399–400, 406–7. Three of these men were earlier epimelelai for the Amphiareia of 329 B.C. (SIG ³ 298). It is tempting to think that Phanodemos, who also appears on the two lists discussed by D. M. Lewis (see n. 38 above), was a member of a wealthy family.

⁴¹ W. Peek, op. cit. 28, 29 n. 2.

⁴² Historia 12 (1963) 39 (additional note); cf. 33–4. He points out that Gargettos, Acharnai, Pallene, and Paiania come from four different trittyes and four different tribes.

⁴³ There is an Olympiodoros from Gargettos in the fourth century: Meritt, B. D., Hesperia 36 (1967) 58–9CrossRefGoogle Scholar (no. 4) line 5 = SEG 116. 5.

⁴⁴ The horos inscriptions from Attika which record security transactions have recently been reconsidered by Millett, P., Opus 1 (1982) 219–49.Google Scholar He reaches substantially the same conclusions on their significance for economic and social history as Finley, M. I., Studies in Land and Credit in Ancient Athens, 500–200 B.C.: The Horos-Inscriptions (New Brunswick 1952).Google Scholar

⁴⁵ Willemsen, F., AM 80 (1965) 100–26 at 122–3 with Beilagen 39–40.Google Scholar

⁴⁶ Eliot 56–8.

⁴⁷ Traill, Studies 168 n. 22. The site was no doubt as impressive to the fifth-century Greeks as to the builders of the chapel.

⁴⁸ Unfortunately Willemsen did not illustrate his no. 6.

⁴⁹ Based on Willemsen's Abb. 3, which in turn is based on Eliot's fig. 5.

⁵⁰ The two inscriptions are discussed together by Lauter, H., AA (1982) 299–314.Google Scholar Lauter knew the earlier report by Eliot (63–4), but an important article on the rock-cut inscriptions near Thiti was published while Lauter's was in press: Traill, Studies 162–71. Lauter, pp. 302–3, reported that in 1980 the two easternmost inscriptions, nos. 1 and 2 on Eliot's sketch (Eliot 57 = nos. 1 and 2 on Traill's map, p. 163), had been destroyed by bulldozers preparing the area for sale to Athenians for ‘weekenders’. However, no. 2 is too high up the slope to have suffered this fate (yet) and was still preserved in March 1983. It is evident that Lauter mistook Eliot's no. 2 for no. 3, for he published (Abb. 3) a photograph of no. 2 (earlier published as no. 2 (p. 122) by F. Willemsen, op. cit. Beil 40:4 and confirmed as no. 2 by Traill, Studies, pl. 21b) and described it as no. 3.

⁵¹ Lauter 308–9, 311.

⁵² IGii² 1204; Eliot 52–3 and n. 15, 55–6, 60–1; Traill, POA 38; Traill, Studies 165.

⁵³ Notably Traill, J. S., Hesperia 47 (1978) 89–109CrossRefGoogle Scholar, Studies 166–71 and Siewert, Trittyen. For the literature on ‘equal trittyes’, an idea beginning with W. E. Thompson, see Rhodes, P. J., A Commentary on the Aristotelian Athenaion Politeia (Oxford 1981) 533Google Scholar and Siewert, Trittyen 5–6 nn. 23–8. Rhodes, himself (Historia 20 [1971] 402)Google Scholar offered an early warning: ‘these dangers are so great as to make it unwise to place any reliance on membership of Thompson's divisions in locating demes in their [sc. regional] .

⁵⁴ D. M. Lewis makes this point in a review of Siewert, 's book in Gnomon 55 (1983) 435Google Scholar; cf. Lauter 314.

⁵⁵ Traill, Studies 164 and n. 3. Although offering essentially the same expansion of the inscriptions as Traill, Lauter (303) calls the S-shaped sign a symbol of abbreviation.

⁵⁶ Lauter 303–4 with notes.

⁵⁷ See, for example, Bicknell, P. J., Antichthon 7 (1973) 1–4CrossRefGoogle Scholar; cf. Lauter 300.

⁵⁸ Milchhöfer, A., Untersuchungen über die Demenordnung des Kleisthenes, (Abh. Königl. Preuss. Akad. Wiss., 1892, Berlin 1892) 12–13Google Scholar with map at back; Löper, R., AM 17 (1892) 319–433Google Scholar at 341, 348–9; Judeich, W., map to RE ii 2 (Stuttgart 1896) 2204 (facing 1999–2000).Google Scholar

⁵⁹ Bicknell, P. J., Mnemosyne 28 (1975) 57–62CrossRefGoogle Scholar at 58 n. 15; Siewert, Trittyen 105 n. 105.

⁶⁰ Eliot 58–9 n. 31 (he is attracted to a site well up in the hills, D on his fig. 4 = Petropoulakou–Pentazos X₇–Y₃ no. 2, for Kedoi or Pambotadai; see 58–9 and 136 n. 1); cf. Siewert, Trittyen 104 n. 102.

⁶¹ See my article ‘The Tribal Reform of Kleisthenes the Alkmeonid’ (n. 7 above), under ‘I. Erekhtheis’.

⁶² Traill, POA 38–9 n. 9.

⁶³ Traill, Studies 167 n. 17; for his earlier view (assignment to the inland ‘modified geographical trittys’) see Hesperia 47 (1978) 105.

⁶⁴ Commuting to Aghios Dhimitrios from a population centre in the plain ending at Aghia Marina would be easier than from Kitsi; cf. Eliot 61.

⁶⁵ The reason why scholars began looking for two words is no doubt that ΠΜ scarcely begins a Greek name (cf. Eliot 64, Lauter 303). Once Vanderpool's earlier suggestion had been published (Eliot 64), it was natural for scholars to seek to replace the suggestion with two words.

⁶⁶ Eliot 64.

⁶⁷ Ober, J., Hesperia 50 (1981) 68–77, 73–7 with pl. 28.CrossRefGoogle Scholar In the short time since Ober wrote the bus-route has changed from 52 to 224, the main avenue has been renamed and inscription no. 2 has evidently been destroyed in the construction of a look-out road down from the peak and along the saddle (March 1983).

⁶⁸ Lauter 314.

⁶⁹ Cf. Lauter 299; J. Ober, loc. cit. 74.

⁷⁰ Cf. Lauter 299 and n. 2 with J. Ober, loc. cit. 76. Lauter would be reluctant to come down to the ‘Kaiserzeit’.

⁷¹ Pittakis, K. S., EphArch (1853) 858 (no. 1378)Google Scholar; IG ii² 1212; Traill, POA 38 and n. 8; Siewert, Trittyen 43 and n. 46, 104–5.

⁷² Papakhristodhoulou, I., ADelt 25 (1970) B'1 123.Google Scholar See also my article ‘The Tribal Reform of Kleisthenes the Alkmeonid’ (n. 7 above), under ‘IV. Leontis’.

⁷³ Geroulanos, M. I., BSA 68 (1973) 446, 448Google Scholar reports that the best honey is produced in June, after the thyme starts to bloom near Athens (his estate is at Trahones, about 5 km south of Alepovouni). This honey is creamed off in the first half of July; a second harvest follows in September.

⁷⁴ Thompson, W. E., Historia 15 (1966) 1–10Google Scholar; Mnemosyne 22 (1969) 137–52; cf. Rhodes, P. J., Historia 20 (1971) 385–404 and n. 53 above.Google Scholar

⁷⁵ The arrangement of the roster in Hesperia 47 (1978) 90–1 repeats that on Agora XV 13. Although Traill uses this repeated arrangement to support (for example) the topographical location of Upper and Lower Potamos, he nevertheless pursues ‘modified geographical trittyes’ in this article and in Studies, as does Siewert in his book.

⁷⁶ Dow, S. in Bonfante, L. and von Heintze, H. (edd.), In Memoriam Otto J. Brendel: Essays in Archaeology and the Humanities (Mainz 1976) 69–84 at 72–9.Google Scholar

⁷⁷ Traill, J. S. first made this suggestion (Hesperia 47 [1978] 109)Google Scholar in an addendum to an article in which he used the term ‘trittyes of prytaneis’ but then stated a preference for ‘modified geographical trittyes’ (ibid., after n. 79).

⁷⁸ Eliot, C. W. J., Phoenix 22 (1968) 3–17CrossRefGoogle Scholar questioned the use of lot by Kleisthenes and it is now generally accepted that there is little room for sortition if unequal trittyes yielded tribes of almost equal size (e.g. Traill, POA 70–1 and n. 31, P. J. Rhodes, op. cit. [n. 53] 253). We must keep in mind, however, that the total number of bouleutai in a trittys may not be a safe guide to its population or to the number of hoplites which that trittys contributed to the tribal contingent in the army. Huxley, G. L., The Times Literary Supplement (14 May 1982) 538Google Scholar emends the suspect to in Athenaion Politeia 21.4, but further changes to the text would then be necessary.

⁷⁹ Traill, POA Maps 1–3; Siewert, Trittyen. His Map 1 repeáts Traill's Map 1, while Map 4 shows his own scheme of arithmetically precise ‘Prytanen-Trittyen’. Traill's maps, including a revised version of Map 2, are available separately from the book.

⁸⁰ Eliot 21, 3–4, 33–4, 45–6, etc.

⁸¹ R. G. Osborne will expound these and other points in his forthcoming book Demos: The Discovery of Classical Attika, to be published by Cambridge University Press in the Cambridge Classical Studies series.

⁸² ‘Dema House’: Jones, J. E., Sackett, L. H., and Graham, A. J., BSA 57 (1962) 75–114Google Scholar; Jones, J. E., Archaeology 16 (1963) 276–83.Google Scholar There is not much evidence of farming activities at this site; enough fragments for a little grinding and a little weaving, and not much else. Note the quest for a second floor to explain the lack of rooms, especially bedrooms: BSA 57 (1962) 111–13; Archaeology 16 (1963) 283. Vari Cave House: Jones, J. E., Graham, A. J., and Sackett, L. H., BSA 68 (1973) 355–452Google Scholar (quotation from 356); date between c. 350 and c. 275 B.C.: Ibid. 360, 363, 364, 374,415–18. The authors note the lack of Attic country houses so far discovered: BSA 57 (1962) 102 andn. 29, BSA 68 (1973) 418, as does Jones, J. E., AAA 7 (1974) 293–313Google Scholar at 294, 304, and in Mussche, H.et al., Thorikos and the Laurion in Archaic and Classical Times [Miscellanea Graeca, 1] (Ghent 1975) 63–136 at 106–16.Google Scholar

⁸³ Young, J. H., Hesperia 10 (1941) 163–91CrossRefGoogle Scholar, especially 178 and 190–1 with figs. 10–11 on 179–80. Young describes a farm ‘boundary wall’ leading towards the farmhouse which he identifies on his map (fig. 1 on 164) with the of the Salaminioi inscription no. 2 (lines 28–9, 33). Inscriptions: Ferguson, W. S., Hesperia 7 (1938) 1–74.CrossRefGoogle Scholar Deme centre: Kordhellas, A., AM 19 (1894) 240–4.Google Scholar For possible farmhouses nearer the deme centre in the Agrileza valley, see Young, J. H., Hesperia 25 (1956) 122–46CrossRefGoogle Scholar; J. E. Jones in Thorikos and the Laurion (see n. 82) 116–19; Langdonand, M. K.Watrous, L. V., Hesperia 46 (1977) 162–77CrossRefGoogle Scholar; Wickens, J. M., Hesperia 52 (1983) 96–9.CrossRefGoogle Scholar

⁸⁴ Lazaridhis, D., AAA 1 (1968) 31–5 at 31Google Scholar; cf. (1960) 30–1, (1961) 37–8.

⁸⁵ For a guide, see Index to Ordnance Survey of Lincolnshire on the Scale of Six Inches to One Mile (Southampton 1891), accompanying the first edition of the survey, Ordnance Survey of England and Wales: Combined Index Shewing Civil Parishes and the Ordnance Survey Maps of Lincolnshire (Southampton 1930) or Stamp, L. D.et al., The Land of Britain: The Report of the Land Utilisation Survey of Britain, vii: Eastern England, Parts 76–7 (London 1942) 478, 502–4.Google Scholar To the west of this set of parishes, between the River Brant and the limestone ridge (‘the Heath’) there is a less marked but similar tendency for elongated parishes between Harmston and Caythorpe. The pattern is not dissimilar to the north of Lincoln, with parishes on either side of Ermine Street running down from the Cliff to the rivers Till in the west and Ancholme in the east. I am grateful to Dr. R. J. Olney for advice about the area around Lincoln.

⁸⁶ Beresford, M., History on the Ground: Six Studies in Maps and Landscapes, 2nd edn. (London 1971) 26–7, 31–3, 35–7.Google Scholar

⁸⁷ Whitlock, R., The Shaping of the Countryside (London 1979) 54–5Google Scholar pictures the Anglo-Saxons coming up the rivers and settling in a kind of ribbon development. Because the only direction for expansion was uphill, many downland farms today constitute a narrow strip of territory, no more than two fields broad but sometimes six miles long, from a meadow or two by the river, through a zone of cultivated fields on the gentle slopes, to an area of downland on the sides and top of the hill. Much work has been done recently on land-use and settlement patterns going back to the Bronze Age in Britain: see, for example, Bowen, H. C. and Fowler, P. J., Early Land Allotment in the British Isles: A Survey of Recent Work [BAR British Series, 48] (Oxford 1978)Google Scholar and Sheppard, J. A., The Origins and Evolution of Field and Settlement Patterns in the Herefordshire Manor of Marden [Dept. of Geography, Queen Mary College, Univ. of London, Occasional Papers, 15] (London 1979), especially 23–37.Google Scholar

⁸⁸ Herodotos 9. 15. 3: Mardonios' army Myres, J. L., Herodotus: Father of History (Oxford 1953) 285–6Google Scholar; Burn, A. R., Persia and the Greeks: The Defence of the West, c. 546–478 B.C. (London 1962) 511.Google Scholar

⁸⁹ Bradford, J., Antiquaries Journal 36 (1956) 172–80CrossRefGoogle Scholar and pl. 9–10; Ancient Landscapes: Studies in Field Archaeology (London 1957) 29–34 and pl. 6, 10.

⁹⁰ So Thompson, W. E., SO 46 (1971) 72–9CrossRefGoogle Scholar and P. J. Rhodes, op. cit. (n. 53) 252. This was true of many ‘city’ demes; for those in the built-up area, however, see Larsen, J. A. O., CP 68 (1973) 45–6.Google Scholar