Acknowledgements and Abbreviations

We are indebted to a number of scholars for their help and advice, particularly to Professor Sir Moses Finley, Professor M. H. Jameson, Professor A. M. Snodgrass, Dr. J. M. Wagstaff, and Mr. B. Quiller. We should like to acknowledge the help given to us in Greece by the British School at Athens and by the former and current Ephors of Antiquities for Lakonia and Arkadia, Mr. G. Steinhauer and Mr. Th. Spyropoulos, who have kindly allowed us to make use of unpublished information from recent work under their direction by the Greek Archaeological Service. We are grateful to Dr. J. L. O'Neil and Dr. J. Roy for allowing us to use material from their University of Cambridge doctoral dissertations; also to P. Halstead and to H. A. Forbes, H. A. Koster, and L. Foxhall for permission to cite articles as yet unpublished. Any remaining errors are our own responsibility.

Earlier versions of this paper were read to the Ancient City seminar at the University of Cambridge in March 1977 and to the Topics in Current Research seminar at the University of Manchester in May 1978. Subsequent fieldwork was generously supported by an award from the British Academy's Small Grants Research Fund in the Humanities.

Special abbreviations used for the citation of modern works are as follows:

Amit = Amit, M., Great and Small Poleis (Brussels, 1973)Google Scholar Part III ‘Sparta and Mantinea’.

Bölte = RE xiv² (1930) 1289–1344, s.v. ‘Mantinea’.

Buck = Buck, C. D., The Greek Dialects (Chicago, 1955).Google Scholar

Fougères, MAO = Fougères, G., Mantinée et l'Arcadie Orientale (Paris, 1898).Google Scholar

Frazer = Frazer, J. G., Pausanias's Description of Greece iv (London, 1898).Google Scholar

Hope Simpson–Dickinson = Simpson, R. Hope and Dickinson, O. T. P. K., A Gazetteer of Aegean Civilisation in the Bronze Age i, Studies in Mediterranean Archaeology lii (1979).Google Scholar

Howell = Howell, R., ‘A Survey of Eastern Arcadia in Prehistory’, BSA lxv (1970) 79–127.Google Scholar

Moggi = Moggi, M., I Sinecismi Interstatali Greci (Pisa, 1976).Google Scholar

O'Neil, GDCOA = O'Neil, J. L., Greek Democratic Constitutions outside Athens (University of Cambridge, Ph.D. dissertation, 1974).Google Scholar

Pritchett = Pritchett, W. K., Studies in Ancient Greek Topography ii, University of California Classical Studies iv (1968) ch. V ‘The Battles of Mantineia’.Google Scholar

Roy, SHA = Roy, J., Studies in the history of Arcadia in the fifth and fourth centuries B.C. (University of Cambridge, Ph.D. dissertation, 1968).Google Scholar

¹ The town itself survived even the enslavement of its population and the importation of new settlers by Antigonus Doson and Aratos in 223 B.C. (Plb. ii. 56. 6–7, 58. 12, 62. 11–12; Plut. Arat. xlv. 4–6). On its final period of inhabitation, Fougères, MAO 518. See also the recent archaeological evidence discovered by Steinhauer, G., ADelt xxix (1973/1974) Chr 296–301, at 300Google Scholar; cf. JHS c (1980) AR (1979/80) 33.

² Moggi 140–56, 251–6 gives references to modern historical discussions and reproduces all the ancient evidence.

³ For the details below, Loring, W., ‘Some Ancient Routes in the Peloponnese’, JHS xv (1895) 25–89.CrossRefGoogle Scholar

⁴ Humphreys, S. C., ‘Town and Country in Ancient Greece’, in Ucko, P. J. et al. (edd.), Man, Settlement and Urbanism (London, 1972) 763–8Google Scholar; Finley, M. I., The Ancient Economy (London, 1975) ch. V.Google Scholar

⁵ Simpson, R. Hope, A Gazetteer and Atlas of Mycenaean Sites, University of London, Institute of Classical Studies, Bulletin Supplement xvi (1965) 39–40 nos. 87–8Google Scholar; Howell 85–8 nos. 9–16; Hope Simpson–Dickinson 79–80, B17–20; Kahrstedt, U., Das wirtschaftliche Gesicht Griechenlands in der Kaiserzeit (Bern, 1954) 132–6Google Scholar; Bölte 1309–16. Cf. also the less systematic discussion of settlement in Rigopoulos, N., Ἱστορία τῆς Μαντινίας (Athens, 1938).Google Scholar

⁶ We visited the Mantinike in March–April 1979, August 1979, and April 1980. This study incorporates information concerning excavations or finds made before April 1980 in so far as it has been available to us. (The most recent relevant volume of ADelt Chr to appear before the time of writing was that covering the years 1973/4) Systematic consultation of periodicals ceased in July 1980.

⁷ Finley, M. I., The Ancient Greeks (Harmondsworth, 1966) 54Google Scholar; cf. ‘The Ancient City: From Fustel de Coulanges to Max Weber and Beyond’, Comparative Studies in Society and History xix (1977) 305–27, at 325.

⁸ Humphreys, op. cit. (n. 4) 764; Moggi, M., ‘Συνοικίζειν in Tucidide’, ASNP Serie iii, v (1975) 915–24.Google Scholar

⁹ For the detailed geography and topography, see esp. Fougères, MAO 1–129; Bölte 1293–316. Howell 85 provides a succinct summary.

¹⁰ Pritchett 43 challenges the usual assumption that the border lay at the narrowest part of the plain, preferring to locate it along the line of katavothras a little way S. The construction of a watch-tower on the spur of Mytika, probably in the fourth century (Loring op. cit. (n. 3) 82–3; Lattermann, H., ‘Nestane und das Argon Pedion’, AA xxviii (1913) 395–428, at 425–7Google Scholar; Pritchett 45–6) suggests that the Mantineians expected secure control of that promontory, since it commands an extensive view over the plain of Mantineia, including the town. Several early nineteenth-century travellers discovered remains of a wall near the base of Mytika which some of them interpreted as a border wall. This explanation is unlikely. (Cf. Pritchett 44–5 and Fougères, , MAO 113, 126 n. 2Google Scholar for alternative explanations and references to earlier interpretations. The remains had disappeared by the time of Fougères' expedition in the late 1880s.) In view of Thucydides' remarks (v 65. 4) on Mantineian–Tegean struggles over the water at their border, it is reasonable to expect that the border may often have been moved.

¹¹ See Section II, B below; Loring, Lattermann, and Pritchett, ibid.

¹² X. HG vi 5. 17–19; Loring, ibid. 86; Lattermann, ibid. 394.

¹³ On these communities, Roy, J., ‘Tribalism in south-western Arcadia in the Classical Period’, Acta Antiqua xx (1972) 43–51.Google Scholar Cf. the existence of separate Mainalian representatives in IG v 2. 1 = Tod 132 (probably dating between 368 and 361) and their ultimate participation in the state of Megalopolis (Paus. viii 27. 3). The E. Mainalian communities were probably foremost among the subject allies who fought with the Mantineians in 418 but whose independence had subsequently to be conceded (Th. v 58. 1; 81. 1).

¹⁴ Termination of the ditch: Pritchett 55–6 with refs. As he remarks (n. 75), no archaeological evidence has ever been adduced to support hypotheses concerning the precise location of the Elisphasioi. Hellenistic coin: Gardner, P., BMC Peloponnesus (London, 1887) 14 no. 163.Google Scholar

¹⁵ Fougères, MAO 127–9; Bölte 1312, 1330. Fougères conjectured that they were a deme in spite of his opinion (128) that ‘ce peuple de bergers se trouvait ainsi rejeté hors des limites de la Mantinique propre, celle-ci finissant à la ligne des katavothres au pied des hauteurs de Simiadès et de Kapsia’.

¹⁶ Walbank, F. W., An Historical Commentary on Polybius (Oxford, 1957–1979) ii 286Google Scholar; Moggi 147–8.

¹⁷ Jameson, M. H., ‘Excavations at Porto Cheli and its Vicinity, Preliminary Report, I: Halieis, 1962–68’, Hesperia xxxvii (1969) 311–42CrossRefGoogle Scholar, at 315 n. 12. The site was not re-occupied until late Roman times.

¹⁸ Ibid. 313–15; Jameson, M. H., ‘The Southern Argolid: the Setting for Historical and Cultural Studies’, in Dimen, M. and Friedl, E. (edd.), Regional Variation in Modern Greece and Cyprus, Annals of the New York Academy of Sciences cclxviii (1976) 74–91, at 83.Google Scholar

¹⁹ Cf. recently Sokolowski, F., Lois Sacrées des Cités Grecques (Paris, 1969) no. 60.Google Scholar

²⁰ Martin, R., ‘Rapports entre les structures urbaines et les modes des division et d'exploitation du territoire’, in Finley, M. I. (ed.), Problèmes de la terre en Grèce ancienne (Paris, 1973) 97–112, at 107–8.Google Scholar The absence of sanctuaries from the Roman period is probably explained by the abandonment of rural habitation attested by Pausanias. Under the Empire large estates owned by town-dwellers or outsiders dominated the Mantinike: Kahrstedt, op. cit. (n. 5) 132–6.

²¹ Howell 79 n. 1 lists the main works of nineteenth-century travellers.

²² Cf. the accounts of Hope Simpson, Howell, and Hope Simpson–Dickinson cited in n. 5 above.

²³ Clark, W. G., Peloponnesus (London, 1858) 128Google Scholar; Fougères, MAO 92–3; Frazer 199–200; Lattermann op. cit. (n. 10) with a plan at 409–10; Rigopoulos, op. cit. (n. 5) 41–2; Hope Simpson, ibid. 39–40 no. 88; Howell 87 no. 14; Levi, P. (trans.), Pausanias, Guide to Greece (Harmondsworth, 1971) 386 n. 49Google Scholar; Winter, F. E., Greek Fortifications (London, 1971) 52 n. 114Google Scholar, 193 n. 10, 216–17 and fig. 215; Hope Simpson–Dickinson 80, B19; personal observation.

²⁴ Lattermann ibid. 415 n. 2, citing Fougères, MAO 142, fig. 21 (section between Gates G and H); cf. n. 53 below.

²⁵ See the map in Lattermann, ibid. 399–400.

²⁶ Apud Stephanus of Byzantium, s.v. Νοστία

²⁷ FGH ii C 27, 103.

²⁸ FGH ii B 31.

²⁹ Fougères, MAO 108–9; Lattermann, op. cit. (n. 10) 425–7; Howell 88 no. 16; Hope Simpson–Dickinson 79, B17; personal observation.

³⁰ Howell and Hope Simpson–Dickinson label the tower Hellenistic without explanation. Prof. A. M. Snodgrass (pers. comm.) agrees with the possibility of a fourth-century date for the masonry, part of which (the N. side of the tower) is illustrated in Plate 47c.

³¹ Loring op. cit. (n. 3) 84–5. Fougères, MAO 121–3 effectively rebutted Loring's case but did not avail himself of the most conclusive argument. Pausanias (viii 12.5) states that to the right of the first route stood a tall hillock called the grave of Penelope. For Loring to be correct, the easterly route must have passed to the E. of Penelope's grave; otherwise it would have had the grave to its right and would have been closer to it than the westerly route. This in turn makes his view dependent upon the identification of Penelope's grave with the hill of Gourtsouli rather than with a lower hill just over 1 km further N. The easterly route cannot have passed to the E. of the northerly hill without difficulty because a ridge connects that hill to the southern slope of Mt. Armenias. Excavation, however, has now proven that Gourtsouli was not the grave of Penelope but the site of Ptolis (Section II, F below). Consequently Penelope's grave will have been the northerly hill. It will have stood close to the right of the easterly route which must therefore have been the first route described by Pausanias. Fougères did not use this argument because of his mistaken identification of the two hills. In his day no ancient remains had yet been found on Gourtsouli. Bölte 1303 gives refs. to earlier discussions of the problem of Pausanias' routes, to which add Frazer 221–2.

³² Howell 86 (f); Levi, op. cit. (n. 23) 402 n. 93; personal observation. Site bearings: Artemision church 35°; summit of Gourtsouli 140°; village of Simiades 235°.

³³ Boblaye, M. E. Puillon, Recherches géographiques sur les ruines de la Morée (Paris, 1835) 149Google Scholar; Fougères, MAO 119; Frazer 222; Meyer, E., RE xix 1 (1928) 605–6Google Scholar, s.v. ‘Maira’; personal observation.

³⁴ The civic centre of the town (agora, theatre, etc.) is the most likely point from which Virlet made his measurement of 5,500 m since it accords best with the location of his site as 1,200 m S. of Kakouri. A measurement of 5,500 m. from the northernmost point of the town walls would place the site within the outskirts of the village.

³⁵ Puillon Boblaye, op. cit. (n. 33) 149; Fougères, MAO 119–20; Frazer 222; information from Mr. G. Steinhauer; personal observation. We are indebted to Prof. M. H. Jameson who inspected these sites with us.

³⁶ J. Pečírka, ‘Homestead Farms in Classical and Hellenistic Hellas’, in M. I. Finley (ed.), op. cit. (n. 20) 113–47.

³⁷ Fougères MAO 84; Bölte 1309–10; Kahrstedt, op. cit. (n. 5) 135; information from Mr. G. Steinhauer; personal observation.

³⁸ The pass was until recently called Portes. For its identification, see Frazer 194–6, with refs.; Fougères, MAO 83–4; Bölte 1299; Levi, op. cit. (n. 23) 383 n. 42.

³⁹ Jost, M., ‘Pausanias en Megalopolitide’, REA lxxv (1973) 241–67, esp. 253–8CrossRefGoogle Scholar; ‘Les traces de Pausanias en Arcadie’, RA (1974) 179–86 esp. 185–6. Bölte gave Melangeia village status.

⁴⁰ Karayiorga, Th., ADelt xviii (1963) Chr 88–9Google Scholar, with references (at 88) to earlier identifications of Ptolis; Daux, G., BCH lxxxvii (1963) 766–7Google Scholar; Hope Simpson, op. cit. (n. 5) 39 no. 87; Simpson, R. Hope and Lazenhy, J. F., The Catalogue of Ships in Homer's Iliad (Oxford, 1970) 92–3Google Scholar; Howell 86–7 no. 11; Hope Simpson–Dickinson 79–80, B18; personal observation.

⁴¹ Ventris, M. and Chadwick, J., Documents in Mycenaean Greek, 2nd ed. (Cambridge, 1973) ch. iii, esp. 83.Google Scholar We are indebted to Paul Cartledge for drawing our attention to these points.

⁴² A single Early Helladic sherd was also discovered among a deposit of sherds ranging from the Geometric period to the sixth century B.C. See below, Site 2.

⁴³ The site notations are our own. For convenient reference Sites 1–4 are discussed in the same order as in Karayiorga's excavation report.

⁴⁴ Daux's report was published before Karayiorga's, which did not appear until 1965, but his wording mostly follows hers so closely that he must have used her account in compiling his report.

⁴⁵ The report by Simpson, Hope and Lazenby, in JHS lxxxii (1962)Google Scholar, AR (1961/2) 31 mentions some archaic cist-graves on the south-west slopes. This seems to have been an initial misinterpretation of what turned out on excavation to be the early sanctuary (Site 4 above).

⁴⁶ Since Karayiorga's report had not been published when Hope Simpson wrote, he repeats Daux's sixth-century terminal date. See comments on Site 3 above.

⁴⁷ The sources are cited and discussed in Moggi 140–57 no. 24, 251–6 no. 40.

⁴⁸ On the 1887–9 excavations, Fougères MAO; BCH xi (1887) 485–90; xiv (1890) 65–90, 245–71, 595–601; cf. also xii (1888) 105–28; (1892) 568–79; xx (1896) 119–66. For more recent work, ADelt xvii (1961/2) Chr 86; xviii (1962/3) Chr 90; xx (1964/5) Chr 177–8; xxix (1973/4) Chr 296–301.

⁴⁹ Fougères had to defend this procedure against the criticisms of Schliemann; see BCH xiv (1890) 271–5; MAO x–xv.

⁵⁰ For the following details, Fougères, MAO 179–81, 187–8; Martin, R., Recherches sur l'agora grecque (Paris, 1951) 380–1.Google Scholar

⁵¹ Bulle, H., Untersuchungen an griechischen Theatern (München, 1928) 248Google Scholar remarks that the analēmma is in the same polygonal style as the town walls, but assigns a fourth-century date to both.

⁵² Its neighbour immediately to the N. may be the sanctuary of Hera which could date from a similar period. We have omitted consideration of the (?) bouleuterion. Martin, op. cit. (n. 50) 381 includes it in his list of fifth-century buildings, but at 467 dates it to the fourth century. Fougères (MAO 133) also mentions ‘les débris d'édifices religieux à qui le style des chapiteaux et des colonnes assigne une date contemporaine du V^e siècle et parfois plus reculée encore’; but he does not specify to which remains he is referring.

⁵³ The account below relies on that of Scranton, R. L., Greek Walls (Cambridge, Mass., 1941) 57–9, 90, 156–7Google Scholar, 163 B1. 29, 172 C6. 12 for the classification of the different types of masonry.

The correlations of types of masonry with stretches of wall, however, are our own, based on personal observation. Scranton fails to specify exactly where each type of masonry is found, using neither compass directions nor Fougères' lettering of the gates. The sole indication he gives is erroneous. He states (57) that the polygonal masonry is found ‘at the lower end of the city, at the point where the river bed in its course around the town is deepest’ which he later identifies as the outlet of the river Ophis (mod. Fidhias) from the city. All these indications point to the N.W. part of the walls; but there is no polygonal masonry in that area. On the contrary, it is found to the S. and S.E. near the point of the inflow of the river around the town. The two photographs of the polygonal walling cited by Scranton himself confirm his error. The first (Fougères, MAO 142 fig. 21) depicts the stretch of wall three curtains W. of Gate G; the second (Scranton 58 fig. 11) that four curtains N.E. of Gate G (personal observation).

Scranton's topographical errors do not of themselves affect his classification of the different masonry styles which our own autopsy has confirmed, or the dates assigned to those styles.

⁵⁴ Winter, op. cit. (n. 23) 217 notes that the design of some of the gatecourts of Mantineia with both inner and outer entrances does not seem to occur before the fourth century; cf. also 227 fig. 233 (Gate A).

⁵⁵ Cf. Winter, ibid. 73 n. 11, 86–7 on the advantages of mud brick walls and trapezoidal masonry for speed of construction.

⁵⁶ Ibid. 80–90, 95–100.

⁵⁷ Fougères, , MAO 162–5, 376–8Google Scholar; Martin, op. cit. (n. 50) 376–82; L'Urbanisme dans la Grèce antique (Paris, 1974) 120–2. Curiously, despite his opinions on the interpretation of circuit and agora and on the latter's fifth-century origin, Fougères (MAO 133) expresses the view that the early town enclosed a smaller area than the later one.

⁵⁸ These roads are shown on the town plan in Fougères, MAO Pl. viii, with the exception of that between Gates B and I, part of which was uncovered by Steinhauer, G. in 1973: ADelt xxix (1973/1974) Chr 296–301, at 296.Google Scholar

⁵⁹ Steinhauer, ibid. and pers. comm.

⁶⁰ (1) IG v 2. 262; Jeffery, L. H., The Local Scripts of Archaic Greece (Oxford, 1961) 216 no. 29Google Scholar, cf. also 212–14, from which the following quotations are taken; Buck 198–9 no. 17. (2) IG v. 2. 261; Jeffery ibid. 216 no. 28.

⁶¹ Buck 198, whose translation is borrowed below.

⁶² In transcribing the original Greek it has been necessary for reasons of printing to represent the digamma by the Roman ‘w’

⁶³ For a recent discussion with compendious references to earlier views, Moggi 150–1.

⁶⁴ Amit 124–8, with references to Beloch and Callmer.

⁶⁵ Cf. the correct assessment of Andrewes, A., ‘Sparta and Arcadia in the Early Fifth Century’, Phoenix vi (1952) 1–5CrossRefGoogle Scholar, at 3 n. 11, ‘there seems to be no evidence for dating the συνοικισμός of Tegea’. Literary and archaeological evidence for the existence of the town does not pre-date the early fourth century: X. HG vi 5. 8; Callmer, C., Studien zur Geschichte Arkadiens (Lund, 1943) 111, 116Google Scholar; Hejnic, J., Pausanias the Periegete and the Archaic History of Arcadia, Rozpravy Československé Akademie Vĕd, Ročník lxxi Sešit xvii (1961) 107–8.Google Scholar

⁶⁶ The evidence of good relations with Sparta (Mantineia's absence from the Arkadian side at Dipaia: Hdt. ix 35. 2; her assistance against the Messenians: X. HG v 2. 3) cannot disprove the possibility of synoecism in these years. It is at least as plausible to suggest, as has Roy (SHA 175–6), that the synoecism may have taken place when Argos was no longer pursuing an anti-Spartan policy. (She did not fight against her at Dipaia; cf. Forrest, W. G., ‘Themistokles and Argos’, CQ x (1960) 221–41, at 221–9CrossRefGoogle Scholar, on the possibility of an internal change in Argos bringing her closer to Sparta.) Sparta may have approved or acquiesced in the strengthening of her only major Arkadian ally. Sparta was not always antagonistic to synoecism (cf. Kleombrotos' or Kleonymos' synoecism of Heraia, perhaps in the 370s: Strabo viii 3. 2; Roy, SHA 176 n. 31) and Argos was not always antagonistic to Sparta (cf. the events of 420 and 418/17: Th. v 40, 76–81). In any case Argos had recovered sufficiently from Sepeia to fight at Tegea (Hdt. ix 35. 2).

⁶⁷ It might be suggested (Amit himself does not do so) that the civic pride displayed by Mantineia's first coinage issue should be ascribed to a new feeling of strength deriving from the synoecism, thus dating it to the early fifth century. This is not impossible, but there are alternative reasons why she may have coined at that time. On Mantineia's early coinage, Head, B. V., Historia Numorum (2nd ed.Oxford, 1911) 449–50Google Scholar; Babelon, E., Traité des monnaies grecques et romaines (Paris, 1901–1933) ii 1. 861 ff.Google Scholar; ii 3. 631 ff.; Kraay, C. M., Archaic and Classical Greek Coins (London, 1976) 96–7.Google Scholar Its starting date is uncertain. The coinage has normally been assigned to the first decade of the century, since it is said to have been superseded by the federal issues of the Arkadian League whose beginning has usually been dated to c. 490. However, as Kraay explains, recent research suggests that the Arkadian coinage may not have started until after the Persian Wars. Kraay gives a date c. 480 for the single Mantineian coin which he depicts (no. 285); but the relation of this coin to the other early issues is not clear. The earliest hoard in which a Mantineian coin appears is GCH 1643 found at Memphis in Egypt and dated to c. 480 B.C. by Dr. G. K. Jenkins. From further information which Dr. Jenkins (pers. comm.) has kindly provided, it appears that this hoard cannot give an independent terminus ante quem for Mantineian coinage since its own dating depends largely upon the date of its Mantineian coin. Based upon a comparison with Kraay no. 285, Dr. Jenkins is now inclined to lower the date of the hoard coin to 470–60 (providing that Kraay's date is correct). This obviously calls into question the conventionally accepted sequence of the Mantineian and Arkadian federal issues. On reasons for coinage issues, see most recently Kraay ch. 13.

⁶⁸ There is nothing in the content of Demoanax' reforms (Hdt. iv 161; D.S. viii 30) to suggest that Mantineia was already synoecized.

⁶⁹ Both Fougères, MAO 360 n. 3 and Walbank op. cit. (n. 16) i. 261 dismiss the claim as tendentious. Fougères refers it to the polis, Walbank to the synoecism.

⁷⁰ The fact that Strabo (viii 3. 2) mentions Mantineia's synoecism together with that of Elis, Tegea, Heraia, Aigion, Patrai, and Dyme is no evidence for a mid-fifth-century date. Heraia's synoecism may have happened only in the 370s (n. 66 above). J. L. O'Neil, ‘The Exile of Themistokles and Democracy in the Peloponnese’, CQ (forthcoming), points out that Strabo gives no implication that the places named together in his list were involved in a common movement.

⁷¹ Examination of the arguments for each specific date would be of little profit. They are all necessarily conjectural owing to the lack of precise evidence. The 470s and 460s are obscure decades because we do not know the dates of the battles of Tegea and Dipaia and the circumstances which attended them.

⁷² This is the attempt of Fougères, MAO 421 n. 4; Martin, op. cit. (n. 50) 377 n. 3.

⁷³ Pausanias' errors: (1) viii 8. 11 implies that Mantineia's change of name to Antigoneia was a voluntary, peaceful action. In fact it took place only after the capture of the town, the enslavement of its citizens, and its resettlement with a new population (Plb. ii 58; Plut. Arat. xxxv 4). (2) viii 10. 5–8 claims that Agis IV was killed in battle against Mantineia, whereas he seems to have died in Sparta at the hands of conspirators (Plut. Agis xix if); cf. Pritchett 61–2 (with refs. to other discussions) who is inclined to reject the details of the battle entirely.

⁷⁴ Bölte 1323 suggests that only the houses of a few cult officials will have been left standing. Fougères, (MAO 132–3, 421)Google Scholar claims that the finds of fifth-century inscriptions, capitals, and columns prove that temples and sanctuaries escaped destruction.

⁷⁵ Scholars are divided as to which to prefer. See the discussion and refs. of Bölte 1323, who prefers Xenophon. More recently Amit 173 n. 200 and Moggi 152 have preferred Ephorus.

⁷⁶ They conflict over whether dioecism was one of Sparta's original demands to Mantineia and over the course of the river Ophis. Xenophon omits Mantineia's appeal for Athenian help and fails to explain why Agesipolis did not flood the town at an earlier stage of the campaign. Diodorus omits the destruction of crops, the early part of the siege, and the negotiations after the Mantineian surrender.

⁷⁷ Fougères, MAO 421 n. 4; Moggi 147.

⁷⁸ Bölte 1323, with reference to Bursian's view.

⁷⁹ Thus Fougères, MAO 128–9, 334–6; Bölte 1311–12; and recently R. Baladie, Budé ed. of Strabo, v 219. Amit 122–3 sees less of a distinction. He also assumes (173 n. 200) that Strabo independently corroborates Ephorus, and implies that Diodorus does as well.

⁸⁰ See most recently G. L. Cawkwell's introduction to the 1978 reprint of R. Warner's translation of the Hellenika in the Penguin classics series: Xenophon, A History of My Times (Harmondsworth).

⁸¹ On this phenomenon, see recently Snodgrass, A. M., Archaic Greece (London, 1980) 31.Google Scholar On the role of Ptolis, Moggi 148–9.

⁸² Bölte 1339, 1344 dates the cult of Maira to an early period.

⁸³ On the Mantineian population, see Section V below. For the suggested density of 150 persons per ha of built up area, M. H. Jameson, op. cit. (n. 18) 85. We assume an average household size of 4–5 without taking slaves into account.

⁸⁴ Howell, Appendix I, Table of Sites, No. 11. 16.

⁸⁵ Since the town is mostly mentioned during narratives of military campaigns, one should beware that some references to the town's inhabitants in such situations may be atypical, e.g. if extra troops from families living in the villages had been concentrated inside the town.

⁸⁶ The following ancient quarries are known to us: (1) at the N.W. edge of the plain of Mantineia ‘near three large katavothrai’ (Pritchett 51 n. 51). These are presumably the ones named Kapsia, Palaiochori and Karaphotia indicated in Pritchett 40 fig. 8; (2) at the W. edge of Mt. Barberi near the hamlet of Milia (Steinhauer, pers. comm.).

⁸⁷ Plb ii. 62. 12. Contrast his estimate that Kleomenes III of Sparta could get up to 300 T out of Megalopolis at a time when most of the free and slave populations had escaped to Messene. Wilhelm, , JOEAI xvii (1914) 108–16Google Scholar, estimated that of the 300 T obtained from Mantineia, the moveable property may have realized only 75 T; cf. Walbank op. cit. (n. 16) i 270.

⁸⁸ SHA Ch. 3. On medicinal herbs, ibid. 69–71, with refs. to the favourable comments of ancient writers, especially Theophrastos, who mentions inter alios a certain Thrasyos, an apothecary from Mantineia.

⁸⁹ Data on the climate of the Mantinike has been obtained from Admiralty, Naval Intelligence Division, Geographical Handbooks, Greece (London, 1944)Google Scholar; Nuttonson, M. Y., Ecological Crop Geography of Greece (Washington, D.C., 1947)Google Scholar; United Nations, Economic Survey of the Western Peloponnesus, Greece. F.A.O. and U.N. Special Fund (Rome, 1965)Google Scholar; Mariolopoulos, E. G., Étude sur le climat de la Grèce (Paris, 1925).Google Scholar This data comes from the meteorological station at Tripolis about 6 km S. of the Mantinike and, at an altitude of 661 m, about 30 m higher than the plain of Mantineia. Similarity of ancient and modern climates: Bintliff, John L., Natural Environment and Human Settlement in Prehistoric Greece. British Archaeological Reports, Supplementary Series xxviii (1977) i 51Google Scholar; Mariolopoulos, ibid. 39–59.

⁹⁰ Annual mean temperatures: Tripolis 57·5 °F; compare Ioannina 57·9 °F, and contrast Athens 63·3 °F and Nauplion 64·5 °F; Admiralty, op cit. iii 481 (Table 3); see also Nuttonson, op. cit. Agroclimatological Chart; United Nations, op. cit. ii 10 and 12 (Table 1). Frosts: United Nations, op. cit. iii 7; Admiralty, op. cit. ii 484 (Table 5): the frequency of 23·3% days of frost between November and April works out at about 42 days. Agroclimatic classifications: United Nations, ibid, iii 5–7 and Map 7.

⁹¹ The mean annual rainfall within the hydrological catchment areas of the different valleys are Nestane 964 mm; Louka 914 mm; Mantineia (figure for the Tripolis catchment area as a whole) 952 mm. Data on rainfall from United Nations, op. cit. ii 8–10, 15–17, 20 (Table 4) and Maps 6 and 8; Admiralty, op. cit. ii 485 (Table 6); cf. i 104; Nuttonson, op. cit. Agroclimatological Chart; Mariolopoulos, op. cit. (n. 89) 18–19 (Tables I–II).

⁹² Admiralty, op. cit. iii. 173.

⁹³ On the geology and hydrology, Philippson, A., Die griechischen Landschaften (Frankfurt am Main, 1959) iii 1. 245–59Google Scholar; Burdon, D. J. and Papakis, N., Greece, Karst Groundwater Investigations, F.A.O., Land and Water Development Division, No. 54607 (Rome, 1964) 46–66Google Scholar; United Nations, op. cit. (n. 89) ii 4–8; iv² 108; Fougères, MAO 13–52; Pritchett 38–42, with a convenient map (40 fig. 8) showing the katavothras; Siderides, N.-A., ‘Les katavothres de Grèce’, Spelunca viii (1911) 201–74Google Scholar; Martel, E.-A., Les Abĭmes (Paris, 1894) ch. XXVIII, esp. 509–12.Google Scholar

⁹⁴ See the discussion of Pritchett 41–4.

⁹⁵ The chronology of Diodorus (xv 12. 1) is preferable to the temporally vague account of Xenophon, (HG v 2. 4–5).Google Scholar

⁹⁶ Pritchett 54–6.

⁹⁷ Pausanias viii 6. 4–5; Fougères, MAO 84–5.

⁹⁸ For the date, Gomme, A. W., Andrewes, A., and Dover, K. J., A Historical Commentary on Thucydides iv (Oxford, 1970) 98.Google Scholar Note, however, their remark that the diversion of water obviously did not cause immediate flooding. It did ‘not compel the Mantineians to fight on the scene of the waterworks but to fight somewhere to enable them to reverse what had been done’.

⁹⁹ Pritchett 42; cf. also his Studies in Ancient Greek Topography i, California Classical Studies i (1965) 13, referring to a change in the water table.

¹⁰⁰ Information from Mr. Christos Theophanoyiannis, a farmer from the village of Simiades, who claims that in earlier days (probably those of his great-grandfather) it was impossible to see the church of Ay. Prokopios in the centre of the plain from the valley edge by the Kapsia katavothra even when on horseback. As a result of the accumulation of silt the church is now visible from the ground. The story is supported by the evidence of organic remains of dead marsh vegetation found amidst the very silty soil of this area.

¹⁰¹ Burdon and Papakis, op. cit. (n. 93) 47.

¹⁰² The foundations both of the fourth-century funerary monument 200–250 m S. of gate G and of the sanctuary of Poseidon Hippios were visible in thesurface of the ground when discovered by Fougères (refs. in Appendix, Site 3 and n. 174). The late Roman tombs recently discovered at Milia were almost at ground level (Th. Spyropoulos, pers. comm.; Appendix, Site 3). The remains of the ‘villa’ site in the N.W. of the plain (Section II, d) lie no more than a few cm beneath ground level. Those of Howell's site S.W. of Artemision (Section II, c) stand above the ground. In the last two cases excavation is needed to ascertain how deep the foundations are buried. The soil level within the town walls is higher than that outside, yet the depth of ancient remains was only 1·5 m: Fougères, , BCH xiv (1890) 246.Google Scholar The Roman buildings (Sector A) excavated by Steinhauer, op. cit. (n. 58) were only 0·20–0·30 m below the surface.

¹⁰³ Bintliff, op. cit. (n. 89) esp. 35–58; cf. Vita-Finzi, C., The Mediterranean Valleys (Cambridge, 1969)Google Scholar; ‘Heredity and Environment in Clastic Sediments: Silt/Clay Depletion’, Bull. Geol. Soc. Am. lxxxii (1971) 187–90; ‘Related Territories and Alluvial Sediments’, Appendix B 225–31 in Higgs, E. S. (ed.), Palaeoeconomy (Cambridge, 1975).Google Scholar

¹⁰⁴ Forbes, H. A., Koster, H. A., Foxhall, L., ‘Terrace Agriculture and Erosion: Environmental Effects of Population Instability in the Mediterranean’, unpublished MS.Google Scholar

¹⁰⁵ Halstead, P., ‘Counting sheep in Neolithic and Bronze Age Greece’, in eds. Hodder, I., Isaac, G., Hammond, N., Pattern of the Past: Essays in honour of David Clarke (Cambridge, forthcoming).Google Scholar

¹⁰⁶ H. A. Forbes, H. A. Koster, ‘Fire, Ax and Plow: Human Influence on Local Plant Communities in the Southern Argolid’, in M. Dimen and E. Friedl (edd.), op. cit. (n. 18) 109–26.

¹⁰⁷ Résultats du recensement de l'agriculture—élevage effectué le 19 mars 1961. Office National de Statistique de Grèce (Athens, 1966), in Greek and French; Results of the Agriculture—Livestock Census of March 14, 1971. National Statistical Service of Greece (Athens, 1978), in Greek and English. The figures in the text are aggregates ofthe figures given for the individual Koinotes of Artemision, Kapsia, Louka, Nestane, Pikerni, Sanga, Simiades, and Skopi. A majority (at least) of the farmers in each community possess holdings within the territory of ancient Mantineia. Since the censuses classify holdings of land according to the residence of their ‘head or owner’ rather than the location of the land itself, the figures probably include certain areas outside Mantineian territory cultivated by farmers from those communities and exclude certain areas inside Mantineian territory whose ‘head or owner’ resides outside the area covered by the above Koinotes, e.g. in the near-by town of Tripolis.

¹⁰⁸ Fougères, MAO 58–9; personal observation.

¹⁰⁹ e.g. Fougères, MAO 55; Frazer 201.

¹¹⁰ Fougères, MAO 55; Admiralty, op. cit. (n. 89) 173; personal observation.

¹¹¹ Fallow land: 581·7 ha in 1961; 953·1 ha in 1971. Grassland: 245·8 ha in 1971 (no comparable figure for 1961).

¹¹² Information from the papas (priest) at Artemision; personal observation.

¹¹³ In our discussion of this topic the limited nature of the information forces us to simplify and to schematize in two ways: by presenting a static picture of the Mantineian population as it was in the fifth and early fourth centuries and by grouping its different social strata into imprecise, broad categories. We are conscious of the deficiencies in our analysis owing to the impossibility of charting both the processes of change and the gradations of wealth and status among and between the different social categories.

¹¹⁴ IG v. 2. 47(i); Hansen, P. A., A List of Greek Verse Inscriptions down to 400 B.C. (Copenhagen, 1975) No. 398.Google Scholar The date of the dedicationis uncertain; its terminus ante quern is c. 465. Parke, H. W., Greek Mercenary Soldiers (Oxford, 1932) 11Google Scholar suggests that Praxiteles served as mercenary commander under the tyrant Gelon.

¹¹⁵ Moretti, L., ‘Olympionikai, i vincitori negli antichi agoni olimpici’, Atti della Accademia Nazionale dei Lincei, 8th Ser., Classe di Scienze morali, storiche e filologiche, Memorie viii (1959) 53–198 Nos. 163, 193, 202, 254, 256, 265.Google Scholar Several patterns in these victories are noteworthy. They were concentrated between the years 500 and 460 (Moretti's dates) and the next known Mantineian victory was not until 296 B.C. (Moretti No. 514). They were all in athletic contests as opposed to chariot racing and four of them were in boxing. With the exception of Dromeus (no. 202) all the victors, including all four boxers, were boys. The late fifth-century lawgiver, Nikodoros, was also a boxer in his youth (Aelian, , Varia Historia ii 22Google Scholar) and a man with foreign connections.

¹¹⁶ X. HG v 2. 3; vi 5.4; Tuplin, C. J., ‘Kyniskos of Mantinea’, LCM ii (1977) 5–10.Google Scholar

¹¹⁷ IG v 2. 323 No. 3 Zakynthios Xanthiau; No. 7 Pantinas Eretriadau; No. 19 Samios -ereto; No. 84 Praxin[os] Seliniō. Hiller von Gaertringen dated the first three to the period 425–385, the last before 226 B.C. Fora recent discussion of the tesserae, Amit 141–7.

¹¹⁸ On the political structure of Mantineian democracy, O'Neil, , GDCOA 42–59, esp. 52–4.Google Scholar See further Section VI.

¹¹⁹ See the brief discussion with references to ancient sources in Walbank, op. cit. (n. 16) ii. 724.

¹²⁰ Beloch, K. J., Griechische Geschichte 2nd ed. (Berlin and Leipzig, 1922–1931) iii I². 280.Google Scholar

¹²¹ Bölte, 1308–9, who summarizes the population estimates of earlier scholars.

¹²² Fougères, MAO 568–72.

¹²³ Having completed the above argument, Fougères then (569–70) unaccountably introduces the arbitrary estimate that able-bodied males between ages 20 and 60 constituted one-sixth of the total Mantineian population, which was therefore 18,000. He proceeds (571) to say that 9,000 of those will have been males, which, deducting children, would mean between 5,000 and 6,000 citizens. He seems unaware of the contradiction between this figure and his previous one of 3,000 able-bodied males between 20 and 60. One can hardly imagine that unfit males between those ages together with old men over 60 years will have numbered 2,000–3,000 citizens! We therefore ignore this higher figure in the text in favour of Fougères's original proposition.

¹²⁴ Die Bevölkerung der griechisch-römischen Welt (Leipzig, 1886) 125–7.

¹²⁵ Th. v 57–73. Only the Boiotian infantry is specifically mentioned as a mixed hoplite and light-armed force (v 57).

¹²⁶ Gomme–Andrewes–Dover, op. cit. (n. 98) 50; cf. Andrewes at 50–1, 106 on the possibility of an alternative source.

¹²⁷ Cooper, A. Burford, ‘The Family Farm in Greece’, CJ lxxiii (1977/1978) 162–75, esp. 168–72.Google Scholar The exact figures expressed in the ancient Greek measurements are, respectively, 200–300 plethra, 40–60 pl., 10 and 20 pl.

¹²⁸ we know of no previous attempt at an estimate of cultivable land. Beloch and Fougères have estimated the total surface area of Mantineian territory at 275 and 325 sq. km respectively (refs. in Bölte 1308). These are overestimates based upon the assumption that Mantineian territory included parts of W. Mainalia. (Against this, see Section II above.) A more realistic figure can be derived from the combined surface area of the individual Koinotes listed in n. 107 (but omitting Skopi whose area lies largely outside Mantineian territory) which is 207 sq. km (20,700 ha): Résultats du recensement de la population effectué le 19 mars 1961, Office National de Statistique de Grèce (Athens, 1964–8). Of this, however, only 6897·7 and 6250·6 ha were recorded as farmland in the agricultural censuses of 1961 and 1971 respectively. The fact that these figures (which include land of farmers from Skopi many of whom own plots within the Mantinike) fall well below our maximum figure of 9,000 ha suggests that we have not underestimated the potential area of cultivation.

¹²⁹ This conclusion would not be altered even if one used the figure at the lowest end of the range of hoplite estate size, 3·6 ha, as the basis of the calculations. The amounts of land required would then still range between 11,232 and 14,400 ha. The conclusion would also remain even if one included (as we would not) the valley of Kapsia within Mantineian territory. The valley contains no more than about 150 ha of poor quality alluvial soil. The remainder of the valley, totalling over-all only about 1,100 ha, consists mainly of barren, steeply sloping conglomerate.

¹³⁰ Although they interpret Lysias' statement differently, in practice their figures coincide, with the exception of their estimates of the over-all citizen body. Fougères' estimate has already been set aside as being in contradiction with his military estimates. Beloch's estimate is also inconsistent with his figure for the hoplite levy, implying that the hoplites formed only 35·7% of the citizen body.

¹³¹ The purpose is to make these figures comparable with the other estimates, which concentrate upon citizens eligible for military service. For simplicity we assume throughout that the number of men over 60 in possession of estates but lacking descendants of military age was negligible.

¹³² Quotation from Jameson, M. H., ‘Agriculture and Slavery in Classical Athens’, CJ lxxiii (1977/1978) 122–45, at 131Google Scholar, who reports the calculations of K. Hopkins that a family of five required 4 ha cropped each year in wheat.

¹³³ Parke, op. cit. (n. 114) via Index, s.v. Arcadians; Griffith, G. T., The Mercenaries of the Hellenistic World (Cambridge, 1935) 237–8Google Scholar; Roy, SHA 97–103 also collects the evidence for other forms of emigration.

¹³⁴ For the numbers of Arkadians among the Ten Thousand, Roy, J. ‘The Mercenaries of Cyrus’, Historia xvi (1967) 287–323, at 308–9.Google Scholar On the fallacy of Xenophon's statement (An. vi 4. 8) that most of the soldiers had not joined up through poverty, see G. L. Cawkwell's Introduction to the Penguin translation of the Anabasis: Xenophon, , The Persian Expedition (Harmondsworth, 1972) 43–4.Google Scholar

¹³⁵ Note the legend in Pindar (Olympian x 69–70) of the four horse chariot victory of Samos the Mantineian at the first Olympic games; cf. also the scholia to this passage, in Drachmann, A. B. (ed.), Scholia Vetera in Pindari Carmina (Leipzig, 1893) i 331–3.Google Scholar The sanctuary of Poseidon Hippios was one of Mantineia's most important cults. Poseidon's trident and dolphin sometimes appeared on Mantineian coinage in the fifth century: Babelon, op. cit. (n. 67) ii 1 nos. 1237, 1238; ii 3 nos. 953 (possibly identical with ii 1 no. 1237), 967. The trident was also the emblem on Mantineian shields (Bacchylides fr. 21 Snell; scholiast on Pindar, Olympian 83, in Drachmann (ed.) i 331.

On the fame of Arkadian horses in the Roman period, see the refs. in Fougères, MAO 59 n. 1, with discussion at 59–61; also Rigopoulos, op. cit. (n. 5) 49. Mantineia had a racecourse in Pausanias' day (viii 10. 1).

¹³⁶ Cf. n. 60 above. The translation is that of Buck 198–9, with correction of his inaccurate rendering of weikiatai as ‘serfs’.

¹³⁷ Cf. the evidence in Roman times of vineyards of 6 plethra and, if H. von Gaertringen's restoration is correct, of 14 plethra (just over ½ and 1¼ ha respectively): IG v 2. 269; 270. In the medieval period Eustathius (Comm. ad Iliadem 302. 3) called Mantineia ‘many-vined’.

¹³⁸ Cf. also the evidence of Archaic and Classical Arkadian bronze statuettes depicting shepherds: Lamb, W., ‘Arkadian Bronze Statuettes’, BSA xxvii (1925/1926) 133–48Google Scholar, who suggests that their centre of production may have been Tegea; Jost, M., ‘Statuettes de bronze provenant de Lykosoura’, BCH xcix (1975) 339–64. at 339–45.CrossRefGoogle Scholar

¹³⁹ Grazing on the stubble was recommended by Varro (RR ii 2. 12), referring to Epirus. It is practised generally in modern Greece; cf. Wagstaff, J. M., ‘Aspects of Land Use on Melos’ (unpublished report, Department of Geography, University of Southampton, 1976) 31.Google Scholar Bjørn Quiller has pointed out to us that the remainder of the cereal crop after removal of the grain will have been of greater food value in both quantity and quality than today. The stalks, shortened by modern breeding, will have been longer and, in the absence of weed-killers, mixed with more weeds whose seeds will have provided extra nutrition. Varro mentions an additional source of nutrition, those ears of grain which had fallen upon the ground.

¹⁴⁰ Cf. Sophocles, Oedipus Rex 1121–41; IG ii² 7180; Robert, L., ‘Epitaphe d'un berger à Thasos’, Hellenica vii (Paris, 1949) 152–60.Google Scholar

¹⁴¹ Roy, SHA 93–5.

¹⁴² Our posing of this question implies rejection of the hard and fast distance of 5 km (or a walking-time of 1 hour) from the settlement claimed in recent studies as the limit for cultivation under systems of subsistence farming; cf. Higgs, E. S. and Vita-Finzi, C., ‘Prehistoric Economy in the Mount Carmel Region of Palestine: Site Catchment Analysis’, Proceedings of the Prehistoric Society xxxvi (1970) 1–37Google Scholar; ‘Prehistoric Economies: a Territorial Approach’, in Higgs, E. S. (ed.), Papers in Economic Prehistory (Cambridge, 1972) 27–36Google Scholar; cf. Chisholm, M., Rural Settlment and Land Use (London, 1962).Google Scholar The appropriateness of this figure for Greece has recently been questioned by Wagstaff, op. cit. (n. 139), who concludes (24) that ‘the application of catchment analysis in Greece requires the use of greater critical distances’. He emphasizes that ‘critical limits should be derived empirically for each region under examination and ideally for each level of technology and different form of economic organisation’.

¹⁴³ Hopkins, op. cit. (Table I, note b) 24–5.

¹⁴⁴ Chayanov, A. V., The Theory of the Peasant Economy, edd. Thorner, D., Kerblay, B. and Smith, R. E. F. (Homewood, Ill. 1966) 56–60Google Scholar; Franklin, S. H., The European Peasantry: the Final Phase (London, 1969) 18–20.Google Scholar

¹⁴⁵ Pepelasis, A. A. and Yotopoulos, P. A., Surplus Labour in Greek Agriculture 1953–1960. Center of Economic Research, Research Monographs Senes ii (Athens, 1962)Google Scholar to which our following discussion is indebted. The variability of seasonal labour demands is evident in Hesiod, Op. 414–617; cf. West, M. L., Hesiod, Works and Days (Oxford, 1978) 52–5.Google Scholar

¹⁴⁶ Jameson op. cit. (n. 132), esp. 126–31.

¹⁴⁷ The fact that the plain of Nestane was untilled in Pausanias' day when rural habitation, including the village of Nestane, was abandoned (cf. n. 20 above), and the population was probably lower, is unreliable evidence for our period, which had continuous village habitation and a high population density. Even today, when the Mantinike, including the modern village of Nestane, is suffering from depopulation, some parts of the plain are sown with winter cereals, despite drainage problems, and others are used for summer-grown maize and tomatoes. There is a hint of cultivation in the plain in Xenophon's account of Agesilaos' invasion of the Mantinike in winter 370/69 (X. HG vi 5. 15–20). The fact that Agesilaos led his army into the plain of Nestane during one of the days that he spent ravaging the countryside suggests that it was fertile; cf. Loring, op. cit. (n. 3) 86.

¹⁴⁸ For the reputation of the Mantineian turnip, Athenaeus i 4c; Pollux vi 63.

¹⁴⁹ H. A. Forbes, ‘“We have a little of Everything”’: the Ecological Basis of Some Agricultural Practices in Methana, Trizinia’, in Dimen and Friedl (edd.), op. cit. (n. 18) 236–50.

¹⁵⁰ Examples in Babelon, op. cit. (n. 67).

¹⁵¹ Our following remarks on Mantineian democracy owe a great deal to O'Neil, GDCOA 42–59; and to his forthcoming article (cited in n. 70).

¹⁵² See e.g. Lézine, A., ‘Sur la population des villes africaines’, Ant. Afr. iii (1969) 69–82, esp. 82CrossRefGoogle Scholar; Jameson, op. cit. (n. 18) 85. On the proportion of town areas taken up by public sectors, Lézine passim.

¹⁵³ IG v 2. 36 = Tod no. 202 = Buck 206–9 no. 22. The date of the inscription is 324 B.C. The gardens were probably in the town, although the inscription does not specify.

¹⁵⁴ R. Martin, op. cit. (n. 20) 109–10.

¹⁵⁵ E.g. Cromna, Tenea, and Asae, at distances of c. 6·5, 10·6, and 5·5 km respectively from the town of Corinth: Wiseman, J., The Land of the Ancient Corinthians, Studies in Mediterranean Archaeology I (1978) 66–70, 92–3, 101.Google Scholar

¹⁵⁶ Amit 121–4, 173–4.

¹⁵⁷ On Sparta, see Cartledge, P., Sparta and Lakonia (London, 1979) ch. 8.CrossRefGoogle Scholar On Corinth, Roebuck, C., ‘Some Aspects of Urbanization in Corinth’, Hesperia xli (1972) 96–127.CrossRefGoogle Scholar On Heraia, , RE viii 1 (1912) 407–18Google Scholar; Suppl. ix (1962) 70. She coined in the Archaic period long before her probably early fourth-century synoecism; cf. n. 66, above; Williams, R. T., ‘The Archaic Coinage of Arcadian Heraea’, Museum Notes xvi (1970) 1 ff.Google Scholar; Kraay, op. cit. (n. 67) 96, 99. Cf. also the inscription (IvO 9 = Tod 5 = Meiggs–Lewis 17) recording their alliance with Elis which is dated c. 500 by Jeffery, op. cit. (n. 60) 220 no. 6.

¹⁵⁸ On ethnics, Bölte 1291 (but the Olympic victors are better dated to the early fifth century). On coinage, n. 67 above.

¹⁵⁹ Paus, viii 8. 4. According to the myth the town was subsequently founded by Antinoē, daughter of Kepheus and granddaughter of Aleos (cf. the shrine of Athena Alea in the town).

¹⁶⁰ Strabo's precise wording suggests that each state had ‘groupings of demes’. In common with most commentators we assume this to be some sort of error and take it that he meant, or should have meant, that each state had a grouping (singular) of demes. For a possible interpretation of the role of the demes, Baladie op. cit. (n. 79). From the mid fourth century at least the population appears to have been grouped into five phylai, according to the evidence of IG v 2. 271, which gives their names. Cf. also the inscription bearing the name Posoidaia, recently discovered by Pritchett (50–2). The five damiourgoi in IG v 2. 1 (variously dated to the fourth or third century, but perhaps best located sometime in the 360s: von Gaertringen, H., AM xxxvi (1911) 349–60Google Scholar; Cary, M., JHS xlii (1920) 188–90)Google Scholar and the two groups of five officials in IG v 2. 278 (probably from the fourth century) may be connected with the phylai system too. Unfortunately, we do not know when the phylai were instituted. Pritchett claimed the date 418 for the Posoidaia inscription, but this has been disputed by Solin, H., ‘Bemerkungen zu einer mantineischen Namenliste’, ZPE xiv (1974) 270–6Google Scholar, who would locate it in the mid fourth century along with IG v 2. 271. Equally we do not know what relationship the phylai might have had to the demes, if the latter existed. On this question, Bölte 1318–19.

¹⁶¹ Op. cit. (n. 157) 97.

¹⁶² Cf. Hopkins, K., ‘Economic Growth and towns in Classical Antiquity’, in Abrams, P. and Wrigley, E. A. (edd.), Towns in Societies (Cambridge, 1978) 35–77, at 68.Google Scholar

¹⁶³ Th. iv 134; v 29. 1, 33, 54. 1, 58. 1, 81. 1; Amit 147–63; Roy, SHA 183.

¹⁶⁴ On the history of the Arkadian league, Larsen, J. A. O., Greek Federal States (Oxford, 1968) 180–95.Google Scholar

¹⁶⁵ On the doubt concerning the date of this episode, Cawkwell, op. cit. (n. 80) note at 212–13. We omit the events in 385 from the text because of their uncertainty. Xenophon, (HG v 2. 4)Google Scholar and Diodorus (xv 5. 5) imply that there was no Mantineian resistance before the siege, but Plutarch (Pel. iv) and Pausanias (viii 8. 6) both mention a battle.

¹⁶⁶ Y. Garlan, ‘La défense du territoire à l'époque classique’, in Finley (ed.), op. cit. (n. 20) 149–60, at 157.

¹⁶⁷ Forbes, Koster, Foxhall, op. cit. (n. 104).

¹⁶⁸ P. Abrams, ‘Towns and Economic Growth: Some Theories and Problems’, in P. Abrams and E. A. Wrigley (edd.), op. cit. (n. 162) 9–33, at 19.

¹⁶⁹ Capitalism and Material Life, 1400–1800 (London, 1967) 439–40.

¹⁷⁰ Lattermann, op. cit. (n. 10) 401–2.

¹⁷¹ Kurtz, D. C. and Boardman, J., Greek Burial Customs (London, 1973) 189–90.Google Scholar

¹⁷² Cf. the findings of Hägg, R., Die Gräber der Argolis in submykenischer, protogeometrischer und geometrischer zeit (Boreas vii 1) Uppsala, 1974.Google Scholar

¹⁷³ Professor A. M. Snodgrass, pers. comm.; Kurtz-Boardman, op. cit. (n. 171) part i.

¹⁷⁴ Pritchett's site cannot therefore be that of the sanctuary of Poseidon Hippios, as he suggests, since according to Polybius (ix 8. 11; xi 11. 6) Poseidon Hippios was 7 stades from Mantineia. It has been located by Fougères about 1200 m from the walls: BCH xiv (1890) 80–2; MAO 103–6. Our relocation of Pritchett's site is confirmed by the fact that the dovecote and abandoned house lie N. of a broad track which runs E.-W. from the S.W. slope of Mt. Barberi to the Artemision—Tripolis road. The track is only 500 m S. of Mantineia town. On Pritchett's Plate 41, a photograph of the dovecote and house (the two buildings near the centre of the picture) taken from the slope of Mt. Barberi, looking approximately W.S.W., part of the track is visible well to the left (i.e. S.) of those buildings. The track is also visible on Pritchett's plate 21, an aerial view of the valley, 8 mm S. of the walls of Mantineia town.

¹⁷⁵ Of the Classical inscriptions Pritchett gave the third monumenta probable fourth century date. The second, a partly preserved list of names under the heading of one of the Mantineian tribes, he ascribed to 418 B.C., suggesting that it was a monument to soldiers who died in the battle of Mantineia in that year. Solin, op. cit. (n. 160) has questioned this interpretation and argues from letter forms and dialect for a mid fourth century date.

¹⁷⁶ This information comes from a report dated 8 Dec. 1979. The discoveries are also described briefly in JHS c (1980) AR (1979/8) 33 which we received at the time of writing this article in Nov. 1980. The description derives from an announcement in the Greek press (Akropolis 14. 12. 79) which we were unaware of and have not been able to consult. Apparently the press announcement suggests that the funerary offerings date from late Classical to Roman times. According to our information, however, some date to the late Classical and others to the late Roman period (i.e. none are datable to intervening years). It also refers to the discovery of abronze vase decorated in relief which our information does not appear to mention.

¹⁷⁷ Glotta lvi (1978) 202–5 and 206–12.