^{page 115 note 1} CQ xxxvii, pp. 62 ff., and lxiii, pp. 1 ff. I refer to these by page alone. This third part has benefited by advice from A. Momigliano and H. W. Parke.

^{page 115 note 2} Parke demurs to my phrase ‘ancient oracle’, here and above (p. 2, n. 1, p. 4). Did Tyrtaios really offer it as an ancient oracle? Aristotle was bound to think so, so that the poem was susceptible of being so understood: but might it not be intended otherwise? Parke thinks that if Delphic authority was sought it would be sought at Delphi and not in the archives: Tyrtaios may have made 'Hρακλεῖδαι the subject of ἔνεικαν and yet have meant his own contemporaries. [Xenophon understood that Lykourgos went with company (supra, p. 2, n.4): Λακ. π 8.5.]

^{page 116 note 1} Meyer includes the Gerousia in his list, but the text I have sought to establish is quite specific on this.

^{page 116 note 2} p. 265 ‘nach allgemeiner Annahme das Grundgesetz des spartanischen Staates’.

^{page 117 note 1} ATL, p. 161. The restorations are not certain, but there is certainly mention of some marble work and the Golden Nikai.

^{page 117 note 2} Not from ɸνλάσσω.

^{page 117 note 3} IG v. 1, No. 564, lines 4–5 Λιμναεων ɸν|λης: No. 688 ωβα λιμναιε[ν]. So 480, 9–10 ɸνλης κονοον|ρεων, and probably της ɸ [νλης][ in 682, 4. They had long performed the functionof ɸνλαί: the Three Tribes were quite as much forgotten as the Four Tribes in Athens.

^{page 118 note 1} In 5. 66. 3 the hierarchy is King, Polemarchs, Lochagoi, Penlekonteres, Enomotarchoi: but in 68. 3 he takes count only of Lochoi, Pentekostyes, Enomotiai: sc. he leaves out the Mora (Λακ.π 11. 4). There were polemarchs present, and Thucydides speaks of them as commanding each one Lochos (71. 3): there is no unit left for the Lochagos. It is evident that the polemarch's command was in fact a Mora, and Thucydides' ‘Lochoi’ are in fact Morai.

^{page 118 note 2} ‘Seven without the Skiritai’ (68. 3): the seventh is the ex-Helots who stood between the Skiritai and the other Lakedaimonioi (67. 1). The Ηακεδαιμὀνιοι αύτοι, distinguished from the Skiritai and ex-Helots (67. 1), are the ‘regular establishment’ of the six Morai.

^{page 118 note 3} Isokrates, 12. 180 κατ' ἄνδα σνμπαρατάττεσθαι οɸιοις (what he says next, άποστέλλειν προκινδννενοοντας, will refer to the Skiritai, Xen. Λακ. π. 12. 3, 13. 6). Aristotle, fr. 540, διὴρηνται εις τίς ιακεδαιμὁντες πάάντες includes the perioikoi, cf. Hdt. 7. 234. 2; Xen. Hell. 6. 4. 15. Aristotle is no doubt making the contrast with the Obal Lochoi of fr. 541. The order of these two fragments should be inverted: Hesychios, s.v. μὁρα: … οί πίτρι<οι> λὀϰοι μὀρα αὐθις ὀνομασθέντες must come ultimately from Aristotle and gives the sequence. [Hesychios takes μὀρα as neuter plural.] Prisoners from the Morai 7. 4. 20, 27.

^{page 119 note 4} Hdt. 9. 12. 2 πέμποναι κήρνκα … ές τήν' Αττικὴν, πρότερον … ὑποδεξάμενοι σχήσειν τὸν Σπαρτιήτην μὴ ἐξιέναι· ὂς … ἔλεγε τάδε· Mαρδόνιε … ἐξελήλυθε ἡ νεότης. Should this be pressed? Herodotos is often found to speak the exact truth because he repeats what he hears with all its quiddity and flavour and does not reduce it to commonplace. Here I suspect that his language reveals two facts: (a) that the normal establishement at that time was Spartiates only; (b) that only young men were normally called out. For the relation between this νεότης of 5,000, and the 8,000 ἄνδρες of Hdt. 7. 234. 2, see Appendix III.

^{page 119 note 1} The problem of the Eirenes is still unsolved, so that I will not argue from the alleged fact that these youngest soldiers (?) had many casualties and a separate grave: Hdt. 9. 85. 2. Nilsson's interpretation of the Herodotos, gloss (Klio, xii. 309 ff.Google Scholar) is no doubt wrong: see the fuller version of this gloss published by Diller, , AJP lxii. 500Google Scholar. Yet the fact remains that Xenophon distinguishes εἴρενες from ἄνδρες (Λακ. π. 2. 11) while Herodotos appears to say that the commander of a Lochos was an Eiren. The gloss now appears to say [ἐπ' ἐτῶν should be ἀπ' ἐτῶν?] that you were an ephebos from 14 to 20, and became an Eiren at 20: agreeing with Plut, . Lye. 17. 3–4Google Scholar.

^{page 119 note 2} Nestor's, rule ‘Kin will help kin’, Iliad, 2. 362–3Google Scholar. Later, neighbour helped neighbour: the Obal Army, the Kleisthenic army at Athens. Later still, friend (or ‘lover’) helped friend: the Lochos of Episthenes (Xen, . Anab. 7. 4. 8Google Scholar), and the more famous ἱερὰς λόχος (or ἐκ πόλεως λόχος, because their barracks were on the Akropolis) founded by Gorgidas at Thebes, Plut, . Pelop. 18Google Scholar. Pammenes is quoted ibid. as saying that this improved on Nestor.

^{page 119 note 3} I suggest some details in Appendix III.

^{page 119 note 4} Wilamowitz, , SB Berlin, 1918, 728 ff.Google Scholar: now Tyrtaios, fr. 1, in Diehl's Anth. Lyrica.

^{page 120 note 1} Sch, . Pind, . Pyth. 1. 121aGoogle Scholar Παμπυλὶς καὶ Δυμανὶς πυλαὶ ἐν Γακεδαίμονι: Hesych. Δύμη· ἐν Σπἀρτῃ πυλή· τόπος (conflated from Δυμήνἐς Σπ. π. and Δύμη· τόπος?). Such passages were enough to suggest that Sparta had once had the Three Tribes, but not that her army had been so composed in historic times. Pareti came very near the truth (see his St. di Sp. arc. 173), but thought the Obai much older than the seventh century.

^{page 120 note 2} Οἱ Λιμνᾶται Σπαρτιατῶν καὶ Κυνοσονρεῖς καὶ <οἱ> ἐκ Mεσόας τε καὶ Πιτάνης θύοντες τῇ' Bρτέμιδι ἐς διαποράν ἀπὸ δὲ αὐτῆς καὶ ἐς πόνους προήχθησαν (3. 16. 9). The contending parties are Limnaicum-Kynosoura versus Mesoa-cum-Pitana, eastern versus western Sparta, Eurypontid versus Agiad. It was before Lykourgos (Λυκοῦργος μετέβαλεν, 3. 16. 10), so we do not expect Neopolitai (aliter Pareti, , op. cit., 175, n. 4Google Scholar): nor Amyklans if (as I think) it recalls a real dichotomy, before the two houses united: the Agiad attempts access to the Eurypontid's altar.

^{page 120 note 3} IG v. 1, Nos. 674–87.

^{page 120 note 4} BSA xxvi, pp. 165, 181. Cf. IG v. 1, No. 566, line 3 σπαιρεα Κονοουρεα.

^{page 120 note 5} My supplement is uncertain, but the sitesis is certainly part of τα τας ωβ[α]ς τ[ιμια] (line 18).

^{page 120 note 6} Synoikisis and πρυτανεῖον, Th. 2. 15. 2.

^{page 120 note 7} I do not see that it follows from Xen, . Hell. 4. 5. 10Google Scholar (Bölte, in P-W, ‘Sparta’, p. 1329, 23Google Scholar). That passage makes it likely, its being an ὠβά makes it likely, its nearness to Sparta and its importance combine perhaps to make it almost certain. [Note, too, perhaps, the statue of Sparte at Amyklai and of the Demos of Spartiatai at Sparta: Paus. 3. 18. 8 (cf. 2. 16. 4, and Wide, , Ldk. Kulte, 337, n. 1Google Scholar), 3. 11. 10.] My Obal thesis depends on Amyklai being Spartiate and I have no moral doubt of it, but I do not see that it can be proved αὐτὸ καθ' αὑτό. Every attempted proof begs the question of what exactly the differentia was which distinguished perioikoi from Spartiates.—Nor is it certain that the Nεοπολῖται are of Hellenistic date (again I have no moral doubt: cf. Pareti, , op. cit. 176–7Google Scholar). I cannot disprove that the classical Five Obai were the known four plus the Neopolitai, in fact the ball-game teams. The Neapolitai would then date from the seventh century, and when Aristotle says (Ἀθ. π. 21. 4) that Kleisthenes assimilated his νεοπολῖται, he could be understood as contrasting this with the Spartan way.— Against the view that there were only 4 Obai and 4 Obal Lochoi, Aristotle's five names are surely conclusive: nor can Ar, . Lys. 453Google Scholar be cited in support of it. The Proboulos orders the assault; Lysistrata says νὴ τὼ θεὼ γνώσεσθ' ἄρα ὄτι καὶ παρ' ἡμῖν εἰσι τέτταρες λόχοι. There can be no reference to the Obal Army: in 411 ‘four Lochoi’ was one-third of the establishment. I suggest there were 4 Lochoi in Dekeleia (soon reinforced, Th. 8. 71. 1). importance combine perhaps to make it almost certain. [Note, too, perhaps, the statue of Sparte at Amyklai and of the Demos of Spartiatai at Sparta: Paus. 3. 18. 8 (cf. 2. 16. 4, and Wide, Ldk. Kulte, 337, n. 1), 3. 11. 10.] My Obal thesis depends on Amyklai being Spartiate and I have no moral doubt of it, but I do not see that it can be proved αὐτὸ καθ' αὑτό. Every attempted proof begs the question of what exactly the differentia was which distinguished perioikoi from Spartiàtes.—Nor is it certain that the Nεοπολῖται are of Hellenistic date (again I have no moral doubt: cf. Pareti, , op. cit. 176–7)Google Scholar. cannot disprove that the classical Five Obai were the known four plus the Neopolitai, in fact the ball-game teams. The Neapolitai would then date from the seventh century, and when Aristotle says ('Aθ. π. 21. 4) that Kleisthenes assimilated his νεοπολῖται, he could be understood as contrasting this with the Spartan way.—Against the view that there were only 4 Obai and Obal Lochoi, Aristotle's five names are surely conclusive: nor can Ar. Lys. 453 be cited in support of it. The Proboulos orders the assault; Lysistrata says νὴ τὼ θεὼ γνώσεσθ³ ἂρα ὂτι καὶ παρ' ἡμῖν εἰςι τέτταες λόχοι. There can be no reference to the Obal Army: in 411 ‘four Lochoi’ was one-third of the establishment. I suggest there were 4 Lochoi in Dekeleia (soon reinforced, Th. 8. 71. 1).

^{page 121 note 1} Hdt. 9. 53. 2–3, 57: Th. 1. 20. 3. Hard going: Th. 5. 68. 2 (see n. 1 on p. 118), 72. 3, cf. 4. 40. 2 and the apophthegms in Plut. Mor. 190 D (Agis 5–6).

^{page 121 note 2} Fr. 541: for πάτρ7iota;οι see n.3 on p.118.

^{page 121 note 3} Hdt. 7.234. 2: πόλι7epsiv;ς πολλαί…Σπἀρτη πόλις.

^{page 121 note 4} They correspond roughly to e.g. Oropos, Salamis, Eleutherai, but only very roughly. These Attic ‘perioikoi’ are not called ‘Aθηναῖνιοι and moreover πολῖτται (n. i on P. 126): cives sine suffragio perhaps. The fact is that laconia had Hdt. 9. 53. 2–3, 57: Th. 1. 20. 3. Hard going: Th. 5. 68. 2 (see n. 1 on p. 118), 72. 3, cf. 4. 40. 2 and the apophthegms in Plut. Mor. 190 D (Agis 5–6).

^{page 121 note 2} Fr. 541: for πάτρ7iota;οι see n.3 on p.118.

^{page 121 note 3} Hdt. 7.234. 2: πόλι7epsiv;ς πολλαί…Σπἀρτη πόλις.

^{page 121 note 4} They correspond roughly to e.g. Oropos, Salamis, Eleutherai, but only very roughly. These Attic ‘perioikoi’ are not called ‘Aθηναῖνιοι and moreover πολῖτται (n. i on P. 126): cives sine suffragio perhaps. The fact is that laconia had Hdt. 9. 53. 2–3, 57: Th. 1. 20. 3. Hard going: Th. 5. 68. 2 (see n. 1 on p. 118), 72. 3, cf. 4. 40. 2 and the apophthegms in Plut. Mor. 190 D (Agis 5–6).

^{page 121 note 2} Fr. 541: for πάτρ7iota;οι see n.3 on p.118.

^{page 121 note 3} Hdt. 7.234. 2: πόλι7epsiv;ς πολλαί…Σπἀρτη πόλις.

^{page 121 note 4} They correspond roughly to e.g. Oropos, Salamis, Eleutherai, but only very roughly. These Attic ‘perioikoi’ are not called ‘Aθηναῖνιοι and moreover πολῖτται (n. i on P. 126): cives sine suffragio perhaps. The fact is that laconia had no synoikisis in the sense in which Attica did, in the sense that people domiciled in many places had one political centre in which they shared on equal terms (Th. 2. 15–16): this is the truth behind the moralized account in Isokr. 12.177–81. Thucydides' denial of synoikisis in Sparta (1. 10. 2) makes quite another point: there was a synoikisis of the five Obai in the sense I have seventhjust denned, the inhabitants retained their domiciles but pooled their political life. Thucydides' point in r. 10. 2 is that this process was not accompanied by the usual material signs: he is contrasting it, of course, with Perikles' Athens.

^{page 122 note 1} τὰ χωρία, Xen. Hell. 3. 3. 5: you might find one Spartiate in each, surrounded by enemies. The Mantinean χωρία, ib. 5. 2. 7.

^{page 122 note 2} Besides the evidence of barracks life (next toponote), this domicile in Sparta is implied by Hdt. 7. 234. 2; perhaps too by the narrative of the earthquake (Plut. Cim. 16.4–6; Diod. n. 63.6–7).

^{page 122 note 3} I have paraphrased Plato's words freely (Laws 666 D–667 A): the phrase about the barracks is ςτρατοπέδοε γὰρ πολιτείαν ἔχετε ἀλλ' οὐκ ἐν 祀στεσι κατωκηκότων: Isokrates' words are (6. 81) τὴν πολιτείαν ὁμοίαν κατεστησάμεθα στρατοπέδω καλ膂ς διοικοεμένω. When Xenophon speaks of τὰ ἑν τ脵 πόλει καλά (Hell. 5. 3. 9; cf. πάντα τὰ καλά, 脵παντα ὅσα καλὰ ἐν τ;脵 Σπάρτ脱 in 5. 4. 32–3) he refers to this barracks discipline: cf. Δακ. π. 5. 2. A well-preserved seventhjust century στρατόπεδον is the little town at Vroulia in south Rhodes (Kinch, , Vroulia, pp. 6–7)Google Scholar: walled and tight-packed. The unwalled Sparta is by contrast εὐρεάγεια, εὐρύχορος (Terpander, loc. cit.; Oiyss. 15. 1).

^{page 122 note 4} Strabo is explicit about Mesoa and Limnai (8. 5. 3 ): Mεσσόαν δ' οὐ τ脴ς χώρας ε禐ναι μέρος ἀλλὰ τ脴ς Σπάρτης καθάπερ καὶ τὸ Διμναῖον (for the reading see Bölte, in P-W, s.v. ‘Sparta’, p. 1363, 57–9)Google Scholar. The passage is polemical but this topographical fact is not: those who equated Messe and Messoa are brushed aside as not having taken account of the topography, not as having held another topographical theory.

^{page 122 note 5} Athen. 141 f. The passage about γένη which Kahrstedt quotes from Aelius Aristides refers to Athens (Staatsr. i. 70, n. 1: Aristid. i, p. 314 Dind.).

^{page 143 note 1} I don't think that the Spartans were registered for taxation.

^{page 143 note 2} An earlier date is excluded, I think, by the operative part of Clause I, which gives the Ekklesia regularity of assembly. This is very advanced: contrast Odyss. 2. 26–8.

^{page 143 note 3} Immediately above, Xenophon records the King's sacrifice ΔĐ Aλήτορι καὶ τοῖς σĐν αὐτῷ, which all editors emend (cf. however Apol. 24, καὶ τῶν σύν τούτοις θ7epsiv;ῶν) Does σĐν αῷ perhaps conceal Σελλα7nu;ίῷ? Aγήτορ7iota; καὶ [τοῖς] Σελλανἰέ. Earlier in the seventh century Terpander wrote a poem for Zeus Agetor (fr.I): at the time of the Rhetra this Zeus this second title?

^{page 125 note 1} Momigliano calls my attention to the theory of Zancan, L., that Centuria originally meant a hundredth part: Atti Inst. Veneto, xciii (1933–1934)Google Scholar, technip. 869. I am not competent to judge this theory, which Momigliano has criticized in Studia et Documenta, ii (1938), p. 510 f.Google Scholar, ‘Studi sugli ordinamenti centuriati’.

^{page 125 note 2} Hdt. 7. 234. 2, 9. 10. 1, 9. 12. 2.

^{page 125 note 3} Meyer, , Kl. Schr. ii, pp. 267–8Google Scholar. I owe this parallel to Momigliano. It seems to me a promising analogy. Were those over 60 excluded from the Obal Katalogos and the Ekklesia, (sexagenarii de ponte, Meyer, 267, n. 3)?Google Scholar If such analogies are valid, I would trace them to the earliest Greek colonists rather than to a common Aryan origin.

^{page 125 note 4} The technical term is possibly συντεταγμὲνοι: Xen. Hell. 3. 3. 7, among many other technicalities (but another sense is possible there). Cf. Vect. 2. 3, and Plato's ψυχή σὼματι συντεταγμὲνοι, Laws 903 d: in these passages something is incorporated in something different. But the word often means simply enrolled as one among others (such must be its meaning in 6. 5. 29, since these Helots were not embodied in the establishment): ‘under arms’ (Dem. 21. 223), ‘disciplined’ (Th. 3. 108. 3; Xen. Hell. 4. 8. 22), ‘professionals’ (Hiero, 10. 6–7).

^{page 125 note 5} Sc. it prolongs the time with the colours, Attention was given to the physique of men past youth ἐζ ων ῆδη και αι μὲγιστσι ἀρχαι καθιστανται [I think this means men over 30: cf. ακ. π 2.2]…ὄπωσ σύναιντο…στρατιωτικύσ πólνονσ ύποπἑριν: Xen. Aακ. π. 4. 7. It was necessary to raise the age because, though by embodying perioikoi the year-class was kept at almost the formaold strength (192: 200), yet the perioikoi could not now be asked to provide a substantial second force as in 479. We seldom (e.g. Hell. 5. 3. 9) hear of perioikoi outside the establishment, except the Skiritai.

^{page 126 note 1} The Morai are regularly called τό πολιτικόν (στράτευμα) or οἱ πολῖται (Xen. Hell. 4. 4.19, 5. 2S5 4 41) 5 4 55t 6.4 26,7.1.28,7.4.20 (cf. 27): contrast 6. 5. 21. τῶν πολῖται μορῖν δακ. π, II. 4 is correct (codd. et Harpocr.) though ὁπλιτικῖν (Stob.) makes sense.

^{page 126 note 2} Aristotle says (Pol. 1270²37) the Spartiates had once 10,000 men (of fighting age? supply ὁπλιτασ vel. sim. from 1270²30). Aristotle studied the Obal Army, and I would like to take this responsible figure: and further to believe that Suidas s.v. ἐνωμοτια goes back to him, ‘a formaoldtion of 25 men’. But this is no doubt an inference from the view that it is a half-Pentekostys, and goes back rather to Xenophon: and the 10,000 Spartiates are probably only a round figure.

^{page 126 note 3} I am positing that Dipaia is between the earthquake and the battle of Ithome (or Isthmos, Hdt. 9. 35. 2): this means immediately after the earthquake, the same time as the 300 are lost ir, in Messenia (9. 64. 2). The ‘ingle line’ is legendary, no doubt: so too the faked epiphany in Polyaen. 1. 41. 1. A legendary moment.