^{page 113 note 1} In Kirchner, , Prosopographia Attica, no early Miltiades with a deme-title fails to come from the Philaid deme, Laciadae.Google Scholar

^{page 113 note 2} Hdt. 1. 76–77: after a campaign in Cappadocia Croesus was brought to battle and returned home, arriving five months before the spring of 545; Sardis fell later in the winter. For the date of the fall of Sardis, see my article in Historia iv (1955), p. 394.Google Scholar

^{page 113 note 3} Various reconstructions of the Philaid stemma are given by Beloch, , G.G. ii. 2. 43Google Scholar (1931); Obst, in R.E., ‘Miltiades’ 1679 (1932); Berve, , ‘Miltiades’ in Hermes Ein-zelschr. ii. 4 (1937);Google Scholar and Schachermeyer, , in R.E., ‘Philaidai’ 2119 (1938).Google Scholar

^{page 114 note 1} Even in Greek Epic, which rejoices in epithets and periphrases, I do not think this journalistic trick is used in the same sentence. Berve, H., loc. cit. 5 with n. 6, seems to assume that Miltiades the uncle and Miltiades the founder are the same person, but he does not discuss Greek usage. It may be suggested that the passage is corrupt; but the suggestion is hard to accept when the sentence is so neatly constructed with internal contrasts.Google Scholar

^{page 114 note 2} Kirchner, , Attische Genealogie, 280 n. 1, dismisses this passage in Aelian as ‘unrich-tig’ But Aelian is citing well-known triads, and it is difficult to see how he could make so glaring an error. In the preceding words Aelian mentions two Perianders; Kirchner can hardly dismiss that as a mistake, because Aristotle mentions two Perianders (Fr. 517).Google Scholar

^{page 114 note 3} F.Gr.H. 3 F 2.

^{page 115 note 1} Cherronesus appears as a ‘polis’ also in the Scholiast to Ar. Eq. 262.

^{page 115 note 2} Hdt. 6. 103. 1 and 4.

^{page 115 note 3} Those who maintain that there were only two Miltiadae see the same person as victor at Hdt. 6. 103. 2 and 6. 36. 1; if so, it is surprising that Herodotus did not name him in each case as the son of Cypselus, his usual appellation of Miltiades I. But the objection to their case lies in the three passages with which I began this article.

^{page 115 note 4} Two codices read seven read As Hude and Budé read the former, I give it here.

^{page 115 note 5} The aorist tense may mean that before his death the subject of the sentence associated Stesagoras with him and handed over authority before his decease; the same phrase occurs at Hdt. 2. 159.

^{page 116 note 1} L.S.J, s.v.; Hude in Teubner ed. ‘post in E spatium novem fere litt.’; E is the earliest and main codex.

^{page 116 note 2} Casaubon, working the other way, emended the words of Marcellinus to read for It is hard to explain a corruption of and, if one does so, it still remains necessary to change Hdt. 6. 103. 4. The fact that Marcellinus cited the family tree of the Philaids so far only as Miltiades I does not show that Miltiades I had no children. We do not know at what date Pherecydes closed his work. In any case Marcellinus only quotes Pherecydes to indicate that Thucydides, being descended from the general Miltiades, came of a noble family going back to Aeacus, son of Zeus; his purpose was served when he reached Miltiades I, for everyone knew that Mil-tiades and Cimon, ‘those most famous generals’, were descended from Miltiades I (2). Finally Marcellinus himself states that Miltiades I had a son (9–10).

^{page 117 note 1} Professor A. D. Nock made this suggestion during a conversation with me.

^{page 117 note 2} Hdt. says not that the Chersonesitae sacrificed to him as the founder but that they sacrificed to him ‘as the custom is in the case of a founder’. I cannot find a parallel in Herodotus, and I doubt whether the distinction in meaning is very significant in itself; but the words are not the same as in Hdt. 6. 103. 4 and Marcellin. V.Th. 10.

^{page 117 note 3} Förster, , G. H., Die olympische Sieger, p. 8.Google Scholar

^{page 117 note 4} Since the mares are more likely to have been in form for eight than for twelve years, we may assume that the victories were successive. As Herodotus tells the story, on the occasion of the first victory Cimon was in exile; on the occasion of the second he allowed Peisistratus to be proclaimed victor and secured his own recall; and on the occasion of the third or soon after it he was murdered by the Peisistratids, Peisistratus himself being no longer alive. We know that Peisistratus was alive at the beginning of the Attic archon-year 528/7 (for he died during the course of its twelve months), and the Olympic Games were held usually a fortnight after the beginning of the Attic archon-year. The odds are therefore high that Peisistratus was alive at the time of the Olympic Games of 528, and that, if Cimon's murder followed close on the proclamation of that victory, Peisistratus would have been responsible. Therefore I think it more reasonable to place Cimon's third victory in 524 with Förster, , op. cit., p. 9,Google Scholar than in 528 with Cadoux, T. J., J.H.S. lxviii. 110 n. 217,Google Scholar and Wade-Gery, H. T., J.H.S. lxxi. 214. I am grateful to Cadoux for a discussion of this matter with me.Google Scholar

^{page 118 note 1} That the word in Hdt. 6. 103. 4 may be used of an adult, is clear from the passage in Hdt. 1. 130. 3 where the career of Cyrus up to and including the capture of Astyages is summarized in the words compare the entries III-V for in L.S.J.

^{page 118 note 2} For the date of Darius' invasion of Scythia see my article in Historia iv (1955), p. 394.Google Scholar

^{page 118 note 3} Stein stultifies the whole passage by adding before NO codex has Legrand follows Stein; Hude does not.

^{page 119 note 1} Hude wrongly brackets thus: The imperfect means that he was continuously in exile and the place, from which he was in exile, is in the accusative; cf. Thuc. 5. 26. 5 Of the codices PDRSV read and AB read which Legrand prefers; C omits the word.

^{page 119 note 2} The text is bracketed or emended by Dobree, Powell, (C.Q. xxix. 159)Google Scholar, and Wade-Gery (see J.H.S. lxxi. 216Google Scholar), but there is nothing in the manuscript tradition to suggest corruption.

^{page 119 note 3} A considerable number of cities in the area of the Hellespont and Bosporus were burnt, and Megabazus' forces probably used the Chersonese as a base; cf. Hdt. 4. 143–4; 5. 1; Ctesias 17; Polyaen. 7. n. 5.

^{page 119 note 4} Ps.-Andocides 4. 33. Paus. 6. 10. 8 says he will recur to a dedication by ‘Miltiades the Athenian’, that is Miltiades III, at Olympia in connexion with statues of horses and chariots, but he does not do so. Förster, G. H., op. cit., p. 9,Google Scholar who does not refer to Paus. 6. 10. 8, thinks that [Andocides] confused the victories of father and son, Miltiades and Cimon, with the three victories of Cimon, son of Stesagoras; but Andocides pins the ostracism and the charge of relations with his sister to the correct Cimon. It is certainly not to be supposed that Andocides meant to refer all the points to Cimon, son of Stesagoras.

^{page 119 note 5} It is probable that in the fifth-century an age-limit was introduced to exclude in fluential young aristocrats.

^{page 119 note 6} Beloch, , G.G. ii. 2. 136,Google Scholar and others condemn Herodotus and acquit the Peisistratidae. But Herodotus was in a better position than Beloch to know the facts and form a judgement.

^{page 120 note 1} The passage in Marcellin. V.Th. 3 is corrupt; Vömel, , Exerc. chronol. de aet. Sol. et Croesi, 16, suggested inserting Tisander between Miltiades and Hippocleides, as I have done.Google Scholar

^{page 120 note 2} Cadoux, , J.H.S. lx. 90 notes 82–86 and 115 n. 249, discusses this phenomenon with reference to other writers. It is probable that the ban on tenure of a second archonship was introduced together with the age-limit and for the same purpose. Cadoux is correct in saying ‘nor, again, do we know if this man or men belonged to the Philaid family’. For, in the absence of evidence either way, we cannot be certain that the Miltiades of 664/3 is the same person as the Miltiades of 659/8 or that both belonged to the Philaid house. But the odds are so high in favour of one of them being a member of that leading family (see p. 113, n. 1) that he should be accommodated in any stemma given for that family. Personally I think the odds are high in favour of the same Philaid Miltiades having been elected twice. In 575 the Philaidae were the richest family in Athens; they were perhaps surpassed later by the Alcmeonidae, when Alcmeon got money from Croesus (Hdt. 6. 127. 4; 125. 2).Google Scholar

^{page 120 note 3} Of the few families we have mentioned, Cypselus and Periander of Corinth practised polygamy (for Periander see Hdt. 3. 50–53 where the point is that Periander had only two sons by Melissa, his otiier sons being by other wives or another wife—a point which Beloch, , G.G. i. 2. 283,Google Scholar seems to miss); Peisi-stratus of Athens, Miltiades III, and his son Cimon may well have done likewise. So too women of noble houses re-married; besides the anonymous wife of Cypselus and Stesa-goras in the Philaid family, we have the example of Timonassa, who was married to Archinus of Ambracia and then to Peisi-stratus. If a noble was already married when he took a second wife, it was not necessarily an insult to the family of the first wife or to that of die second wife. Melissa was killed when her sons were 18 and 17 years old; yet Periander had other sons of similar age, who were therefore born of odier wives or another wife during Melissa's lifetime (F.Gr.H. 90 F 59), and the tension between Periander and Procles only came to a head after die deadi of Melissa. So, too, when Miltiades, already die father of sons by an Adienian wife, married the daughter of the Thracian king Olorus, it was no insult to Olorus whether his Athenian wife was alive or deceased. In die present case Uiere is PO reason to suppose that Hippocleides' fadierhood was any bar to his wooing of Agariste, or even to suppose Uiat he was a widower (C.A.H. iii. 765Google Scholar). The aristocrats of the seventh and sixdi centuries were a law unto themselves; the closest parallel is the aristocracy of Macedonia.

^{page 120 note 4} It is a mistake to suppose that all families in all states or in any one state observed a minimum age of marriage, especially if it is put at 30 years on the basis of Solon fr. 27. 9 who says 28–35 is a good age for marriage (e.g. C.A.H. iii. 765Google Scholar). All we can say is diat in general Solon, Plato, and odiers advised a man to marry between 28 and 40; die practice we can learn better from die stem-mata. For the Alcmeonids see C.Q. vi. 47.Google Scholar

^{page 122 note 1} Cardia = Scymn. 699; Strabo fr. 52 (citing Hdt.); Schol in Dem. 63. 16. Pactye = Scymn. 711. Crithote = ibid.; Ephorus in F.Gr.H. 70 F 40. Kahrstedt, U., ‘Beiträge zur Geschichte der Thrakischen Ghersones’ (in Deutsche Beiträge zur Altertums-wissenschaft, vol. 6 (1954), p. 6) regards Miltiades I as a ruler of the Dolonci, a ‘Stammeshaupt’.Google Scholar

^{page 122 note 2} Frazer commenting on Paus. 6. 19. 6 says that Nepos made a mistake; Frazer thinks Nepos should have said ‘Miltiades, son of Cypselus’. This is quite arbitrary; for the story in Nepos is entirely different from that in Hdt. 6. 34–36 both in detail and in chronology.

^{page 122 note 3} Nepos may have misunderstood the allusion in Hdt. 7. 6. 3 to the islets adjacent to Lemnos; see Berve, , op. cit. pp. 22–23.Google Scholar

^{page 123 note 1} Herodotus is careful to distinguish types of warship from one another. At this stage most of the Athenian fleet consisted of pente-conters (Thuc. I. 14. 3).

^{page 123 note 2} Herodotus mentions the trireme in which Miltiades sailed; part of the ‘picked band’ may have sailed on other ships, and the Athenian settlers probably came later.

^{page 123 note 3} F.Gr.H. 70 F 40 (Ephorus) and Suidas, and Scymn. 711 (adding Pactye) may refer to this period of settlement. Mil tiades I took only some Athenian volunteers in 556; and Miltiades III in 496–493 would have had difficulty in persuading many Athenians to settle in the Chersonese during the Ionian revolt. The setUement at Crithote is, then, an example of Nepos' phrase ‘multitudinem, quam secum duxerat, in agris collocavit’ (Miltiades 2. 1).

^{page 123 note 4} On this passage the usual comment (Schubart, Hitzig-Blümner, Frazer, Berve, etc.) is erravit Pausanias; ‘Miltiades son of Cypselus was the first Athenian tyrant of the Chersonese’. But let us see what Pausanias says. In the Teubner ed. Spiro bracketed in order to read so that the meaning would become ‘held office’. However, the manuscripts all read Further, Pausanias does not use the imperfect tense but the aorist just as Hdt. 6. 39. 2 used the vivid present in Th. 54. 2). Pausanias' point is that Miltiades son of Cimon first seized power in the Chersonese. The dedication in Attic script was not made by native Chersonesians, as Frazer seems to suppose, but by Athenians operating from the town Chersonesus; why should the native Chersonesians capture dieir own district or fort, whichever Aratou Teichos was? The inscription in ‘ancient Attic letters’ may have resembled the Phanodicus inscription (Syll. 1 ³. 2). For commentary see Fried-länder, P., Epigrammata (1948), nr. 52,Google Scholar who notes that the style suits a date late in the sixth century. For the most recent discussion of the city Cherronesus and its coins see Kahrstedt, U., op. cit., pp. 8–9.Google Scholar

^{page 124 note 1} Both verbs are in the imperfect tense. For with a personal object see Hdt. 1. 71 fin., 1. 130 fin. et alibi, where the meaning varies from conquering to de stroying. The meaning here is ‘destroying’, because conquest in the aorist tense ( four times) has already been mentioned in the preceding chapter and because here we require the stronger meaning in the phrase

^{page 124 note 2} Rawlinson reverses the order of enslaving and conquering, i.e. as I take it ‘destroying’. Legrand in the Budé edition has ‘ce Lycarétos mourut à Lemnos dans 1'exercice de son commandement … voici ce qui motivait ses actes: il réduisit…’ (with a note ‘il = Otanes’). Godley in the Loeb edition translates ‘this Lycaretus came to his end while ruling in Lemnos; this was because he strove to enslave and subdue all the people, accusing some… ’ The difficulty here is diat Godley requires a pun on and , which in Herodotus' idiom of writing must have the same meaning; that it is an unintelligible reason for Lycaretus' death with Persia's might and methods behind him; and that ‘all the people’, whoever they are, have been subdued by Otanes in the previous chapter.

^{page 124 note 3} Nor does it make sense historically; for some of these states had already been punished by Darius and Megabazus for their part in the Scythian campaign and its aftermath (Hdt. 4. 144; 5. 1; Polyaen. 7. 11. 5; Ctesias 17), Calchedon in particular having been sieged and burnt by Darius on his return.

^{page 125 note 1} In this sentence the following words stress the resumption after a digression:

^{page 125 note 2} Of innumerable examples it may suffice to quote Hdt. 6. 40 fin. where refers to Miltiades' restoration.

^{page 125 note 3} The meaning in the lacuna may be ‘during his rule in Lemnos Lycaretus puts an end to the Pelasgians’. This would justify the vivid present and lead up to what follows. One or two words only may have dropped out (e.g. if such a usage is possible, which I doubt, meaning ‘puts an end to them’, or else meaning ‘ends by destroying them’).

^{page 125 note 4} Herodotus' method of exposition resembles that of Thucydides to which I drew attention in C.Q. NS ii (1952), 127 f. One of many examples on a larger scale may be noted in 6. 33 to 6. 41, where the arrangement of the narrative is as follows. The Phoenicians conquered with the exception of Cardia all the cities of the Chersonese (6. 33. 3). Their ruler was Miltiades son of Cimon, the position of ruler having been won earlier by Miltiades son of Cypselus (6. 34. 1). The activities of Miltiades son of Cypselus are then narrated (6. 34. 2 to 6. 38. 1 init. ); next the narrative is brought down to the first establishment of Miltiades son of Cimon in the Chersonese (6. 38. 1 to 6. 40. 1 ); next comes the narrative which brings the Phoenicians towards the Chersonese and the flight of Miltiades from Cardia (6. 40. 1 to 6. 41. 1). The order of topics is ABCCBA.

^{page 125 note 5} The Samians who escaped the massacre under Syloson probably joined Lycaretus the Samian in Lemnos.

^{page 126 note 1} Meyer, , Forschungen (1892), i. 14,Google Scholar holds that the Athenians expelled all the natives and installed a cleruchy. Hdt. does not say so; in any case cleruchs used natives to till the land in most cases. Bérard, , Revue des études anciennes, li (1949), 228, also holds that Miltiades expelled all the Pelasgians.Google Scholar

^{page 126 note 2} Berve, , op. cit., p. 50,Google Scholar dating Lemnos' fall between 510 and 505, does not mention Nepos, , Miltiades, 1–3.Google Scholar

^{page 127 note 1} Herodotus cited Hecataeus and set against his account that of the Athenians in 6. 137. He then ceased to cite Hecataeus and the account in 138–140 seems to be a purely Athenian one. In theory Nepos might have drawn on Hecataeus; but in fact Hecataeus probably did not include historical narrative of this type. For Hellanicus' interest as a Lesbian in Pelasgians cf. Pearson, L., Early Ionian Historians, pp. 157 f. on the Phoronis; for this narrative Nepos would have drawn on Hellanicus' Atthis, and, if it is so, then this is the only passage concerning late sixth-century history which may have been derived from Hellanicus.Google Scholar

^{page 127 note 2} It is interesting that a similar difficulty arises in Marmot Parium 46 , for which and have been suggested.

^{page 128 note 1} That the word in Ath. Pol. 20. 1 means ‘akin’ is confirmed by its use in 18. 4 and 18. 5 because there the point is that the tyrants will commit by killing their own kindred, and by its use in 22. 4 where refers to Hipparchus being a member cf. 22. 6. The kinship of the Peisistratidae, who came from the deme Philaidae, with the Philaid house is stated in Marcellin. V.Th. 18.

^{page 128 note 2} Kirchner, , Prosopogr. Attica, s.v., does so tacidy.Google Scholar

^{page 128 note 3} Mustilli, D., ‘L'occupazione ateniese di Lemnos e gli scavi di Hephaistia‘ in Studi.. offerti a E. Ciaceri (1940), p. 149,Google Scholar and especially, pp. 156–8. I was able to see this rare book through die courtesy of Bodley's Librarian.

^{page 128 note 4} Haspels, C. H. E., Attic B-F Lekythoi (1936), pp. 33 and 41.Google Scholar

^{page 128 note 5} This is also true of Wade-Gery's attractive comparison of the Acropolis mounted archer widi the archer on die Ashmolean plate, J.H.S. lxxi (1951), 220.Google Scholar

^{page 129 note 1} Richter, G. M. A., Attic Red-figured Vases (1946), p. 44 with refs.Google Scholar in n. 21; Langlotz, E., Zur Zeitbestimmung der str. Vasenm. (1920), pp. 59 f.Google Scholar; Nachod, , R.E. ‘Stesagoras’ 2457 (1929).Google Scholar