¹ Mr. Harrison, however, writes that he is not sure that there is linguistic objection: ‘the second τήν may mark the fact that the first genitive is double, the second single; and it avoids ambiguity.’ Such duplication of the article does of course occur, but rarely.

² It is quite beyond my understanding that Adcock, (Klio, xii. p. 5)Google Scholar can suggest that this chapter may be ‘the result of research by Aristotle’; still more that Seltman (p. 15, 4) can call him ‘probably the first scientific numismatist.’ Here, if anywhere, ‘er hat weder sich selbst noch seinen Lesern ein Bild jener Verfassung zu entwerfen versucht, sondern sich begnügt eine sehr kurze und ungleichförmig gearbeitete Skizze fast ausschliesslich auf Grund der Darstellungen zu liefern, die er bei den Atthidographen fand’ (Wilamowitz, i. 55). Indeed, this is the only justification for one who is not a numismatist discussing the chapter, that Aristotle was not one either. (Mr. Seltman writes to me: ‘My point would be, Aristotle was quite a good numismatist (in the same degree as he was a good marine biologist), but a bad popular-handbook-writer. Perhaps he had the facts of the Solonian reform clear in his own head, and perhaps he thought he had put those facts down with admirable clarity, the while his phrasing was distinctly esoteric.’ But I am not convinced.)

³ Th. Reinach and Beloch are exceptions: see below.

⁴ Beloch, (Gr. Gesch. ² i. 2. 335 ff.)Google Scholar holds not only that Aristotle thought this, but that it is true and that this is the meaning of On the value of his documentary evidence in support of this, I am not qualified to speak (see Reinach, Th., B.C.H. 1896, pp. 251–6, 385, and 1904, p. 18Google Scholar: Reinach's former explanation seems on the face of it the more probable). But Beloch goes on to argue that there isno divergence between Aristotle and Androtion: in the last sentence in c. 10 the former says the talent weighed 63 minae—that is, says Beloch, the new talent weighed 63 old minae, each of which had 70 dr.; 63 × 70 = 4410 old dr. to the new talent; whereas Androtion says the old mina weighed 73 dr.—therefore 60 × 73 = 4380 old dr. to the new talent. The only difference then is that Aristotle makes the relation of the old dr. to the new 73½%, Androtion 73%, and the loss of the fraction in the latter is very likely due to Plutarch. But this argument is doubly false: refers, as everyone has seen, to a change in the trade-talent (expressed in terms of the coinage-mina), not in the coinage; and if in Aristotle means that the old mina had 70 dr., in Androtion must mean that the old mina had 73 dr., more particularly if, as Beloch supposes, Aristotle is only copying Androtion without under-standing him.

⁵ Adcock does not notice that in the use of πρώτην here, which is not necessary to the narrative as it stands, there is some slight support for his view that this is the date of the second expulsion.

⁶ It is also, of course, inconsistent, and in a different way, with the theory that Peisistratus was only exiled once.

⁷ In explaining Aristotle we should not lay any stress on the fact that Hdt. i. 61 and Plut. Cat. mai. 24 say that Hippias and Hipparchus were grown up when Peisistratus married Megacles' daughter (Hdt.) and when he married Timonassa (Plut.). With Adcock's dates this would mean that they were both over 18 in 560, and Hippias therefore 90 or more at the battle of Marathon. (It would be better on Adecock's own theory to alter his first dates: say 560–59 first expulsion, 557 first return. This would more easily explain the doubts as to the date of this marriage, and at the same time suits better the story of Megacles' daughter.)

Incidentally, it is time we ceased arguing, as Sandys for example does, that Peisistratus' marriage with ‘the Argive woman’ more probably took place during his exile than during his first rule, ‘when her presence in the palace would not have ingratiated him … with his wedded wife.’ no more means ‘the Argive woman’ (as opposed to ‘wife’ or ‘lady’) than means ‘the Argive fellow’ and the Athenians were not polygamists. Peisistratus' first wife was dead or divorced, when he married Timonassa. At a later time, when there was a more cogent distinction between marriage with an Athenian and with a foreigner, Pericles divorced his wife before marrying Aspasia.

⁸ One should note also 8. 4 τὴν δὲ τῶν ᾿Αρεοπαγιτῶν ἔταξεν ἐπὶ τὸ νομοφυλακεῖν ἄσπερ ὑπῆρχεν καὶ πρότερον ἐπίσκοπος οὐσα τῆς πολι τείας and observe that this ‘correction’ points back not only to the Dracontic constitution, 4. 4, but also to the pre-Dracontic, 3. 6. Incidentally, not all of c. iv may be ‘interpolated’: we cannot help noticing how admirably §§ 4–5 ν῾ δὲ βουλὴ ἠ ἐξ ῾ Αρείου πάγου φύλαξ ἠν τῶν νόμων κτλ would suit § 1 Δράκων τοὺς θεσμοὺς ἔθηκεν hadsome account of the θεσμοί or of their publication and its effect (or of the ἐφέται and the φυλοβασιλεῖς—Wil. i. 94) followed these words, instead of ἡ δὲ τάξις αὕτη κτλ

⁹ Not to mention the slightly divergent account in the Politics, which is clearly not from Herodotus, for the figures are different. This in itself is sufficient to show that the whole story of Peisistratus' reign in the Atthis was not dependent on Herodotus.

¹⁰ It should however be noted that in giving Hegesias as the archon for this year, Aristotle is at variance with Marm. Par., which names Euthydemus as fifth after Comeas; as has been pointed out by Pomtow, (Rh. Mus. 1896, pp. 570 ff.).Google Scholar

¹¹ I do not see why Adcock should object, ‘How is it likely that Aristotle or any other Athenian authority at his disposal should know how many months Peisistratus ruled ‥ And why, if Aristotle did know this chronology so exactly, did he qualify the phrase by a tentative and modest It is quite easy to suppose a tradition that Peisistratus quarrelled with Megacles after ‘about six months.’ Thucydides gives very few dates in the Pentecontaëtia, but he can tell us that Oenophyta was fought 61 days after Tanagra, and that the Corinthians turned out again to face the Athenians after the previous battle. Aristotle knew, or thought he knew, that the Cypselid tyranny lasted 73 years and 6 months (Pol. v. 12. 22).

¹² Pomtow—l.c.: see Adcock, p. 175—should not have argued that the ordinal numbers in the Ath. Pol. deserve special respect as they are not liable to the confusion of figures, being written in full: in 54. 7 we have for πέμητη, in 32. 1 for for 47. 4, and in 63. 2 the false reading τριακοστοῦ as an interpretation of λ. He also says, as we should all like to say, that it is arbitrary to take this or that figure and declare it corrupt. Yet some figures in the Ath. Pol. are certainly corrupt—πέμττῳ in 22. 2 (unless there is a big lacuna before), 22. 8 (unless this is a slip on Aristotle's part), and ἑπτακοσίους 24. 3; and this last case does not make all the other figures in the context suspect. Certainly Pomtow's own suggestion that μηνί has twice been corrupted into ἔτει within 15 lines (which still leaves him with an impossible δωδίκάτῳ on his hands) is just as arbitrary.

¹³ Characteristically chronologers like Eratosthenes and Eusebius give him 12 years' reign, reckoning by the dates only (Jacoby, , Marm. Par., p. 195Google Scholar).

¹⁴ Cf. however Jacoby's ingenious emendation of Dem. xxi. 154—ΔΔΔΠ for ΔΔΔΠ (Apollodors Chronik, 57).