^{page 97 note 2} Against e.g. Bury, J. B., CAH iv, 1926, 484Google Scholar. 1 (and others), ‘It is more probable that the reference is to the eclipse of April 15th, 657 B.C., which was total in or near Rhodes and of which the news would have reached Paros (cp. Hauvette, Archiloque 13 sqqGoogle Scholar.)’. Why is that more probable?

^{page 97 note 3} Ginzel, , Spezieller Kanon, 1899, 167Google Scholar; Abh. Berl. Akad. 1918, no. 4Google Scholar. What we need is a computation for the eclipse of 648/7 B.C. of the same accuracy as the computation made by Fotheringham on behalf of Blakeway for the eclipse of 711 B.C., 14 March. According to Ginzel the eclipse of 648 was nearly, though not quite, total (11·72″) at Paros; according to Fotheringham (Blakeway, 36) it was total, while in 711 B.C. it was ‘total at Thasos and not far from total at Paros’. See n. 5.

^{page 97 note 4} Practically, I believe, we have only to choose between Thasos and (more probably) Paros. See p. 107, n. 3.

^{page 97 note 5} ‘The date of Archilochos’ in Greek Life and Poetry, 1936, pp. 34–55.

^{page 97 note 6} loc. cit. p. 36. If he further states that ‘at the most the eclipse of 711 B.C. approaches more nearly to the description of Archilochos than that of 648’, I do not know whether he is right. If μεσημβρ⋯η is astronomically inaccurate, but an impressive poetical expression for the hour of, the day or (as Archilochos had no clock) for the forenoon, a few minutes more or less do not matter much. More important is another consideration. Fotheringham gives 10.16 a.m. for the middle totality local solar time at Thasos in 711, ‘which is nearer to midday than the time when the eclipse of 648 B.C. April 6th was total at Paros’. What time was that? In any case it seems to me that we should compare the middle totalities for Thasos for both eclipses on the one hand, and for Paros on the other. The question is not so much when and if the eclipse of 711 B.C. was total in Paros, but whether the eclipse of 648 B.C. failed to reach totality in Thasos or Paros. It may be added that the eclipse is only a terminus post, as the speaker in the eclipse poem F 74 is Lykambes: Archilochos may have seen the eclipse at Thasos, but the poem may have been composed (not so very much) later at Paros. To judge from the most valuable report of Kritias (Aelian, , VH 10. 13 = Vorsokr. 88 [81] B 44Google Scholar) about the personality of Archilochos, the Neobule affair should be later than the emigration of the poet to Thasos; it belongs to the time when he had returned to his native island.

^{page 97 note 7} loc. cit. 55. I am not sure that this statement is borne out by the facts. True that, e.g. Beloch, , Gr. G. ² i. 2. 350Google Scholar, called the eclipse ‘the decisive testimony’ and Wilamowitz, , Gr. Lit. ³ 1912, 30Google Scholar, said that ‘his time is fixed by the mention of the eclipse of 648 B.C.’, and others too expressed themselves in a similar manner. But they did so or have done so for the sake of brevity. In fact, the earlier scholars were influenced by the ancient date for Gyges which was wrong, and voted for the eclipse of 711 because it seemed to fit in with this date. Some of them even formally rejected the eclipse of 648 because they believed that a contemporary of Gyges could not have seen it. Only after Gelzer, , Rh. M. xxx, 1875, 230ff.Google Scholar, had established the true chronology of the Lydian kings from the Assyrian annals, were scholars driven to assume that the Archilochos eclipse was the eclipse of 648, because most of them quite rightly did not doubt that the eclipse was seen by Archilochos himself, and then 711 seemed much too early. Neither is there in my opinion much of a bias, conscious or unconscious, in the use of literary evidence by modern scholars. On the contrary, Blakeway seems to me to have a bias for the earlier date for reasons which he enumerates in the beginning of his paper (l.c. 34 f.), and strives to fit the literary evidence to this preconceived opinion. In doing so, I am afraid, his knowledge of the development, the manner, and the methods of ancient chronography is not quite up to date, as shown by the review of ‘the external literary evidence’ on p. 35, the treatment of it on pp. 39 ff., and even more by the summing up on pp. 54 ff., when he declares that the rejection of the two authorities Nepos and Eusebios (who, in fact, represent only one witness, viz. Apollodoros of Athens—unus sed led) ‘is a far smaller sacrifice of evidence’ than the discarding of Cicero, Tatian, Cyril, Oinomaos, and Clement. The valuation of sources here is wrong. Apparently Blakeway rejects, or at least fails to do justice to the very great progress made here by Gutschmid, Rohde, Diels, and (ϕορτικ⋯ν τ⋯ ⋯ληθ⋯ς) by me in the Apollodoros (PhU xvi, 1902)Google Scholar and the FGrHist i–ii (1923/1930)Google Scholar. As he does not quote the last-named books and only once mentions the ‘ingenious and attractive theory of Rohde’, with a very characteristic note (II. 40, 2), I am inclined to believe that he consciously rejects their opinions as well as their method; and that he consciously renews what Crusius, (RE ii, 1896, 489 f.)Google Scholar describes as ‘die unkritischen vermittlungsversuche älterer gelehrter die Archilochos steinalt werden lassen, nur um alle ansätze der alten beibehalten zu können’. Further, in trying to fix the date of Archilochos, one cannot quite avoid ‘embarking on the vexed problem of the chronology of the Lydian kings’ (p. 44), even if one excludes detail which is not pertinent. Neither can one quite avoid taking into account the problem of relative chronology in Greek literary history between, say, 750 and 650, at least as far as it is concerned with the relations between Archilochos and Hesiod, Archilochos and Terpander, Archilochos and Kallinos.

^{page 98 note 1} PhU xvi. 144. 3. Personally I believe that Demeas' dates for Archilochos are worth not much more than those of Sosikrates for Epimenides or Anacharsis, who came to Athens κατ⋯ τ⋯ν ⋯λυμπιάδα ⋯π' ἄρχοντος Εὐκρ⋯τους (ibid. 168. 8).

^{page 98 note 2} F 51 Diehl; von Gaertringen, Hiller, Nocheinmal das Archilochosdenkmal von Paros. GGN 1934Google Scholar.

^{page 99 note 1} Herodt. 1. 12. 2 Γ⋯ϒγς, το⋯ κα⋯ Ἀρχ⋯λοχος ⋯ Π⋯ριος, κατ⋯ τ⋯ν αὐτ⋯ν χρ⋯νον γεν⋯μενος, ⋯ν ἰ⋯μβωι τριμ⋯τρωι ⋯πεμν⋯σθη. Hippias, , FGrHist 6 F 6Google Scholar ⋯ψ⋯ ποτε το⋯δε το⋯ ⋯ν⋯ματος (SC. τ⋯ραννος). εἰς τοὺς Ἓλληνας διαδοθ⋯ντος κατ⋯ το⋯ς Ἀρχιλ⋯χου χρ⋯νους, καθ⋯περ Ἱππ⋯ας ⋯ σοϕιστ⋯ς ϕησιν Proklos, , Chrest. Phot. Bibl. 239, p. 319b 27Google Scholar ἰ⋯μβων δ⋯ ποιητα⋯ Ἀρχ⋯λοχ⋯ς τε ⋯ Π⋯ριος ἄριστος κα⋯ Σιμων⋯δης ⋯ Ἀμ⋯ργιος … κα⋯ Ἱππ⋯ναξ ⋯ Ἐϕ⋯σιος, ὧν ⋯ μ⋯ν πρ⋯τος ⋯π⋯ Γὺγου … ἣκμαζεν. The synchronism Homer–Archilochos in Alex, Clemens. Strom. I. 117. 8/9Google Scholar: να⋯ μ⋯ν Θε⋯πομπος (115 F 205) μ⋯ν ⋯ν τ⋯ι τ⋯ν Φιλιππικ⋯ν μετ⋯ ἒτη πεντακ⋯σια τ⋯ν ⋯π⋯ Ἰλ⋯ωι στρατευσ⋯ντων γεγον⋯ναι τ⋯ν Ὂμηρον ἱστορεῖ. Εὐϕορ⋯ων δ⋯ ⋯ν τ⋯ι Περ⋯ Ἀλευαδ⋯ν (F 23 Scheidw.) κατ⋯ Γὺγην αὐτ⋯ν τ⋯θησι γεγον⋯ναι ὅς βασιλεὺειν ἣρξατο ⋯π⋯ τ⋯ς ⋯λυμπι⋯ος (708/4), ὃν κα⋯ ϕησι πρ⋯τον ὠνομ⋯σθαι τὺραννον (the Olympiad is added by Clement from his chronological handbook, probably the Χρ⋯νοι of Dionysios of Halikarnassos; see infra, Strom. 1. 131.7; FGrHist 251 F 3). Tatian, , ad Gr. 31Google Scholar (Euseb, . PE 10. 11. 4Google Scholar) ἓτεροι δ⋯ κ⋯τω τ⋯ν χρ⋯νον ὐπ⋯γαγον, σὐν Ἀρχιλ⋯χωι τ⋯ν Ὂμηρον εἰπ⋯ντες ⋯ δ⋯ Ἀρχ⋯λοχος ἣκμασε περ⋯ ⋯λνμπι⋯δα (688/5), κατ⋯ г⋯γην τ⋯ν Λυδ⋯ν ὓστερον τ⋯ν Ἰλιακ⋯ν ἓτεσι πεντακοσ⋯οις (490 years: Suda s.v. Σιμων⋯δης Κρ⋯νεω) and others. The introduction of Archilochos' name into the Marmor Parium (239 A 33 a. 681/0 or 682/1 B.C.) i s rather dubious; to supply here the colonization of Thasos or (rather) a reinforcement led by Archilochos (von Gaertringen, Hiller, REV A 1311Google Scholar; cf. CAH iii. 654 and others) is unprofitable guesswork.

^{page 99 note 2} Strab. 14. 1. 40, p. 647/8 (from Kallisthenes) κα⋯ τ⋯ παλαι⋯ν δ⋯ συν⋯βη τοῖς Μ⋯γνησιν ὐπ⋯ Τρηρ⋯ν ἄρδην ⋯ναιρεθ⋯ναι, Κιμμερικο⋯ ἓθνος, †εὐτυχ⋯σαντος (-τας F; see p. I04, n. 4) πολὺν χρ⋯νον τ⋯ι δ' ⋯ξ⋯ς ἓτει Μιλησ⋯ους (τ⋯ δ' ⋯ξ⋯ς ⋯ϕεσσ⋯ους Kramer) κατασχεῖν τ⋯ν τ⋯πον. Καλλῖνος μ⋯ν ο⋯ν ὠς εὐτυχοὺντων ἓτι τ⋯ν Μαγν⋯των μ⋯μνηται κα⋯ κατορθοὺντων ⋯ν τ⋯ι πρ⋯ς το⋯ς Ἐϕεσ⋯ους πολ⋯μωι, Ἀρχ⋯λοχος δ⋯ ἣδη ϕα⋯νεται γνορ⋯ζων τ⋯ν γενομ⋯νην αὐτοῖς σνμϕορ⋯ν “κλα⋯ω τ⋯ Θασ⋯ων (Tzschucke, κλα⋯ειν θ⋯σσων Strabo) οὐ (Tyrwhit, οὗ Strabo) τ⋯ Μαγν⋯των κακ⋯”. ⋯ξ οὖ κα⋯ τ⋯ νεώττερον εἶναι το⋯ Καλλ⋯νου τεκμα⋯ρεσθαι π⋯ρεστιν. ἄλλης δ⋯ τινος ⋯ϕ⋯δου τ⋯ν Κιμμερ⋯ων μ⋯μνηται πρεσβυτ⋯ρας ⋯ Καλλῖνος, ⋯π⋯ν ϕ⋯ι “ν⋯ν δ' ⋯π⋯ Κιμμερ⋯ων στρατ⋯ς ἒρχεται ⋯βριμοεργ⋯ν”, ⋯ν ἦι τ⋯ν Σ⋯ρδεων ἄλωσιν δηλοῖ (cf. Kallisthenes 124 F 29). The same, very much abbreviated, in Aristotle-Herakleides, , Pol. 23Google Scholar. 1 and Athen. 12. 29, p. 525c (it is not necessary to discuss here the textual difficulties in the Strabo and Athenaios passages). A detailed discussion on the chronology of Archilochos which did not deal only with the Magnesian affair (misjudged by Blakeway, 50 f., who overlooked the κα⋯) is abbreviated by , Clem. Alex.Strom. 1. 131Google Scholar. 17 Ξ⋯νθος δ⋯ ⋯ Λυδ⋯ς περ⋯ τ⋯ν ⋯λυμπι⋯δα (700/4) —ὡς δ⋯ Διονὺσιος (251 F 3) περ⋯ τ⋯ν(720/17)—Θ⋯σον ⋯κτ⋯σθαι, ὡς εἶναι συμϕαν⋯ς τ⋯ν Ἀρχ⋯λοχον μετ⋯ τ⋯ν ἣδη γνωρ⋯ζεσθαι ⋯λυμπι⋯δα (700/696) μ⋯μνηται γο⋯ν κα⋯ (!) τ⋯ς Μαγν⋯των ⋯πωλε⋯ας προσϕ⋯τως (that means ‘recently’) γεγενημ⋯νης. Σιμων⋯δης (that is Semonides of Amorgos; see Supra, n. 1) μ⋯ν οὖν κατ⋯ Ἀρχ⋯λοχον ϕ⋯ρεται, καλλῖνος δ⋯ πρεσβ⋯τερος οὐ μακρ⋯ι τ⋯ν γ⋯ρ Μαγν⋯των ⋯ μ⋯ν Ἀρχ⋯λοχος ⋯πολωλ⋯των, ⋯ δ⋯ εὐημερο⋯ντων μ⋯μνηται.

^{page 99 note 3} Gelzer, , Rh. M. xxx, 1875, 230 ff.Google Scholar; Lehmann-Haupt, , RE vii, 1912, 1956 ff.Google Scholar; Hogarth, , CAH iii. 507Google Scholar. If not 652, then a little later; there is no need to go into the detail.

^{page 99 note 4} 38 years in Herodotos, 36 in Africanus and in Eusebios', Canon, 35Google Scholar in his introduction. This compromise is accepted by Hogarth, , CAH iii. 501Google Scholar and others.

^{page 99 note 5} Preserved in the Armenian translation, p. 32 f. Karst. In the other sources Gyges' reign begins and ends earlier: 716–679 Herodotos, 698–663 Africanus (699–664 Eusebios in the Canon).

^{page 99 note 6} Gellius, , N A 17. 21. 8Google Scholar.

^{page 99 note 7} Phil. Unters. xvi. 142; FGrHist 244 F 336.

^{page 100 note 1} PhU xvi. 148. The authority for Apollodoros is almost certainly the work Περ⋯ τ⋯ν ⋯ρχα⋯ων ποιητ⋯ν κα⋯ μονσικ⋯ν by Glaukos of Rhegion whom he quotes also for the life of Empedokles (244 F 32). About Glaukos see Jacoby, , RE vii, 1912, col. 1417, no. 36Google Scholar; his outline of musical history col. 1419 f.; cf. Jacoby, , Das Marmor Parium, 1904, 54 f.Google Scholar; 98 f. He was a practising musician, probably a flute-player, who lived (partly at least in Athens) about 400 B.C. As an authority he is used by Herakleides of Pontos in his Συναγωγ⋯ τ⋯ν ⋯ν μουσικ⋯ι, and through him found his way into Plutarch's treatise περ⋯ μονσικ⋯ς. The pertinent passage for the chronological question is ch. 4, p. 1132E ἒνικε δ⋯ κατ⋯ τ⋯ν τ⋯χνην τ⋯ν κιθαρωιδικ⋯ν ⋯ Τ⋯ρπανδρος διενηοχ⋯ναι … κα⋯ τοῖς χρ⋯νοις δ⋯ σϕ⋯δρα παλαι⋯ς ⋯στ πρεσβ⋯τερον γο⋯ν αὐτ⋯ν Ἀρχιλ⋯χου ⋯ποϕα⋯νει Γλα⋯κος ⋯ ⋯στι πρɛσβ⋯τερον γο⋯ν αὐτ⋯ν Ἀρχιλ⋯χου ⋯ποϕα⋯νɛι Γλα⋯κος ⋯ ⋯ξ Ἰταλ⋯ας ⋯ν συγγρ⋯μματ⋯ τινι Περ⋯ τ⋯ν ⋯ρχα⋯ων ποιητ⋯ν τε κα⋯ μουσικ⋯ν ϕησ⋯ γ⋯ρ αὐτ⋯ν δε⋯τερον γεγον⋯ναι μετ⋯ τοὐς μρώτους ποι⋯σαντας αὐλωιδ⋯αν (αὐλητικ⋯ν Westphal); cf. ch. 5, p. 1133A Κλον⋯ς δ⋯, ⋯ τ⋯ν αὐλωιδικ⋯ν ν⋯μων ποιητ⋯ς ⋯ ⋯λ⋯γωι ὕστερον Τερπ⋯νδρου γεν⋯μενος … μετ⋯ δ⋯ Τ⋯ρπανδρον κα⋯ Κλον⋯ν Ἀρχ⋯λοχος παραδ⋯δοταιγεν⋯σθαι. As Glaukos Stood up for flute-music, we do not expect him to date Terpandros earlier than necessary. We do not know whether Hellanikos, who published the chronicle of the Karneia (4 F 85/6), was of the same opinion as Glaukos; but at least he called him the ‘very first’ victor at the Karneia (which were instituted in the 26th Olympiad = 676/3 B.C.; again the Olympiad is a later addition, which is expressly ascribed to Sosibios the Laconian's Περ⋯ χρ⋯νων) and dated him κατ⋯ Μ⋯δαν. Midas reigned according to Eusebios from c. 738 till 696, and the Assyrian Annals confirm this date (Eitrem, , RE xv, 1931, 1538, no. 3Google Scholar; Ed. Meyer, , Gd A ² iii, 1937, 35Google Scholar); that is, he began earlier than Gyges even in the wrong lists. Therefore it is probable that Hellanikos too thought Terpandros the older man of the two. Certain it is for Hieronymos Περ⋯ κιθαρωιδ⋯ν, who dated Terpander κατ⋯ Αυκο⋯ρυον τ⋯ν νομοθ⋯την On the other hand, Phainias of Eresos (Clemens, , Strom. 1. 131. 6Google Scholar) made him Ἀρχιλ⋯χου νεώτερον. Possibly the author of the Parian Marble did so too, if in A 33 a. 681/0 we introduce the rather dubious supplement [Ἀρχ⋯λοχ]ο[ς]; for Terpander, the inventor of the ν⋯μοι κιθαρωιδικο⋯, who τ⋯ν ἒμπροσθεν μουσικ⋯ν μετ⋯στησεν, he dates (A 34) very late in 644/3 (a date which we find also in Eusebios', Canon, sub Ol. 34.4Google Scholar; 641/0). For a possible explanation see Jacoby, , Marm. Par. 189Google Scholar; FGrHist ii. D, p. 686. I would have suggested parochialism on his part, if it were not for Phainias, whom we should expect to stand up for his Lesbian countryman. So he, too, may have given reasons drawn from musical history; but we do not know what they were.

^{page 100 note 2} From Varro?

^{page 100 note 3} It is Clement, not Xanthos or Dionysios, who infers the time of Archilochos from the date of the colonization of Thasos. If there are 20 years between the two dates, he knew from Kritias that Archilochos emigrated to Thasos as a young man. I cannot agree with Blakeway, 50 f., at all.

^{page 101 note 1} Again there is no need to go into detail, as these authorities are all agreed on the last decades of the eighth or the first years of the seventh centuries. But it may be remarked that Eusebios' Canon puts the floruit of Archilochos (together with the other iambographers Semonides and Aristoxenos) in the very last year of Gyges 664 B.C. It is quite on the cards that this, too, is the doctrine of Apollodoros employing the usual artifice of assigning the floruit of a literary man, for whom an accurate date was not available, to the first, the middle, or the last year of a contemporary political personage and thereby not only upholding the old synchronism Archilochos-Gyges, but removing the poet as far as possible under this synchronism from Terpander. The expounder of Homer and many other poetical texts had sense enough to see that it was extremely improbable that the Gyges poem should date from the beginning of the king's reign.

^{page 100 note 2} See p. 107, n. 4; 108, n. 3.

^{page 101 note 3} He was, in accordance with his social standing, no marshal at all, but a simple ‘son of Mars’ (θερ⋯πων Ἐνυαλ⋯οιο ἄνακτος), as he was no founder of a colony (see infra, p. 102 f.). What I mean (and it is nothing new) is quite distinct from Blakeway's mention (p. 55, n. 1) of the ‘modernity’ of Archilochos. This so-called modernity should, in fact, not be used to date the poet as late as possible. So far I quite agree with Blakeway. But if other signs point to a later time, it may be taken into account. The question does not reflect so much on Archilochos as on Hesiod, who cannot very well have lived much later than in the last third of the eighth century. The tradition about the death of Amphidamas in the Lelantine war (infra, p. 107 f.)—Plutarch, , Sept. Sap. Conv. 10, p. 153EGoogle Scholar; Schol, . Hes. Opp. 648, p. 304Google Scholar Gaisford (see Hesiodi Carmina rec. Jacoby, i. 119)—is possibly a corroboration of this date. Hesiod's poems were well known to Archilochos.

^{page 101 note 4} p. 34 f.

^{page 101 note 5} See p. 108, n. 1.

^{page 102 note 1} Infra, p. 103.

^{page 102 note 2} Athen. 4. 63, p. 167D τοιο⋯τος ⋯γ⋯νετο κα⋯ Αἰθ⋯οψ ⋯ Κορ⋯νθιος, ὥς ϕηοι Δημ⋯τριος ⋯ Σκ⋯ψιος, οὖ μνημονεὑει Ἀρχ⋯λοχος (F 145 Bgk.) ὑπ⋯ ϕιληδον⋯ας γ⋯ρ κα⋯ ⋯κρασ⋯ας κα⋯ οὖτος μετ' Ἀρχ⋯ου πλ⋯ων εἰς Σικελ⋯αν, ὅτ' ἔμελλεν κτ⋯ζειν Συρακο⋯σας, τ⋯ι ⋯αυτο⋯ συσσ⋯τωι μελιτοὑττης ⋯π⋯δοτο τ⋯ν κλ⋯ρον, ὅν ⋯ν Συρακο⋯σαις λαχὼν ἒμελλεν ἒξειν.

^{page 102 note 3} For the proverb ⋯ το⋯ ἒνου δ⋯ψος, where the ass barters away the Zeus-given ϕ⋯ρμακον γ⋯ρως ⋯μυντ⋯ριον for a drink of water, Aelian, , N A 6. 51Google Scholar, quotes the Sicilians Ibykos and Deinolochos, Sophokles and two comic poets. Archilochos illustrates the same sort of mind by an historical example. Thousands of men retold or alluded to the story of Esau (Gen. xxv. 27–34), who sold his birthright for a mess of pottage. I am sure there are more parallels in the collections of the Greek proverbs.

^{page 102 note 4} The date given by Xanthos we cannot recover, as we do not know how he expressed it. We are in the same predicament for the dates of Hellanikos, Theopompos, and Euphorion (p. 99, n. 1; p. 103, n. 1), or those of Demeas (p. 98, n. 1), and many others. Probably he gave the year of a Lydian king, which the source of Clement converted into an Olympiad, which is of no use to us. Dionysios gives another Olympiad, which makes it quite clear that either there was no real tradition (PhU xvi. 144. 3) or the chronographers were not able to fix the Parian eponym. The tradition about the colonization is also rather slight: the founder, according to an oracle (Oinom. Euseb, . PE 6. 7. 8Google Scholar; Steph. Byz. s. Θ⋯σος) is one Telesikles, father of the poet in the epigram on the Parian Archilochos monument. Thasian tradition, if tradition there was, we should expect from the Thasian Polygnotos (Pausan. 10. 28. 3), who, strictly speaking (and Blakeway, p. 49. 2, stresses the point), did not paint the colonization, but the introduction of the Demeter-cult by Kleoboia, wife of Tellis, who also appears in the picture ⋯λικ⋯αν ⋯ϕ⋯βου γεγονώς. This Tellis is, according to the source of Pausanias (Polemon)—he says ἢκουσα—grandfather of the poet. Accordingly Blakeway thinks of pre-colonization contacts between Paros and Thasos, while Hiller von Gaertringen believes that Polemon wrongly differentiated between Tellis and Telesikles. As to the date, Blakeway, p. 53, puts the colonization at 720 B.C., the date of Dionysios, which is rather early, to judge from the excavations at Limenia (see Blakeway 51. 1 on the authority of Miss E. Haspels: ‘The French excavations have produced a fair quantity of Greek pottery of the first half of the 7th century <see A.A. 47, 165; 48, 244>, and a little which can perhaps <!> be referred to the end of the 8th’); Hiller at 682/0, which seems to be too late and is founded on his more than doubtful supplement of Marm. Par. A 33. If the ancient dates are about right, that is, if Thasos was colonized between 720 and 700, the leader of the colonists may very well have been the grandfather of the poet. It should be remembered that the oracle says nothing about a son of the founder; it simply tells Telesikles to inform the Parians of its order to colonize Thasos. I have not much doubt as to the authenticity of the oracle; the attendant circumstances are of no great interest, and before Oinomaos nobody knows anything about the participation of Archilochos. Blakeway's compromise— ‘there is no need to take Παρ⋯ους ⋯ξεν⋯γησεν as implying that Oinomaos thought that Archilochos was leader of the colony; ⋯ξɛν⋯γησɛν means no more than ‘acted as a guide’—seems to me neither helpful nor probable.

^{page 103 note 1} Euseb, . PE 6. 7Google Scholar. 8 (text of the oracle) ⋯γγελ⋯, ν⋯ Δ⋯α (ϕ⋯σει τ⋯χα ποὺ τ⋯ς τετυϕωμ⋯νος ἥ σ⋯⋯λ⋯γγχων), κἃν μ⋯ κελεὺσηις π⋯πρωται γ⋯ρ. κα⋯ ἒστι Θ⋯σος μ⋯ν ⋯ Ἠερ⋯α ν⋯σος ἣξουσι δ' ⋯π' αὐτ⋯ν Π⋯ριοι, Ἀρκιλ⋯χου το⋯ ⋯μο⋯ υἱο⋯ ϕρ⋯σαντος ⋯τι ⋯ ν⋯σος αὕτη πρ⋯ν Ἠερ⋯α ⋯καλεῖτο. σὺ οὖν, δειν⋯ς γ⋯ρ ⋯ξελθεῖν, οὐκ ⋯ν⋯ξηι (οἶμαι) αὐτο⋯ οὒτως ⋯ντος ⋯χαρ⋯στου κα⋯ ϕρασ⋯ος, ὃς, εἰ μ⋯ σὺ μηνὺσαι αὐτ⋯ι ⋯βουλ⋯θης οἅκ ἄν ποτε ἢγγειλεν, οὐδ' ἄν Ἀρχ⋯λοχος ⋯ υἱ⋯ς αὐτο⋯ Παρ⋯ους ⋯ξεν⋯γησεν, οὐδ' ἄν οἱ Π⋯ριοι Θ⋯σον ⋯ικησαν.

^{page 103 note 2} This conclusion of Blakeway rests, as far as I understand it, on a combination of Oinomaos' ⋯ξεν⋯γησεν and the dates given by Clement—two very different premises. In fact, Clement does not say anything about a participation of Archilochos in the colonization of Thasos, but his source, in trying to determine the date of the poet (see p. 99, n. 1), takes the colonization as a terminus post for the birth of Archilochos, who—the poems show and Kritias confirms—was not a child when he wrote the Thasian poems: hence γνωρ⋯ζεσθαι. Even if Archilochos wrote a poem such as Blakeway imagines, does not follow that he was an adult at the of the colonization. There were, surely, many opportunities in the Thasian poems or elsewhere, to mention the foundation (see e.g. F 54?), either shortly, as he mentioned the Μαγν⋯των κακ⋯ (See p. 107), or even at some length in the elegies. But we have no authority for it, and I am sure that Kal if he had given more than the name we hear of it: the ancient grammarians looked for historical facts in the poets very carefully. Strabo 10. 5. 7 knows nothing but the bare of the Parian colonization, and Pausanias (1028. 3) only knows what the picture by Polygnotus gave; to explain it he has recourse an ἤκουσα, and what he ‘heard’ refers only to the genealogy of the poet. I personally not believe in a more or less historical poem Archilochos.

^{page 103 note 3} Aelian, , VH 10. 13 = Vorsokr. 88 [81] B 44Google Scholar εἰ γ⋯ρ μ⋯ (ϕησιν) ⋯κεῖνος τοιαὺτην δ⋯ξαν ὑπ⋯ρ ⋯αυτο⋯ εἰς τοὺς Ἔλληνας ἢνεγκεν, οὐκ ἂν ⋯πυθ⋯μεθα ⋯μεῖς οὔτε ὅτι Ἐνιπο⋯ς υἱ⋯ς ἦν τ⋯ς δοὺλης οὔθ' ὅτι καταλιπὼν Π⋯ρον δι⋯ πεν⋯αν κα⋯ ⋯πορ⋯αν ⋯λθεν εἰς Θ⋯σον οὔθ' ⋯τι ⋯λθὼν τοῖς ⋯ντα⋯θα ⋯χθρ⋯ς ⋯γ⋯νετο κτλ. Cf. Pindar, P 2. 54 εἶδον γ⋯ρ ⋯κ⋯ς ⋯ὼν ⋯ν ⋯μαχαν⋯αι | ψογερ⋯ν Ἀρχ⋯λοχον βαρυλ⋯γοις ἒχθεσιν πιαιν⋯μεν.

^{page 104 note 1} Then we should have to ask whether any of the passages quoted could refer to Magnesia at the Sipylos, which I do not believe. As it is, I have simply collected all events which Archilochos might possibly have thought of. Buerchner's article on Magnesia on the Maeander, (RE xiv. 471, no. 2Google Scholar) is, as usual, quite insufficient. The town at the Sipylos is usually distinguished from the Maeandrian by an epithet (Hellanikos 4 F 191 and others).

^{page 104 note 2} F 55. Even if the Μαγν⋯των κακ⋯ were already used proverbially in Archilochos' time (Hudson-Williams, , The Elegies of Theognis, p. 262Google Scholar; Geffcken, , Gr. Lit.-Gesch. i, 1926, 72Google Scholar, and others), it certainly would not mean that we can bring down ‘indefinitely’ the poem in which he uses the phrase. So far I agree with Blakeway, p. 45. 3; and the more so, as not even Theognis 603/4 and 1103/4 are sufficient proof for proverbial use: they adapt, after the manner of the ‘Theognideans’ an elegiac poem by Kallinos to a moral maxim. They do not restrict themselves to Magnesia, but (1103/4) join with it other towns famous or infamous for luxurious living. The real question is not how far to bring the poem down, but how far the event alluded to by Archilochos may be brought up. I feel confident that it was not a ‘Queen Anne is dead’ or ‘Charlemagne has lost a back tooth’ sort of allusion; and again I find myself in virtual agreement with Blakeway, who believes ‘that Archilochos' poem followed soon after the event’ and protests against an interpretation like the one sometimes proposed: ‘I weep for the miseries that beset my own time and country, not for proverbial sufferings of far away and long ago.’ In such an interpretation the ‘long ago’ at least is, in fact, a gratuitous addition.

^{page 104 note 3} NH 35. 55 quid? quod in confesso perinde est Bularchi (-achi B -iarchi V) pictoris tabulam, in qua erat Magnetum proelium, a Candaule rege Lydiae Heraclidarum novissimo, qui et Myrsilus vocitatus est, repensam auro? tanta iam dignatio picturae erat. circa Romuli id aetatem accideril necesse est: <et>enim duodevicensima olympiade (708/4) inieriit Candaules aut, ut quidam tradunt, eodem anno quo Romulus (716), nisi fallor, manifesta iam tunc claritate artis, adeo absolutione. 7.126 Candaules rex Bularchi picturam Magnetum exitii, hand mediocris spatii [pari (del. Mayhoff)] rependit auro.

^{page 104 note 4} The true date for Kandaules seems to dispose of some doubts about the authenticity of the whole story, as ‘a Greek “battle-piece” picture in the first quarter of the 7th century is archaeologically perfectly possible’ (Blakeway, p. 46, quoting Payne, , Protokorinth. Vasenmalerei, pl. 10Google Scholar, nos. 1, 5, 6). Evidence for the third (hardly for the second) quarter is extant. Archaeology, as Paul Jacobsthal informs me, (Pfuhl, , Malerei und Zeichnung, §§ 528, 535, 538Google Scholar) can hardly decide whether proelium or exitium is right. If the pinax was worth the gold king Kandaules paid for it, it was probably not so poor a picture as those on the Protokorinthian aryballoi of 700–675 B.C. to which Blakeway refers, even if it was just a fight. If it was the sack of Magnesia, Bularchos could easily buildit up with the material provided by the Phoenician silver bowls and the like which were well within his horizon: on them one sees not only lively battle-scenes but also pictures of beleaguered towns, etc. Should not the fight on the Chigi vase, painted in the third quarter of the century, follow the same models rather than reflect large wall-paintings, as Payne 1. 1. 14, suggests?' But the rules of Textkritik obviously favour the proelium of the main passage against the exitium in the short and probably additional notice in the 7th book. I do not propose to change exitium into exercitum because I believe that Pliny (in hastily adding to the section artibus excellentes this oldest example from the main passage in book 35 or in abbreviating the extract he had made for it) confused the motif of the picture with the much better known and often quoted Μαγν⋯των κακ⋯—surely an excusable error. Blakeway, of course, prefers exitium, which he does ‘not for a moment believe wrong’—‘strong words, strong words sir, said the black gentleman’. He does not weigh the claims of proelium and the probability of the main passage being more trustworthy for the detail, but spurns logic with the assertion that ‘even if exitium is wrong, we have this body of evidence for Woes of Magnesia before the accession of Gyges’. There is no ‘Woe’, if proelium is right, which possibly even means a victory. I am very much inclined to believe that Bularchos painted the Amazon war, which afterwards Magnes sang (infra, n. 2); and there is nothing in our evidence to support a destruction of Magnesia by Kandaules or anybody. On the contrary, Strabo 14.10.40 (see p. 99, n. 2) knows only one συμϕορ⋯, which means the same event as ⋯πώλεια in the Clement passage (see p. 99, n. 2). Perhaps he even positively excludes a former destruction: εὐτυχ⋯σαντος, said of the Treres, is rather curious; much better the εὐτυχ⋯σαντος πολὺν χρ⋯νον of codex F, referring to ουν⋯βη τοῖς Μ⋯γνησιν ἄρδην ⋯ναιρεθ⋯ναι and looking forward to the following εὐτυχοὺντων. It is grammatically tolerable, and I prefer it to the obvious correction into εὐτυχ⋯σασιν the accusative would explain the corruption into the seemingly logical εὐτυχ⋯σαντος. It is a conclusion e silentio, and I only give it for what it is worth. But in any case, Clement, who, according to Blakeway ‘obviously thought of the destruction of Magnesia as being not far removed from 700 B.C.’ (the italics are mine) cannot be claimed as supporting an exitium at so early a time; and Blakeway himself (p. 50 f.) is not happy about the text which, nevertheless, he used also for reconstructing the poem about the colonization of Thasos.

^{page 105 note 1} FGrHist 90 F 62; Commentary ii C, pp. 233 ff.; 239 ff.; 244 ff. If Blakeway had read the Fragmente—but the pertinent fact was known long before, viz. that Nikolaos drew on a good source of a clearly epichorial character and had access to the Lydian court journals which gave detailed accounts about the exploits of the kings—he would, perhaps, not have relegated this piece of evidence to a note, as being ‘in itself, without support from a better authority, not of much value’. But his suggestion of filling up one of the many gaps in the Lydian history of Herodotos (1.14. 4) from it (not so much from Theognis 1103/4) seems to me a sound one.

^{page 105 note 2} One might be inclined to doubt the name of the Smyrnaean poet, as the Excerpta are rather unreliable in this respect and full of confusions. I do not believe in what Crusius, , RE ii. 489Google Scholar, tries to elicit from the vague words of the Latin metrist Juba, De Archilocho Pario, qui Gygae fabulam optime complexus est (briefly but rightly rejected by Bergk, , PLGr ii.⁴390Google Scholar): Archilochos and a poem of his making fun of the Magnesian (sic) colleague, as the source of Nikolaos.

^{page 105 note 3} Incidentally, as we know from F 22 that Archilochos lived in (or after) the reign of Gyges, and as the Nikolaos war is not assigned to a certain year or period in this reign, it would, in any case, not add much to our knowledge of the poet's date.

^{page 105 note 4} 14. 1. 40. The source is obviously the same passage of Kallisthenes (124 F 29) which he quotes, 13. 4. 8, for the captures of Sardes by Cimmerians, Treres, and Lycians. The passage shows that Kallisthenes was acquainted with the elegies of Kallinos, and that long before Strabo the grammarians (Demetrios of Skepsis is quoted here) discussed the bearing which the historical and chronological questions of the Cimmerian invasions had upon the relative date of the poets who mentioned them. Kallisthenes dealt with ancient history of the Greek cities in Asia Minor in his Ἀλεξ⋯νδρον πρ⋯ξεις. For an episode from this or these wars see Aelian, , VH 14. 46Google Scholar.

^{page 106 note 1} Athen. 12. 29, p. 525c; Strabo, 14. 1. 40 (see p. 104, n. 4).

^{page 106 note 2} F 2–4 D. Σμυρνα⋯ους(= Ἐϕεσ⋯ους) δ' ⋯λ⋯ησον | μν⋯σαι δ' εἴ κοτ⋯ τοι μηρ⋯α καλ⋯ βο⋯ν <Σμυρναῖοι κατ⋯κηαν> | ν⋯ν δ' ⋯π⋯ Κιμμερ⋯ων στρατ⋯ς ἒρχεται ⋯βριμοεργ⋯ν | Τρ⋯ρεας ἄνδρας ἄγων. There are striking likenesses, not noted by Hudson-Williams, to this Λ⋯γος πρ⋯ς Δ⋯α. (Strabo 14. 1. 4) in the prayers with which the second ‘Theognis’ of Megara (Jacoby, , Sb. Berlin, 1931, 138 ff.Google Scholar) opens his book: Ζεὺς μ⋯ν τ⋯σδε π⋯ληος ὑπειρ⋯χοι(757)—Φοῖβε ἄναξ … αὐτ⋯ς δ⋯ στρατ⋯ν ὑβριστ⋯ν Μ⋯δων ⋯π⋯ρυκε τ⋯σδε π⋯λευς, ἴνα σοι λαο⋯ … κλειτ⋯ς π⋯μπωσ' ⋯κατ⋯μβας (773 ff.) κτλ.

^{page 106 note 3} Athen. 12. 29 ⋯πώλοντο δ⋯ κα⋯ Μ⋯γνητες οἱ πρ⋯ς τ⋯ι Μαι⋯νδρωι δι⋯ τ⋯ πλ⋯ον ⋯νεθ⋯ναι, ὥς ϕησι Καλλῖνος ⋯ν τοῖς Ἐλεγε⋯οις. This elegy ‘Theognis’ condenses in his couplet 603/4 (see p. 104, n. 2). It is astonishing that even Bergk did not see that this is a fragment of Kallinos, but only quotes the passage in his note on F 3. On the luxury (?) of the Magnesian aristocracy Aelian, , VH 14. 46Google Scholar.

^{page 106 note 4} They were attacked by Lygdamis, and the great temple of Artemis was burnt down (Kallimachos, , H. 3. 251 ff.Google Scholar; Hesych. s. Λὺγδαμις). Is he the subject of Kallinos F 4 Τρ⋯ρεας ἄνδρας ἄγων? See Strabo 1.3.21 Λὺγδαμις δ⋯ τοὺς αὐτο⋯ ἄγων μ⋯χρι Λυδ⋯ας κα⋯ Ἰων⋯ας ἤλασε κα⋯ Σ⋯ρδεις εἶλεν ⋯ν Κιλικ⋯αι δ⋯ διεϕθ⋯ρη.

^{page 106 note 5} The evidence (see Busolt, , Gr. G. ² ii. 461 ff.Google Scholar) is slender for all detail, and, to judge from Strabo, there was not much more to be got from the complete elegies of Kallinos, which the grammarians used for determining the relative chronology of Kallinos and Archilochos. Obviously there was nothing at all to be got from the latter; if there had been more than the short allusion to the Magnesian Woes, the verses would have been quoted in the discussion.

^{page 106 note 6} It is curious what different standards Blakeway applies to the sources: he finds no difficulty in crediting the learned Strabo (who, in fact, represents here Kallisthenes and the best Hellenistic science) with an incredible confusion and extreme carelessness, but he cannot bring himself to believe (p. 55) that Clement, who in all secular subject-matter is a mere compiler, ‘was mistaken in his dating of Archilochos' reference to the fall of Magnesia’. In fact, Clement dated it with Strabo.

^{page 107 note 1} There is no terminus ante; for the ⋯ξ⋯ς ἒτος in the corrupt text of Strabo 14. 1. 40 is of no use. When in 545 Mazares, the general of Kyros, subjected Ionia to the Persian sway, Magnesia was one of the cities captured by him (Herodt. i. 161).

^{page 107 note 2} Blakeway, p. 46.

^{page 107 note 3} This result does not even make it probable that he was still in Thasos at the time of the eclipse of 648 B.C. (see p. 108, n. 3).

^{page 107 note 4} P. 53 f. The Gyges poem is not dated in the table. In Blakeway's opinion it should be the last or one of the last efforts of Archilochos; and the death of the poet, which he dates (though his reasons are not clear to me) ‘before 670/60 B.C.’, should follow almost immediately. That would at least give a more probable age—say between 45 and 55, and one might as well choose the lesser number—than the ‘not more than 65’ (which it is rather generous not to deem ‘old’), for the man who (as Blakeway and many others believe) was slain in a battle. And it would not damage Blakeway's line of argument at all; rather the contrary.

^{page 108 note 1} As Blakeway has it, p. 47 (cf. p. 53): ‘to endeavour to extract chronological data for Archilochos from the “Lelantine War” may seem to some historians like an attempt to illuminate twilight by complete darkness. For them the only certain chronological datum for that war is the mention of it by Archilochos’ (the italics are mine). Like many others he is under the influence of an age-long discussion (this influence makes itself very curiously felt, e.g. in Crusius', article, RE ii. 495Google Scholar), and he does not quite seem to realize that he is on the whole merely returning to an opinion prevalent about 50–100 years ago. On the other hand, the reference of F 3 to the Lelantine war seems to receive an appreciable corroboration by Dr. Bowra's most attractive interpretation of F 56 (Class. Rev. liv, 1940, 127Google Scholar. When almost forty years ago I gave my first lecture on Greek elegy, I called this poem ‘a prophecy of coming war’, but did not put Bowra's pertinent question, where that war may have been): Γλα⋯χ' ὃρα. βαθὺς γ⋯ρ ἤδη κὺμασιν ταρ⋯σσεται | π⋯ντος, ⋯μϕ⋯ δ' ἄκρα Γυρ⋯ων (Xylander γὺρεν, γὺρεν MSS.) ⋯ρθ⋯ν ἴσταται ν⋯ϕος, | σ⋯μα χειμ⋯νος. κιχ⋯νει δ' ⋯ξ ⋯ελπτ⋯ης ϕ⋯βος. There remains, it is true, a slight doubt, which I shall not develop here. But even granted Dr. Bowra's explanation (and I for one rather like it) the same objection holds good as against F 3: is it ‘the’ Lelantine war Archilochos means by the cloud arising from Euboia? I do not think we have a right to credit Archilochos with the political foresight to see that this war would be an almost panhellenic war. Any feud would do for him.

^{page 108 note 2} See Busolt, , Gr. G. ² i. 457Google Scholar; Ed. Meyer, , GdA ii. 231AGoogle Scholar. The help of Thessalian horsemen: Beloch, , Gr. G. ² i. 1. 339Google Scholar.

^{page 108 note 3} One should not forget that there is an alternative. Kritias (p. 103, n. 3) does not help, as he only says that in Thasos too the poet τοῖς ⋯ντα⋯θα ⋯χθρ⋯ς ⋯γ⋯νετο, and does not State for what destination he eventually left the island. I believe that he returned to Paros, and there and then had his affair with Neobule. The eclipse poem would have been written in Paros; but it is not quite certain that he saw the eclipse there. That Archilochos died in Euboia is, to be sure, not impossible, but it is mere guesswork: the man who slew him was a Naxian (that at least the evidence is agreed upon), and the Demeas chronicle knows of feuds between Paros and Naxos.

^{page 108 note 4} As is the case with F 56, if we take it with Dr. Bowra as allegorical.

^{page 108 note 5} P. S3 with the alternative that ‘it was in progress or about to take place’. From the data collected p. 47 f. one would expect that the war filled about the last third of the eighth century; but Blakeway has to reckon with his ‘conjectural date’ of Archilochos' birth, which is 740/30. The opinion of the historians as to the date of the war is as divergent as can be, and Cary, , CAH. iii, 1925, 622Google Scholar (who is himself content with a vague ‘in the 8th or 7th century’), declares ‘a more precise dating, such as modern scholars attempted on the basis of some very uncertain literary allusions’ to be ‘hardly possible with the means at our disposal’. Ed. Meyer, , GdA ii, 1893, § 342Google Scholar, puts ‘the culminating point’ of the war in the middle of the seventh century, Beloch, , Gr. G. ² i. 1 (1912) 338 fGoogle Scholar. at about 670. The most cautious view is developed by Busolt, , Gr. G. i ², 1893,456Google Scholar; but his result that ‘the main Archistruggle was probably at the end of the 8th and in the first half of the 7th century’ is not very helpful. I am personally inclined to agree with Blakeway. But Thukyd. 1. 13. 3 is not really decisive (what about 13. 4 ? And 15. 3 seems to imply that there were already several fleets in existence at the time of the war); and I cannot dismiss either Herodotos (6. 127. 4) as lightly as Ed. Meyer does or Theognis 891/4. But the whole question is really of no importance for Archistruggle lochos: we cannot date him from the war nor the war from him.