[482] Vol. III. 287, 438. Even after the discussions of Gelzer ("Rheinisches Museum," 1875, s. 264 ff.) on the date of the eclipse, I believe that Oltmann and Bailly's calculation may hold good for it, until it is proved astronomically that in the year 610 B.C. an eclipse of the sun would not have been visible in Asia Minor. If this were proved, Herodotus' dates for Cyaxares, who not only in his work but on the evidence of the inscription of Behistun, was the founder of the Median empire, would have to be thrown back more than half a century, which the date of Cyrus does not allow. To assume a confusion of Cyaxares with Astyages in Herodotus, is impossible, for Cyaxares is twice expressly mentioned (1, 74, and 103), and moreover Astyages is spoken of as the son of Cyaxares to whom Aryanis was married. Nor can I regard it as finally proved that the double capture of Sardis rests simply on Callisthenes, and a deduction from Strabo. Gelzer agrees that the incursions of the Cimmerians into Asia Minor and their establishment in Cappadocia must be placed at the least before the year 705 B.C. ("Z. Aegypt. Sprache," 1875, s. 18); the devastation of Phrygia by the Cimmerians he puts in the year 696 or 676 B.C. According to the dates of Eusebius Midas (the husband of Damodike) began to reign in Olymp. 10, 3 = 738, and took his own life in Ol. 21, 2 = 695 (Euseb. ed. Schöne, 2, 82, 85); his reign extended therefore from 738 to 695 B.C. Hence the devastation of Phrygia by the Cimmerians must have taken place in the year 695. If they were masters of Phrygia at this date, it is not easy to see why these successes did not carry them on into Lydia. As a fact, this is far from improbable; and if the image at Nymsi is their work, they would not have had any time for it in 630 B.C., for that incursion was merely a "plundering raid," and the change in the dynasty of Lydia, the accession of Gyges in the year 689 B.C. (Vol. III. 416), seems to me to point to some previous violent change. Besides, Strabo's words, p. 61, and p. 647, are plain and conclusive enough, so that I see no reason to attach much weight to the interpretation of the passage, p. 627. Cf. Cæsar, "Ind. lect. Marb. Sem. aestivum," 1876. [483] Vol. III. 284 ff., 291. [484] Diod. 2, 34; Nicol. Damasc. fragm. 66. [485] Nicol. Damasc. fragm. 9, 10, ed. Müller; Diod. 2, 33; Ctes. Fragm. 52, ed. Müller. [486] Roxane and Roxanace are both formed from the old Bactrian _raokshna_. Müllenhoff, "Monatsberichte Berl. Akadem." 1866, s. 562. [487] Ctes. Fragm. 25-28, ed. Müller; Nic. Damasc. fragm. 12 ed. Müller. [488] Oppert gives the form of the second version as Çakuka Iskunka. [489] Choerilus in Strabo, p. 303; Herod. 3, 93; 7, 64; 9, 71; Ptolem. 6, 13; Curtius, 7, 4, 6; Arrian, "Anab." 3, 13; Cf. Plut. "Crassus," 24. [490] Nicol. Damasc. fragm. 66, ed. Müller. [491] "Ind." 1, 1. [492] Diod. 2, 13; 17, 110; Strabo, p. 127. [493] Diod. 2, 13. [494] Herod. 1, 98, 99. [495] Polyb. 10, 27.