^{page 9 note 1} J.H.S., 1920, p. 89. I cannot follow him farther.

^{page 9 note 2} I speak only of Asia Minor.

^{page 9 note 3} For ‘literalist’ interpreters of Herodotus, his route must, of course, cross the Halys twice. Herodotus himself was no doubt unconscious of this. As Mr. J. A. R. Munro points out to me, he may have imagined that all east-to-west routes in eastern Asia Minor crossed the Halys.

^{page 9 note 4} Hdt. I. 72; II. 34.

^{page 10 note 1} Herodotus (ibid.) defines the parasang as =30 stades, or 3 miles 787½ yards.

^{page 10 note 2} Apud Macan, loc. cit. Cf. E. M. Walker, ibid. p. 303.

^{page 11 note 1} This and other figures will appear large when compared with the map. In Asia Minor, with its varied surface, the map is no guide to the mileage along the roads. Thus the airline from Sardis to Akroenos is 135 miles; the official railway measurement is 186 miles. Iconium to Ancyra is 150 miles on the map; 159 miles (measured by trocheameter) along the road.

^{page 11 note 2} By railway the distance from Sardis to the Gates is 534 miles (521 miles to Bozanty, 13 miles thence by road to the Gates). This figure guarantees die general accuracy of my (independent) estimate of the road mileage.

^{page 11 note 3} Savatra (misplaced on all maps) lay 39 miles east-north-east of Iconium, on the direct road: from Cybistra to Laodicea Combusta. The. direct route from the Gates, westwards, passes through Cybistra, but avoids Iconium.

^{page 11 note 4} By minor adjustments—e.g., by leading the post road directly from Adana to the Gates, omitting Tarsus—the two figures could be made to agree almost exactly.

^{page 11 note 5} My distances are: Samsat–Adiaman, 18 miles; Adiaman–Malatia, 68 miles; Malatia–Darende, 56½ miles; Darende–Görün, 19 miles; Görün–Azizie, 62 miles; Azizie–Mazaka, 52 miles; Mazaka–Tavium, at least 119 miles; Tavium–Halys, 70 miles; Halys-Sardis, 410 miles. Total, 872½ miles. Herodotus' figure is 738½ miles. An additional objection to the route by Mazaka is that its natural continuation westwards is by Archelais or by Cybistra. Kiepert's route, grotesque as it is, is really more logical.

^{page 11 note 6} Hdt. V. S3 fin. To Nisibis, Samosata is 20 or 30 miles nearer than Zeugma, leaving a net balance, for the whole journey, of over 100 miles against Messrs. Hogarth and Walker.