[890] On the antiquity of the horse and chariot, see what is said above, [p. 320, note 3].
[891] Cf. [Pl. LXXXVIII.], from the north wall of the temple of Rameses II. at Abydos.
[892] Abydos temple, N. wall, the Hittite prisoners.
[893] Cf. [p. 5, note 1].
[894] See [p. 34, note 2].
[897] Cappadociæ, later distinguished always from Comana of Pontus.
[898] See pp. [24], [45]. As to the problem of the direction followed by the Persian Posts in later times, we have formed no opinion, and it is beyond our subject. The suggestion made by Prof. Kiepert that it led over by Sebasteia to the valley of the Tochma Su, and so past Malatia, seems to be supported by the fact that no second crossing of the Halys was considered noteworthy in the record. Mr. Hogarth’s summary (Macan’s Herodotus, 1895, vol. ii. App. xiii. §§ 8, 9) in favour of a route by Mazaca and Comana, descending on Samosata (Samsat), satisfies all the conditions, but seems to us to be improbable owing to its difficulties and to a lack of internal evidence of its importance. Prof. Ramsay’s original preference for a route by the Cilician gates is seemingly substantiated by our new evidence of a visible section northwards from Injessu, which corresponds so nearly to that portion of the Royal Road which he has traced on the Phrygian uplands ([Pl. XXIV.]). We do not think the material at present sufficient to solve the problem, which we believe must in any case be attacked upon the lines laid down by Prof. Myres in a paper read before the Roy. Geog. Soc. 1896, in which he attempted to reconstruct the Maps of Herodotus.