shortest route through the interior of Thrace to Byzantium, from whence he passed into Asia: the interior tribes, unconquered and predatory, harassed his retreat considerably; but we shall find long afterwards Persian garrisons in possession of many principal places on the Thracian coast.[373] It will be seen that Artabazus afterwards rose higher than ever in the estimation of Xerxes. [369] Herodot. ix, 70; Demosthenês cont. Timokrat. p. 741, c. 33. Pausanias (i, 27, 2) doubts whether this was really the cimeter of Mardonius, contending that the Lacedæmonians would never have permitted the Athenians to take it. [370] Herodot. ix, 70: compare Æschyl. Pers. 805-824. He singles out “the Dorian spear” as the great weapon of destruction to the Persians at Platæa,—very justly. Dr. Blomfield is surprised at this compliment; but it is to be recollected that all the earlier part of the tragedy had been employed in setting forth the glory of Athens at Salamis, and he might well afford to give the Peloponnesians the credit which they derived at Platæa. Pindar distributes the honor between Sparta and Athens in like manner (Pyth. i, 76). [371] Plutarch, Aristeidês, c. 19. Kleidemus, quoted by Plutarch, stated that all the fifty-two Athenians who perished belonged to the tribe Æantis, which distinguished itself in the Athenian ranks. But it seems impossible to believe that no citizens belonging to the other nine tribes were killed. [372] Diodorus, indeed, states that Pausanias was so apprehensive of the numbers of the Persians, that he forbade his soldiers to give quarter or take any prisoners (xi, 32); but this is hardly to be believed, in spite of his assertion. His statement that the Greeks lost ten thousand men is still less admissible. [373] Herodot. ix, 89. The allusions of Demosthenês to Perdikkas king of Macedonia, who is said to have attacked the Persians on their flight from Platæa, and to have rendered their ruin complete, are too loose to deserve attention; more especially as Perdikkas was _not then_ king of Macedonia (Demosthenês cont. Aristokrat. pp. 687, c. 51; and περὶ Συντάξεως, p. 173, c. 9).