It seems impracticable, from the existing evidence, to ascertain how great a deduction ought to be made from the calculations of Herodotus, as to the numbers of the invading army; but it is easy to believe that his authorities, which unquestionably were authentic in what relates to the description of the forces, might lead him astray, without any fault on his part. Or probably, as the numbers exceeded the facilities of common computation, some conjectural mode of calculation was adopted by the contemporary Greeks, which might easily exceed the truth.--For example, the length of time occupied by the barbarian train in passing certain defiles:--or the very fallacious mode of reckoning employed by the Persians was perhaps followed:--this, as Herodotus describes it, consisted in counting ten thousand men, who were packed in a circle as closely as possible, and a fence formed round them: they were then removed, and the entire army, in turns, was made to pass within the inclosure: the whole was thus counted into ten thousands. But how probable is it, that, by the inattention of the persons who conducted this process, the successive packages were less and less dense.--Seven thousand men might easily seem to fill the space in which ten had been at first crammed. Nor is it at all safe to argue _à priori_ on the supposition that so many could not have been supported on the march. The power which drew a large levy of men from twenty-nine nations, might also drain those nations of their grain. A vast fleet of flat-bottomed barges attended the army along the coast; and as soon as this fleet was separated from it, all the extremities of famine were suffered by the retreating host. This armament is not fairly compared with those which, in later times, have traversed the continent of Asia; for in these instances the aid of an attendant fleet was not available. Without this aid the distant movement of five hundred thousand men is scarcely practicable; with it, three or four times that number might with little difficulty be led a distance of three or six months’ march. This important difference has not been duly regarded by those who have discussed the question. If then such a deduction from the army of Xerxes is made as may readily be accounted for from the inaccurate mode of computation employed by the Persians or the Greeks; and if the attendance of so large a fleet of store ships is considered, we may well hold Herodotus excused from the charge, either of deliberate falsification, or of intended exaggeration.