[3546] Sir J. Evans, Anc. Stone Implements, 1897, p. 586; Worthington G. Smith, Man, the Primeval Savage, pp. 190, 214.
[3547] J. Evans, Anc. Bronze Implements, pp. 95, 158, 174-5, 245, 248-9, 272, 278-81, 303, 312, 321, 327-8, 330, 339, 351, 356, 400-1, 411, 424, 450, 467; Coins of the Anc. Britons, pp. 70, 83, 122, 125, 232; ib., Suppl., p. 559; Guide to the Ant. of the Early Iron Age (Brit. Museum), p. 98. Mr. F. W. Reader (Archaeol. Journal, lx, 1903, p. 213) argues that ‘it is difficult to conceive that if any considerable British town preceded [the Roman] Londinium, all traces of it in the shape of pottery fragments, &c., should ... have been so entirely obliterated’, &c. But the same argument would apply to Calleva, Camulodunum, and other towns which were certainly British.
[3548] See p. 359, supra.
[3549] Cf. Ammianus Marcellinus, xxvii, 8, § 7,—Lundinum, vetus oppidum quod Augustam posteritas appellavit; ib., xxviii, 3, § 7,—Augusta, quam veteres appellavere Lundinum.
[3550] See pp. 600-3, supra.
[3551] B. G., iv, 29-36.
[3552] Le Verrier apud Napoleon III, Hist. de Jules César, ii, 522. See the next footnote.
[3553] T. Bergk (Jahrbücher für classische Philologie, 13 Supplementband, 1884, p. 618, n. 2) remarks that Caesar himself regarded the 24th, not the 26th, of September as the date of the equinox. His authority is, I suppose, Vegetius, iv, 39, who says that the autumnal equinox occurred VIII Kal. Oct.: but, according to Pliny (Nat. Hist., xviii, 25 [59], §§ 220-1), it fell on the 28th of September, and according to Varro (Rerum rust., i, 28, §§ 1-2), on the 27th. Columella (De re rust., ix, 14) places it about the 24th of September (circa VIII calend. Octobris); and the 24th was, according to Mommsen (Die röm. Chron. bis auf Caesar, 1859, p. 301), the date adopted in the Julian calendar. But that date was fixed by the calculations of Sosigenes: what right, then, has Bergk to assume that Caesar regarded it as the date of the equinox in 54 B.C., nine years before his reform of the calendar took effect?
[3554] Napoleon’s reasoning is based upon assumptions, one of which is certainly incorrect, while all are doubtful. We know that Caesar started on his return voyage soon after midnight (B. G., iv, 36, § 3). ‘If,’ says Napoleon (Hist. de Jules César, ii, 180, note), ‘we assume that he had a favourable wind, as he had on his return from the second expedition, and that his voyage lasted nine hours, Caesar would have reached Boulogne about nine o’clock in the morning. As the fleet could only enter the harbour on a rising tide, all that we need do in order to ascertain approximately the date of his return, is to find out on what day of September, 699, there was high tide at that hour at Boulogne. Now in that harbour there is always a high tide about nine o’clock in the morning two or three days before full moon and before new moon. Therefore, as the full moon of September, 699, took place on the 14th of the month, Caesar must have returned to Gaul about the 11th or 12th of September.’
There is no fault to be found with the conclusion (except that it is uncertain), but much with the argument. To begin with, as there had been a full moon on the 31st of August, it is obvious that not the full moon but the new moon of September took place on the 14th of the month. This error, indeed, is immaterial; but Napoleon has no right to assume that Caesar reached Boulogne about nine o’clock in the morning, for the circumstances of his return voyage in the second expedition were totally different from those of the preceding year. In 54 B.C. there was a dead calm (summa tranquillitate, B. G., v, 23, § 6), and the ships were rowed: in 55 they sailed. Moreover, it is untrue that the fleet could only enter the harbour of Boulogne at high tide (see p. 586, supra).