A HISTORY OF LONDON the map seems also to prove that the town was not laid out in military- fashion. Roman fortified stations such as Colchester or Dorchester (Dorset) retain their main features and can be easily recognized as camps on the Roman model, but London is stated by Tacitus to have been the resort of merchants travelling/^' and had no need of a large garrison to guard the river-passage. Though it is a hopeless task to trace the insulae or blocks of buildings as has been done at Silchester, it is worthy of remark that there is in the heart of London a rectangular space that does not seem to have yielded a single burial. On the west it follows the bank of the Walbrook north-by-east from about London Stone ; and Gresham House is about the centre of the north side. The east side would be in a line with East India Avenue, and the south would approximately coincide with the actual Roman road found in Great Eastcheap. The area inclosed is about that of a legionary camp (50—60 acres), and the vast wall-foundations discovered near the churches of St. Michael and St. Peter on Cornhill correspond well enough to the position of the praetorium or head quarters of a Roman camp. This may be a mere coincidence, but the foundations are described as deeply set, and evidently intended to carry a building of considerable altitude and importance ; and the massive masonry in the centre of Richborough camp may be quoted in illustration. It may be that at the time of the Claudian conquest a legion was posted here to guard the river-passage, but soon passed on to the front, leaving the camp to form the nucleus of London. The low-lying land of Southwark was evidently protected from the tide, for buildings are found close to the river ; but the dimensions of these dykes may have been exaggerated by writers whose remarks with regard to Southwark might perhaps apply to the condition of things in palaeolithic times, but do not explain the facts of the Roman period. According to Mr. Spurrell, who has made a special study of the subject, there is no need to wonder at the early embankments of the Thames. If such works were needed in Roman early times, they were of minor importance in the upper part of the estuary and near London. The height to which we see them rise now is due to the gradual increase from slighter banks, and this increase needs little exertion though regular attention. The vast lake opposite London, spoken of by several writers, ' resolves itself into the supposition of a few inches of water rising over saltings for a few minutes on a few days of the month.' He doubts the existence during the Roman period of tidal marshes or saltings near London or above Erith, and thinks the estuary did not reach as far west as at the present day. Since the Roman occupation the present channel of the river through its alluvium has remained in almost exact relative position with respect to the earthland foot or hard banks from Lambeth to East Tilbury, and certainly so with regard to the more important hards and landing-places on the main stream now existing.^*" Excavations in Southwark have shown that the Roman level was at, or just above. Ordnance datum, and considerably below what is now the high- water mark. At Guy's Hospital Mr. Spurrell records that Roman refuse was "' Annals, xiv, 3 3 . "" Arch. Journ. xlii (1885), 274, 301 ; xlvi, 76 ; xlvii, 43, 170 ; Proc. Geol. Assoc, xi (1891), 224 ; for sections in Bermondsey see Wm. Herbert, Hist, of St. Michael's, 16 (note). 42
