Mediæval Aldgate had two round-fronted towers; these are shown in the Survey of Holy Trinity Priory mentioned above, and they are so similar to the bastions of the wall that I was led to suggest that the double gateway and towers were probably substantially Roman work (Fig. 40). Some confirmation of this is given in _V.C.H._, but compare _Archæologia_, xliii. Fitzstephen, writing at the end of the twelfth century, says that London had “double gates,” and this was doubtless so from Roman days. [Illustration: FIG. 40. ] The Roman ditch outside Aldersgate, with a foundation for a bridge pointing towards the gate, was found about thirty years ago, and this is evidence for a Roman gate on this site (_Archæol._ lii.). Ludgate is guaranteed as Roman by the antiquity of the Strand and Fleet Street. Stow says that in 1595 he observed on the north side of Fleet Street from Chancery Lane to St. Dunstan’s Church, 4 ft. below the surface, “a pavement of hard stone, more sufficient than the first, under which they found in the made ground piles of timber almost close together, the same being black as pitch and rotten, which proved that the ground there, as sundry other places of the City, had been a marsh.” Close piling was such a common Roman procedure that it may not be doubted that what Stow observed was the Roman road to Ludgate. Mediæval Aldgate can be restored very fully by comparing the plan mentioned above with the view of the City given by Braun and Hogenberg (_c._ 1550). The gate is so accurately represented that two stair turrets appear over the positions where stairs are shown in the plan. If this gate is so accurately drawn, then the other indications may be accepted. In the Pepys collection, Cambridge, is an engraved view of a gate dated 1688; in the list of contents this is described as Cripplegate, but I believe it is rather Bishopsgate. It was an unaltered mediæval structure, with corbelled battlements and three statues in niches, one on each of the towers and one in the centre. Newgate is also represented in a woodcut view of about the same time, and in an engraving of considerable accuracy, from a book entitled _Herba Parietis_; here even Whittington’s coat-of-arms plainly appears. For a possible view of the Bridge gate, _c._ 1416, see an article by Mr. Weale in the _Burlington Magazine_, 1904. A Roman road on piles has recently been found in Southwark (_Archæol._ lxiii.). Adding the Bridge gate, we now have evidence for the existence in Roman days of the six chief gates of Londinium. It has been suggested that there may have been an earth bank inside the walls, as at Silchester, but the different relation of the fronts of the gates to the walls in London are contrary arguments.