ROMANO-BRITISH LONDON Fig. It 1 6. — Architectural Fragments from Bastion, HoUNDSDITCH IS be SO and unfortunate it as seems that the description of that the bastion had meagre, that the Roman ditch was here met with •condition. At this point the wall formed the boundary between the houses in Bevis Marks and the yards of those in Houndsditch (Plan C, 19). The base of a bastion was found at ' the north-east of these excavations,' and it is described as ' of later date and rougher, but still probably of Roman work.' It projected i 8 ft. 6 in. from the outer face of the City wall. The width is said to have been as much as 40 ft., but there seems to be some doubt as to this measurement, some other masonry having been built on to it. Its face was ' a flat segment of a circle,' unlike the bold projection of others that have been found. Built up into it were some fragments of Roman architectural work including a circular base of a column, a portion of a column shaft with ■diagonal bands and lo- zenges, and an inscribed «tone (Fig. 16). A massive channel of solid stone, i ft. 6 in. broad and I ft. 3 in. deep, led from the centre of the bastion to the ditch, and ■' traces of a raised earthen bank like an external val- lum to the ditch ' were found these discoveries should some unusual characters. (Plan C, 20). Further along Bevis Marks, the little street now known as Goring Street, which has been cut through the wall into Houndsditch, marks the position of a bastion (Plan C, 21.) It was formerly called Castle Street, and it is sugges- tive that two streets associated with the wall bore this name, and at both of them bastions have been found. One cannot but regret therefore that a name which probably recalls a fact of such interest should disappear. The base of this bastion was uncovered in 1884,^* and its discovery was the occasion of a great outburst of enthusiasm, funds being raised for its explora- tion and an influential committee formed to protect and record London antiquities ; but little further has been heard either of the bastion or of the committee. In a slight notice which appeared at the time, however, it is , said to have been composed of important fragments of Roman sculpture taken from buildings ; a large stone coffin was also discovered. From Bevis Marks the wall passes behind the houses of Camomile Street, forming the boundary of the little graveyard of St. Martin Outwich, and until 1905 some of the old stones were to be seen under the buildings which abutted on the graveyard'^ (Plan C, 22). The houses in Houndsditch to which these belonged were then pulled down, disclosing a good piece of the wall, the base with the chamfered plinth being very perfect. At one part was the most considerable height of the Roman work that has been recorded, there being in the core remains of four bonding courses, the upper one of which was I4jft. above the original ground level. The stonework '* Antiquary, x, 134. " Arch. Ix, 179. 55
