140 ROMAN FORTIFICATION. 



walls would add greatly to their strength. The distance of the two 

 walls from each other would, of course, determine the width of the 

 terrace upon the rampart between them ; this was not always the 

 same. Vegetius, as we see, gives 20 feet as a convenient width. 

 Vitruvius, after describing the manner of building the outer wall, 

 adds " moreover the foundation of the substructure on the inner 

 side should be so far from the outer (wall) as to afford sufficient 

 space within that the cohort may stand on the breadth of the 

 rampart for defence as it is drawn up in line of battle." The width 

 of the terrace at Pompeii is about 15 feet. Generally the inner 

 was much lower than the outer wall, though in some cases it was 

 higher, as at Pompeii, and thicker, as in the fortified camp of 

 Saalburg, in the Taunus mountains, near Homburg, the outer wall 

 of which is only five feet thick, whilst the inner is seven feet. 

 (" Lives of the Greeks and Romans," described from ancient 

 monuments, Guhl and Koner.) The inner wall at Dorchester may 

 have been higher and thicker than the outer ; but high or low, 

 thick or thin, the general rule was to make a rampart for the 

 defence of a town with a wall on either side of it and a wide ditch 

 outside ; and the point to which your attention is specially directed 

 is that we have the remains of one wall only at Dorchester. Where 

 was the other 1 The fragment that remains has some appearance 

 of having been part of the inner wall. Many persons will remember 

 the remains of the ditch, the hollow road now filled up, parallel with 

 the walks and some yards from the wall ; the outer wall should 

 have been on the edge of this ditch. We have further evidence of 

 the masonry now standing being the inner wall from the excavations 

 made here by the Dorset Field Club some years ago, when a Roman 

 paved way was found at the foot of the wall on the inside four feet 

 below the surface. It is a great pity the excavations were not 

 carried further to find out the width of the paved way and to obtain 

 conclusive proof that it was level with the ancient town, as we 

 presume it was, and therefore answered to the broad way in 

 stationary camps upon which large bodies of troops could be 

 manoeuvred, and along which they could be sent to any point of the 



