142 ROMAN FORTIFICATION. 



towers right and left." Whether the walls which surrounded 

 Dorchester had or had not towers could only be decided by 

 uncovering the foundations, if they still exist ; as the fortifications 

 are so strong in other respects there is no reason for assuming that 

 they were deficient in this.* 



"The intervals between the towers must be so contrived that 

 one must not be further from another than an arrow's flight, so 

 that if any of them is attacked the enemy may be repelled by 

 scorpions and other pieces of artillery from the towers to the right 

 and left. . . . Also over against the interior sides of the 

 towers the wall must be divided by intervals as wide as the towers, 

 that the footways bridging them may be within the towers, and 

 these must not be fastened with iron, so that if the enemy has 

 seized any part of the wall the defenders will cut it away. If they 

 do this promptly they will prevent the enemy from penetrating the 

 rest of the towers and wall without casting themselves headlong 

 (into the hollow of the tower)." These wooden drawbridges 

 were a very ingenious way of isolating the portion of the wall 

 attacked. Of gates, there were in all probability four at least. 

 We know the position of two of them. The foundations of the 

 west gate were observed at the top of High-street (1) * where they 

 are marked on the Ordnance Survey map ; and no one appears to 

 question that the south gate was at the end of South-street. 



There is a little difficulty about the position of the east gate. 

 Hutchins says " In making the new road (i.e., the portion of the 

 London-road leading out of Dorchester to the east) a little to the 



* " There were probably towers at the corners ; the mounds and the 

 curves which the walls formed there instead of angles which can still be 

 traced, are some evidence of it. It was a common practice to round the 

 corners of the fortifications of Roman camps and towns, and on the 

 mounds within these rounded curves of fortifications, towers were 

 frequently built. The Roman tower which still remains in the fortifica- 

 tions of York is in this position (see Archaeological Journal, vol. 31, 

 p. 226). 



* (1) Dr. Stuckeley, in his map of Durnovaria, dated August 22nd, 1723, 

 represents the Durngate-street as continued through the town, with the 

 west gate at the end of it, on a supposed road to Ischalis, 



