ROMAN FORTIFICATION. 137 



Now in the first place we do not know that Vespasian found it 

 necessary to construct a camp here at all ; if he did it would not in 

 all probability be of a more formidable character than the Roman 

 camp commonly was. According to Vegetius, when the danger 

 was not imminent a camp was entrenched in this way ; a slight 

 ditch was carried round the whole circuit only 9ft. broad and 7ft. 

 deep ; with the turf taken from it a breastwork 3ft. high was 

 formed within the ditch ; when there was reason to fear an attack 

 by the enemy, the camp was surrounded by a regular ditch 12ft. 

 broad and 9ft. deep. A parapet four feet in height was then raised 

 on the side next the camp, with hurdles and fascines properly 

 covered and secured by the earth taken out of the ditch ; the 

 height of the entrenchment was thus 13ft. On the top of the 

 whole strong palisades, which the soldiers carried with them for 

 the purpose, were planted. Spades, pickaxes, wicker baskets, and 

 tools of all kinds were carried by the army for the purpose. 

 (Vegetius " Epitoma rei militaris" Lib. I. cap. 24). 



A second difficulty is the irregular form of the space included 

 within -the rampart which surrounded Dorchester. Roman camps 

 were not always parallelograms, but they were generally of regular 

 form. " The form of the camp," says Vegetius, " must be determined 

 by the nature of the country, in conformity to which they must be 

 rectangular, triangular, or oval." The common form was the 

 rectangle, and there was no reason, with the choice of ground before 

 them, that the Romans (if they had a camp on this spot) should 

 have formed it otherwise. One can scarcely imagine a more orderly 

 and symmetrical arrangement than was to be found in the plan of a 

 Roman camp. When the camp was marked out and the troops 

 marched upon the ground every centurion could march his century 

 straight to the spot it was to occupy ; but how could he do this in 

 a camp with four unequal sides, in which the troops on one side 

 of the pretorian street must be arranged differently to the troops 

 on the other side, and especially if the form of the camp was 

 changed day after day 1 A third, and it seems to me conclusive, 

 proof that the ramparts surrounding Dorset did not previously 



