and only dug a fosse afterwards if occasion required it.1
Obviously the outer mound bears closely on the questions
suggested by these classical passages. It proves to demonstra-
tion that as with Josephus, Hadrian, and Vegetius, so with
the makers of the Antonine Wall the ditch was a separable
work, which perhaps the term opus valli on the inscriptions did
not necessarily cover. Our sections leave us unable to conclude
whether the making of vallum and ditch went on side by side
at the same time, or whether one of them preceded the other.
All the sections of the outer mound, indeed, reveal at the base
of it traces more or less continuous of an original surface,
attested by a dark line, the existence of which is scarcely
consistent with the turf having been removed prior to the
depositing of the upcast from the ditch. But even were that
conclusion much clearer and safer than it is, it would still
constitute a very ambiguous basis for further inference.
7. "A Priori" Considerations on the Methods of Defence
of the Rampart.
Vitruvius, as we have already seen,2 indicates the existence of
a recognised standard for the thickness of the rampart defending
a town, viz., a thickness sufficient to allow the easy passage of
two armed men. This standard was conceivably enough appli-
cable also to a frontier rampart as well.
For the defence of an ordinary vallum external obstacles of
various kinds were employed, some of them in the ditch, and
some, no doubt, on both its outside and inside edges as well as
on the wall and battlement itself. Arms of all kinds were
employed by the defenders—sword, spear, bows and arrows,
javelins, and the like. These varied means of defence and
counter-attack were only a part of the warlike resources of the
Roman garrison of a camp. We can hardly err in believing that
what Hyginus, Vitruvius, and Vegetius describe as befitting the
defence of cities and camps was more or less applicable to
1 See p. 31, supra.
2 Supra, p. 129.