GLASGOW AftCHiEOLOGICAL SOCIETY.
placed upon them ; and (6) that, if these things be so, the guard-
ing of the vallum by artillery, being a great increase to the
power as well as to the economy of the defence, must be reckoned
upon as a factor in the whole design.
There are attractive features about this argument, but
some of its links require strengthening and supplementing
by further facts. Thus, although it goes without saying that
the camps and stations were defended by artillery, stronger
evidence is needed than is yet forthcoming that the vallum
itself was so defended. It is true that the Croy Hill stone
ball was found at a spot where it cannot reasonably be supposed
to have been shot from any camp; but still one ball is a very
slender foundation for whole batteries of artillery. Withal,
however, had that stone been found in conjunction with one of
the expansions or supposed " tribunals," it might have been
more difficult to resist the cumulative force of the induction,
notwithstanding the undoubted force of the two leading
objections stated to it.
It has been urged that it is not easy to see what arrangements
were used for housing and guarding the machines, which could
hardly have been brought on each emergency from the forts.
It is contended that as the ground immediately north of the
outer mound is in many places covered from view and from fire,
the vallum is therefore not adapted to a distant defence by
engines. And it is pointed out that the expansion on section 12a
Croy Hill, to take an instance, is a very ineffective position for
artillery: the cliff in front being nearly inaccessible, so that
machines would have been superfluous, and the possible ascents
not covered by artillery fire.1
The older opinion that these mounds were watch-towers gains
nothing directly from the sections made through them, although
in one case, that at section Croy No. 11, the situation is well suited
for such a purpose. There is no trace of a chambered tower—
nothing but a solid cespiticious mass—in the structure so far as
revealed by the cuttings. There certainly was no stone structure
1 See, however, p. 140, supra, as to the non-use by the Romans of a glacis on
scientific lines. Besides, note extra width of bertn between Croy Hill sections
11 and 12a.