in Britain and in Northern Gaul,
but which is unknown in Rome itself. Here are the familiar layers of
small stones with the alternate ranges of bricks. We enter where the
eastern gate has been, and find a second line of defence, a wall of
earth, square, or nearly so, but with its angles rounded off, with its
single entrance near the south-east angle carefully kept away from
either of the approaches in the outer wall. Within this again is the
fortress itself, again quadrangular, with projections at the angles. The
more finished parts of its walls, the gateways, and the parts adjoining
them, give us specimens of Roman masonry whose vast stones carry us
back, be it to the wall of _Roma Quadrata_ at one end or to the Black
Gate of Trier at the other, and which specially call back the latter in
the marks of the metal clamps which have been torn away. Details must
be studied on the spot or in the works of M. Barbe, which is nearly the
same thing, as they seem to be had only on the spot. But there are not
many remains of Roman work more striking than this, and it is more
striking still if we try to make out its probable history from the
internal evidence, which is all that we have to guide us.

That this fortress does not belong to any early period of the Roman
occupation is clear from its construction, the alternate layers of brick
and stone, and the bricks with wide joints of masonry between them, as
in all the later Roman work. And again, the fact that among the
materials of the fortress have been found pieces of other buildings used
up again might suggest that it was not built till after some time of
change, perhaps of destruction, had come over the city. But it is the
numismatic evidence which clearly parts off the history of the fortress
from the general history of the city. Jublains has no inscriptions to
show, but its numismatic wealth is great. Among the many coins found,
not many are earlier than the time of Nero, and those which there are
are chiefly coins of Germanicus. From Nero to Constantine coins of all
dates are common. It is M. Barbe's inference that it was in Nero's reign
that the place began to be of importance, and that its great temple was
built. But the numismatic stores of the fortress taken by itself tell
quite another story. There, not a coin has been found earlier than
Domitian, nor one later than Aurelian, saving a chance find of two
Carolingian pieces of Charles the Bald and a modern French piece o