brick, has crumbled, and though I can remember nothing upstanding and
patently of the Empire between the gate of Rheims and the frontier of
Artois, yet one feature--the Roman road--is here so evident, so
multiple, and so enduring that it makes up for all the rest.

One discovers the old roads upon the map, one after the other, with a
sort of surprise. The scheme develops before one as one looks, and
always when one thinks one has completed the web another and yet another
straight arrow of a line reveals itself across the page.

The map is a sort of palimpsest. A mass of fine modern roads, a whole
red blur of lanes and local ways, the big, rare black lines of the
railway--these are the recent writing, as it were; but underneath the
whole, more and more apparent and in greater and greater numbers as one
learns to discover them, are the strict, taut lines which Rome stretched
over all those plains.

There is something most fascinating in noting them, and discovering them
one after the other.

For they need discovering. No one of them is still in complete use. The
greater part must be pieced together from lengths of lanes which turn
into broad roads, and then suddenly sink again into footpaths, rights of
way, or green forest rides.

Often, as with our rarer Roman roads in England, all trace of the thing
disappears under the plough or in the soft crossings of the river
valleys; one marks them by the straightness of their alignment, by the
place names which lie upon them (the repeated name Estree, for instance,
which is like the place name "street" upon the Roman roads of England);