aller than that of Coutances is a fact of less importance than it
would be in England. A characteristic of French architecture is the
constant reproduction of the designs of great churches on a much smaller
scale. This is a thing which we know nothing of in England, where the
parish church and the minster are buildings of two different types, each
of which may be equally good in its own way. The church of Saint Peter
at Coutances, much smaller than that of Saint-Lo, will illustrate this
position. And there are plenty of instances, from graceful miniatures
like Norrey and Les Petits Andelys up to churches of considerable size.
But at Saint-Lo, whatever little outline the church has apart from its
spires it gets from a series of gables along the aisles, something like
those of Saint Giles at Oxford. Inside we have a not very successful
_hallenkirche_, three bodies without a clerestory, Bristol-fashion. Much
of the work is good enough of its kind, and the late stained glass is
worth studying; but, as soon as we leave the west front behind there is
a strange lack of design in the whole building, inside and out.

But Notre-Dame is not the only church at Saint-Lo. Both De Caumont and
Gally Knight have a good deal to tell us about the church of Saint
Cross, which it seems that some antiquaries had carried back to the days
of Charles the Great. _Distinguendum est._ To carry back a piece of
Romanesque of any date to a date too early, but still within Romanesque
times, is a mistake of quite another kind from attributing finished work
of the thirteenth century to Geoffrey of Mowbray in the eleventh. Gally
Knight himself erred more slightly in the same way. He knew very well
that the work at Saint Cross could not be of the eighth century; but he
took it for the eleventh instead of the twelfth. No one can blame him
for that at the time when he wrote. But both Gally Knight and De Caumont
saw some things at Saint Cross which are not to be seen now, and some
things are to be seen now which they did not see. They saw a
twelfth-century church which had gone through some changes and
additions, and they also saw some considerable monastic buildings, of
part of which, a vault with what seems to be a rather classical column,
De Caumont gives a drawing. Here it is, if anywhere, that one would look
for the earlier date of Romanesque. But all outside the church itself
has perished. The church itself has, since De Caumont's visit, been
greatly en