takes as must (I think I may say inevitably) occur in
a very large number of compressed statements about matter often in
itself of great minuteness and complexity. I have found some such
mistakes, and I make no doubt that I have left some.

In the process of examination I have had the assistance of two detailed
reviews of parts of the book by two French critics, each of very high
repute in his way. The first of these, by M. Gaston Paris, in _Romania_
(XII, 602 _sqq._), devoted to the mediaeval section only, actually
appeared before my second edition: but accident prevented my availing
myself of it fully, though some important corrections suggested by it
were made on a slip inserted in most of the copies of that issue. The
assistance thus given by M. Paris (whose forbearance in using his great
learning as a specialist I have most cordially to acknowledge) has been
supplemented by the appearance, quite recently, of an admirable
condensed sketch of his own[2], which, compact as it is, is a very
storehouse of information on the subject. If in this book I have not
invariably accepted M. Paris' views or embodied his corrections, it is
merely because in points of opinion and inference as opposed to
ascertained fact, the use of independent judgment seems to me always
advisable.

The other criticism (in this case of the later part of my book), by M.
Edmond Scherer, would not seem to have been written in the same spirit.
M. Scherer holds very different views from mine on literature in general
and French literature in particular; he seems (which is perhaps natural)
not to be able to forgive me the difference, and to imagine (which if
not unnatural is perhaps a little unreasonable, a little uncharitable,
and even, considering an express statement in my preface, a little
impolite) that I cannot have read the works on which we differ. I am
however grateful to him for showing that a decidedly hostile
examination, conducted with great minuteness and carefully confined to
those parts of the subject with which the critic is best acquainted,
resulted in nothing but the discovery of about half a dozen or a dozen
misprints and slips of fact[3]. One only of these (the very unpardonable
blunder of letting Madame de Stael's _Considerations_ appear as an
early work, which I do not know how I came either to commit or to
overlook) is of real importance. Such slips I have corrected with due
gratitude. But I have not altered passages where M. Sche