bviously his interest to make the most of his labor,
his projects, and his performance. Even his contemporary, though elder,
Southey, the hardest-working and the most scrupulously honest man of
letters in England who could pretend to genius, seems constantly to have
exaggerated the idea of what he could perform, if not of what he had
performed in a given time. The most definite statement of Balzac's that
I remember is one which claims the second number of _Sur Catherine de
Medicis_, "La Confidence des Ruggieri," as the production of a single
night, and not one of the most extravagant of his nights. Now, "La
Confidence des Ruggieri" fills, in the small edition, eighty pages
of nearer four hundred than three hundred words each, or some thirty
thousand words in all. Nobody in the longest of nights could manage
that, except by dictating it to shorthand clerks. But in the very
context of this assertion Balzac assigns a much longer period to the
correction than to the composition, and this brings us to one of the
most curious and one of the most famous points of his literary history.

Some doubts have, I believe, been thrown on the most minute account of
his ways of composition which we have, that of the publisher Werdet. But
there is too great a consensus of evidence as to his general system to
make the received description of it doubtful. According to this,
the first draft of Balzac's work never presented it in anything like
fulness, and sometimes it did not amount to a quarter of the bulk
finally published. This being returned to him from the printer in
"slip" on sheets with very large margins, he would set to work on the
correction; that is to say, on the practical rewriting of the thing,
with excisions, alterations, and above all, additions. A "revise"
being executed, he would attack this revise in the same manner, and not
unfrequently more than once, so that the expenses of mere composition
and correction of the press were enormously heavy (so heavy as to eat
into not merely his publisher's but his own profits), and that the last
state of the book, when published, was something utterly different from
its first state in manuscript. And it will be obvious that if anything
like this was usual with him, it is quite impossible to judge his actual
rapidity of composition by the extent of the published result.

However this may be (and it is at least certain that in the years above
referred to he must have worked his very hard