"three" into "four," and insert "the Idols of the Theatre" at the end
of the sentence.

And this reminds me (since I shall have so much to do with these
questions of date) to suggest a general caution with regard to them
all; namely, that in the case of fragments like these, the comparison
of isolated passages can hardly ever be relied upon for evidence of
the date or order of composition, or of the progressive developement
of the writer's views; and for this simple reason,--we can never be
sure that the passages as they now stand formed part of the original
writing. The copy of the fragment which we have may be (as there is
reason to believe this was) a transcript from several loose papers,
written at different periods and containing alterations or additions
made from time to time. We may know perhaps that when Bacon
published the ADVANCEMCNT OF LEARNING he was ignorant of some fact
with which he afterwards became acquainted; we may find in one of
these fragments,--say the TEMPORIS PARTUS MASCULUS,--a passage
implying acquaintance with that fact. Does it follow that the
TEMPORIS PARTUS MASCULUS was written after the ADVANCEMENT OF
LEARNING? No; for in looking over the manuscript long after it was
written, he may have observed and corrected the error. And we cannot
conclude that he at the same time altered the whole composition so as
to bring it into accordance with the views he then held; for that
might be too long a work. He may have inserted a particular
correction, but meant to rewrite the whole; and if so, in spite of
the later date indicated by that particular passage, the body of the
work would still represent a stage in his opinions anterior to the
ADVANCEMENT OF LEARNING.

I have felt some doubt whether in printing this fragment, I should