the reader may be surprised at finding that while often I only give the
date under which the reference will be found, frequently I am able to
quote the page and volume. The explanation is a simple one: two sets of
compositors were generally at work, and two volumes were passing through
the press simultaneously.

In the selection of the text which I should adopt I hesitated for some
time. In ordinary cases the edition which received the author's final
revision is the one which all future editors should follow. The second
edition, which was the last that was brought out in Boswell's life-time,
could not, I became convinced, be conveniently reproduced. As it was
passing through the press he obtained many additional anecdotes and
letters. These he somewhat awkwardly inserted in an Introduction and an
Appendix. He was engaged on his third edition when he died. 'He had
pointed out where some of these materials should be inserted,' and 'in
the margin of the copy which he had in part revised he had written
notes[36].' His interrupted labours were completed by Edmond Malone, to
whom he had read aloud almost the whole of his original manuscript, and
who had helped him in the revision of the first half of the book when it
was in type[37]. 'These notes,' says Malone, 'are faithfully preserved.'
He adds that 'every new remark, not written by the author, for the sake
of distinction has been enclosed within crotchets[38].' In the third
edition therefore we have the work in the condition in which it would
have most approved itself to Boswell's own judgment. In one point only,
and that a trifling one, had Malone to exercise his judgment. But so
skilful an editor was very unlikely to go wrong in those few cases in
which he was called upon to insert in their proper places the additional
material which the author had already published in his second edition.
Malone did not, however, correct the proof-sheets. I thought it my duty,
therefore, in revising my work to have the text of Boswell's second