ferent parts, written at intervals during seven years, harmonize so well
as they do. There would have been nothing particular in this, if the parts
had related only to matters well-ascertained before any of them were
written:--but as each professes to contain something of original discovery,
or of correction of received views, it does surprise even my partiality,
that they should have the degree of consistency and apparent general
accuracy which they seem to me to present.

I have made some alterations in the text, but they have been altogether of
a typographical or grammatical character; and even where greatest, have
been intended to explain the sense, not to alter it. I have often added
Notes at the bottom of the page, as to paragraphs 59, 360, 439, 521, 552,
555, 598, 657, 883, for the correction of errors, and also the purpose of
illustration: but these are all distinguished from the Original Notes of
the Researches by the date of _Dec. 1838_.

The date of a scientific paper containing any pretensions to discovery is
frequently a matter of serious importance, and it is a great misfortune
that there are many most valuable communications, essential to the history
and progress of science, with respect to which this point cannot now be
ascertained. This arises from the circumstance of the papers having no
dates attached to them individually, and of the journals in which they
appear having such as are inaccurate, i.e. dates of a period earlier than
that of publication. I may refer to the note at the end of the First
Series, as an illustration of the kind of confusion thus produced. These
circumstances have induced me to affix a date at the top of every other
page, and I have thought myself justified in using that placed by the
Secretary of the Royal Society on each paper as it was received. An author
has no right, perhaps, to claim an earlier one, unless it has received
confirmation by some public act or officer.

Before concluding these lines I would beg leave to make a reference or two;
first, to my own Papers on Electro-magnetic Rotations in the Quarterly
Journal of Science, 1822. xii. 74. 186. 283. 416, and also to my Letter on
Magneto-electric Induction in the Annales de Chimie, li. p. 404. These
might, as to the matter, very properly have appeared in this volume, but
they would have interfered with it as a simple reprint of the "Experimental
Researches" of the Philosophical Transactions.

Then I wish to refer, in