lexity. Rules founded on the analogy of similar words, and
sanctioned by the usage of careful writers, must be taken as our guides;
because common practice is often found to be capricious, contradictory, and
uncertain. That errors and inconsistencies abound, even in the books which
are proposed to the world as _standards_ of English orthography, is a
position which scarcely needs proof. It is true, to a greater or less
extent, of all the spelling-books and dictionaries that I have seen, and
probably of all that have ever been published. And as all authors are
liable to mistakes, which others may copy, general rules should have more
weight than particular examples to the contrary. "The right spelling of a
word may be said to be that which agrees the best with its pronunciation,
its etymology, and with the analogy of the particular class of words to
which it belongs."--_Philological Museum_, Vol. i, p. 647.

OBS. 2.--I do not deny that great respect is due to the authority of our
lexicographers, or that great improvement was made in the orthography of
our language when Dr. Johnson put his hand to the work. But sometimes one
man's authority may offset an other's; and he that is inconsistent with
himself, destroys his own: for, surely, his example cannot be paramount to
his principles. Much has been idly said, both for and against the adoption
of Johnson's Dictionary, or Webster's, as _the criterion_ of what is right
or wrong in spelling; but it would seem that no one man's learning is
sufficiently extensive, or his memory sufficiently accurate, to be solely
relied on to furnish _a standard_ by which we may in all cases be governed.
Johnson was generally right; but, like other men, he was sometimes wrong.
He erred sometimes in his _principles_, or in their application; as when he
adopted the _k_ in such words as _rhetorick_, and _demoniack_; or when he
inserted the _u_ in such words as _governour, warriour, superiour_. Neither
of these modes of spelling was ever generally adopted, in any thing like
the number of words to which he applied them; or ever will be; though some
indiscreet compilers are still zealously endeavouring to impose them upon
the public, as the true way of spelling. He also erred sometimes _by
accident_, or _oversight_; as when he spelled thus: "_recall_ and _miscal,
inthrall_ and _bethral, windfall_ and _downfal, laystall_ and _thumbstal,
waterfall_ and _overfal, molehill_ and _dunghil, windmill_ and _twibi