e consonant, the syllable is often _more or
less short_, as it ends with a _single consonant_, or with more than one:
as, 'Sadly, robber; persist, matchless.'

(3.) "When the accent is on a semi-vowel, the time of the syllable may be
protracted, by dwelling upon the _semi-vowel_: as, 'Cur, can, f~ulfil'
but when the accent falls on a mute, the syllable _cannot be lengthened in
the same manner_: as, 'Bubble, captain, totter.'"--_L. Murray's Gram._,
8vo, p. 240; 12mo, 193.

(4.) "In this work, and in the author's Spelling-book, the vowels _e_ and
_o_, in the first syllable of such words as, behave, prejudge, domain,
propose; and in the second syllable of such as pulley, turkey, borrow,
follow; are considered as _long vowels_. The second syllables in such words
as, baby, spicy, holy, fury, are also considered as _long
syllables_."--_Ib._, 8vo, p. 241.

(5.) "In the words _scarecrow, wherefore_, both the syllables are
_unquestionably long_, but not of equal length. We presume _therefore_,
that the syllables under consideration, [i.e., those which end with the
sound of _e_ or _o_ without accent,] may also be properly styled _long
syllables_, though their length is not equal to that of some
others."--_Murray's Octavo Gram._, p. 241.

OBS. 19.--Sheridan's "_infallible rule_, that no vowel ever has a long
sound in an unaccented syllable," is in striking contrast with three of
these positions, and the exact truth of the matter is with neither author.
But, for the accuracy of his doctrine, Murray appeals to "the authority of
the judicious Walker," which he thinks sufficient to prove any syllable
long whose vowel is called so; while the important distinction suggested by
Walker, in his Principles, No. 529, between "the length or shortness of the
vowels," and "that quantity which constitutes poetry," is entirely
overlooked. It is safe to affirm, that all the accented syllables occurring
in the examples above, are _long_; and all the unaccented ones, _short_:
for Murray's long syllables vary in length, and his short ones in
shortness, till not only the just proportion, but the actual relation, of
long and short, is evidently lost with some of them. Does not _match_ in
"_matchless_," _sad_ in "_sadly_," or _bub_ in "_bubble_," require more
time, than _so_ in "_also_," _key_ in "_turkey_," or _ly_ in "_holy_"?
If so, four of the preceding positions are very faulty. And so, indeed, is
the remaining one; for where is the sense of say