OBS. 10.—­Some of the vowels are usually uttered in longer time than others; but whether the former are naturally long, and the latter naturally short, may be doubted:  the common opinion is, that they are.  But one author at least denies it; and says, “We must explode the pretended natural epithets short and long given to our vowels, independent on accent:  and we must observe that our silent e final lengthens not its syllable, unless the preceding vowel be accented.”—­Mackintosh’s Essay on E. Gram., p. 232.  The distinction of long and short vowels which has generally obtained, and the correspondences which some writers have laboured to establish between them, have always been to me sources of much embarrassment.  It would appear, that in one or two instances, sounds that differ only in length, or time, are commonly recognized as different elements; and that grammarians and orthoepists, perceiving this, have attempted to carry out the analogy, and to find among what they call the long vowels a parent sound for each of the short ones.  In doing this, they have either neglected to consult the ear, or have not chosen to abide by its verdict.  I suppose the vowels heard in pull and pool would be necessarily identified, if the former were protracted or the latter shortened; and perhaps there would be a like coalescence of those heard in of and all, were they tried in the same way, though I am not sure of it.  In protracting the e in met, and the i in ship, ignorance or carelessness might perhaps, with the help of our orthoepists, convert the former word into mate and the latter into sheep; and, as this would breed confusion in the language, the avoiding of the similarity may perhaps be a sufficient reason for confining these two sounds of e and i, to that short quantity in which they cannot be mistaken.  But to suppose, as some do, that the protraction of u in tun would identify it with the o in tone, surpasses any notion I have of what stupidity may misconceive.  With one or two exceptions, therefore, it appears to me that each of the pure vowel sounds is of such a nature, that it may be readily recognized by its own peculiar quality or tone, though it be made as long or as short as it is possible for any sound of the human voice to be.  It is manifest that each of the vowel sounds heard in ate, at, arm, all, eel, old, ooze, us, may be protracted to the entire extent of a full breath slowly expended, and still be precisely the same one simple sound;[103] and, on the contrary, that all but one may be shortened to the very minimum of vocality, and still be severally known without danger of mistake.  The prolation of a pure vowel places the organs of utterance in that particular position which the sound of the letter requires, and then holds them unmoved till we have given to it all the length we choose.