as pronounced in ab, ad, ag. If we try to dwell upon the consonants which end these words, we find ourselves unable to do so but for a short time, and even then it requires some muscular exertion. In each of the three cases the tone seems to be modified by a closed cavity, no aperture being left for the breath to escape by. In pronouncing b, the lips are closed, and the vibrations are confined to the throat and mouth; in pronouncing d, the tongue is raised to the palate, and the throat and hinder portion of the mouth are the only open cavities; in pronouncing y, the tone seems to be modified merely by the hollow of the throat. We shall call these letters from the

circumstances of their formation the close letters.

The letters b, d, g have a very near connexion with the three nasals

The only difference in their formation is, that in pronouncing the latter, the breath passes freely through the nostril. With this exception the organs are disposed precisely in the same way for pronouncing m, n, ng, as for pronouncing b, d, g. As the nostril affords a free passage for the breath, we may dwell on these letters during a whole respiration.

v, dh.t

have the strongest affinity to b and d. The peculiarity of their formation lies in the free passage of the breath through the interstices of the upper teeth. To the edge of these teeth we raise the lip in pronouncing v, and the tongue in pronouncing dh, instead of joining the lips, or raising the tongue to the palate. As these teeth form part of the enclosure which modifies the voice, the breath may pass between them, and we may dwell upon the letters during a whole respiration, as is seen in pronouncing the words av, adh.

* This character represents the sound which ends such words as loving, telling, &c.

+ dh represents the vocal sound of th as heard in the, their, those, &c.

w, y,

are never heard in pronunciation except at the beginning of a syllable and before some other vowel. They seem merely to represent the short vowels i and u (as heard in put and pit), melting into their several dipthongs. They are generally considered as consonants; but if the y of your be a consonant, so must also be the e of Enrope.

l, r.

The peculiarity in the formation of these letters is a certain trembling or vibration of the tongue, whence they may be called the trembling letters. In pronouncing 7 the tongue is raised to the palate, as in forming the letter d, but the breath is allowed to escape between it and the side teeth, and thereby causes the loose ends of the tongue to vibrate. In pronouncing the letter r the tongue is raised towards the palate without touching it, and the breath in passing causes it to vibrate.

These tremblings or vibrations of the tongue are quite distinct from the vibrations of the voice, and may be produced during a whisper when the voice is absent.

The only two vocal sounds which remain to be considered are

In pronouncing z the tongue is raised to the palate in nearly the same position it occupies in pronouncing e, save that, instead of lying hollow so as to form a tube or funnel for the voice, the surface rises in a convex shape and leaves but a narrow slit or aperture between it and the roof of the mouth. By lengthening the aperture we get the sound of zh. These letters may be called the sibilants or hissing letters.

* By the character zh is represented the sound of z in azure.

Hitherto we have spoken only of vocal letters, or, in other words, of the different modifications of the voice. If the vocal ligaments be so inclined to each other as not to vibrate, the emission of breath from the lungs produces merely a whisper. This whisper may be modified in like manner as the voice, by similar arrangements of the organs; and every vocal sound has its corresponding whisper-sound, that might, if custom had so willed it, have constituted a distinct letter.

It is, however, doubtful if there ever was a language which had its whisper letters perfect. In our own the number of whisper letters is nine. The three close letters, the two dentals or teeth-breathing letters, the two sibilants, and the letter w, have each of them their whisper letters, and the aspirate h is the ninth.

We have lost all distinction between dh and th in our spelling, though we still distinguish them in pronunciation, as is seen at once in comparing the sound of th in this, then, clothes, to loathe-with its sound in thistle, thin, cloths, loath.

* The distinction here taken between vocal and whisper letters appears to me important. I once thought it was original; but in conversing on this subject with a respected friend, to whose instructions I owe much, I found his views so nearly coinciding with my own, that I have now but little doubt the hint was borrowed.

The distinction also between the connected letter sounds zh and sh does not appear in our orthography, though at once sensible to the ear in comparing the sound of azure with that of Ashur.

That wh represents the whisper sound of w will, I think, be clear, if we compare the initial sounds of where, when, while, with those of were, wen, wile. It is probable that in the Anglo-Saxon hwær, hwen, hwile, the w may have been vocal, and the h may have represented a distinct breathing; but it would be difficult to account for the change of hw into wh, which took place at so early a period (perhaps as early as the 12th century), unless it indicated a change in the pronunciation; and this change would naturally be to the whisper sound of the w.

In this view of the case w may put in a fair claim to the title of consonant. If the true definition of a vowel be, that it is a letter which makes any part of a word, into which it enters, a distinct syllable, then w has clearly no right to the title of vowel. Nor can we reasonably call the initial sounds of were, wen, wile dipthongal, unless we allow the initial sounds of where, when, while, to be dipthongs also. But were this so, we should have part of a dipthong a mere whisper while the other part remained vocal. Our w then, amid a choice of difficulties, may, perhaps, be allowed the title of consonant; but the same reasoning does not apply to the y. The latter, I think, can only be considered as a letter indicating the initial sound of a dipthong.

The whisper sounds of the two liquids 1, r, constitute two distinct letters in Welsh, and in several other languages. I am also inclined to think that the Latin rh, if not the Greek 'p, indicated merely the whisper sound of the r.

That these letters p, t, k, f, &c. are the whisper sounds of b, d, g, v, &c. may, I think, be shown without much difficulty. If we try to pronounce the words ab, ad, ag, av, &c. in a whisper they cannot be distinguished from ap,

at, ak, af, &c. Again, the vibrations of the organs, which are obvious while we are pronouncing a vocal letter, cease immediately we change to the whisper sound; but the disposition of the organs remains unchanged. Thus, in pronouncing the v of av, if we change to a whisper, the vibrations of the lips and teeth cease; and without any change in the position of the organs we find ourselves pronouncing f.

The number then of English consonantal sounds, if we consider was one, amounts to twenty-two; whereof thirteen are vocal and nine mere whisper sounds.

The vowels are eleven in number. The long a, e, o, u, as heard in father, reel, roll, rule; au and a as heard in aught, ate; and the short a, e, i, o, u, as heard in pat, pet, pit, pot, put. The dipthongs are twelve, ei, oi and ou, as heard in height, hoity, out; and eleven others formed by prefixing y to the eleven vowels. These are heard in the following words, yardn, yean, yoke, yule, yawn, yare, yap, yell, yif, yon, young.

Having said thus much on the formation of our elementary sounds, we will now consider in what way and to what extent they may be rendered useful, in embellishing and perfecting the rhythm.

If, as is often the case, besides the idea which the usage of language has connected with certain words, there are others which are naturally associated with the sounds or with the peculiarities of their formation, it is obvious, that the impression on the mind must be the most vivid, when the natural associations can be made to coincide with such as are merely artificial and conventional. In all languages there are certain words in which this coincidence is perfect. In our own we have hiss, kaw, bah, and a few others, in which the natural sound so closely resembles the articulate sound which represents it, that many have fallen into the error of supposing the latter a mere imitation of the former. The number, however, of these imitative sounds in any language is but scanty, and