duced by this means; they are multiple resonances of the column of air, and not the vibrations of the" (metallic) "tongue itself, as was formerly supposed.

“Similar results are obtained when the vibrating tongue of an Æolina is brought before a tube, and its length is altered, and this case resembles Mr. Willis's arrangement. The same multiple resonances are produced also when the cavity of the mouth is substituted for the tube. In these cases the fundamental sound is louder than when the tongue of the Jew's harp was employed. The sound of the larynx itself may be substituted for that of the vibrating tongue, and similar harmonic sounds will be heard; this experiment may easily be made by placing a piston tube before the mouth, while the voice continues to sound rather a low note. About two years ago (1835?) a young man named Richmond exhibited a novel kind of musical performance with the voice: on examining the circumstances under which the sounds were produced, it was ascertained that the continued sound or drone was produced by the larynx and that he had acquired the art of adjusting the cavity of the mouth so as to fit it for resounding to any multiple. In this way he was able to command these subordinate sounds in any succession, and even to dwell upon them; and he could thus perform a great number of airs.

"Some kind of sounds are better suited to produce these multiple resonances than others, and it is a universal fact, that wherever these subordinate sounds can be distinguished, there also the vowel qualities are heard; and reciprocally, when a sound puts on successively different vowel qualities, these multiple resonances are audible. The tongue of a Jew's harp, which so readily gives rise to these subordinate sounds, is obedient not only to the vowel sounds, but to almost all the articulations of speech. The free reed or æolina tongue, when it is such as can enter readily into vibration, is affected in a similar manner; but when it is too rigid, though it may produce as clear a musical sound as before, the multiple resonances and vowel qualities are equally lost; not, perhaps, be

he returned to London, and had again no success. About this time, "the iron of the Jew's harp had affected his teeth, and produced general decay," and he suffered great pain from any attempt to play upon this instrument. Since this time, he has performed comparatively rarely, and always with considerable suffering; but he still continued to give concerts down to 1833. His last was given at Clifton, "the breaking of his last tooth obliging him to relinquish." He soon after married a German lady; and in September, 1834, established himself, as a teacher of the German language and the guitar, at Bath, where he still resides (1844). These particulars are taken from a little work entitled "A Sketch of the Life of C. Eulenstein, the celebrated performer on the Jew's Harp. 2nd edition; London, 1840;" p. 57.

cause they do not exist, but because they are overpowered by the original sound of the reed.

"We do not mean to assert that each multiple resonance is a distinct vowel sound. But we infer, that when a tube is added to a reed or vibrating tongue, whatever may be its length, a quality is added to the original sound, which depends on the feeble vibrations of the air in the added tube: these increase in number in proportion to the shortness of the tube; and when the number of vibrations thus excited is any multiple of the original vibrations of the reed, the energy of the resonance is so greatly augmented as to produce the effect of a superadded musical sound. "Thus it is evident that the vowel qualities and multiple resonances are different forms of the same phenomena."

The reader is now in possession of the principal modern discoveries concerning the general nature of the vowel sounds. The subject cannot be said to be entirely cleared up, but much has been done towards its elucidation. An examination of the principal vowel qualities we are acquainted with, together with a notation by which every possible shade of sound may be accurately represented will form the subject of Chapter V.

CHAPTER 4.-ON THE CONSONANTS IN GENERAL.

In the preceding chapter we took a general survey of the modifications produced in the sound of the voice by such alterations in the forms of the cavities of the mouth, &c., as did not amount to a total, or very nearly a total, stoppage of the aperture through which the air had to pass on leaving the lungs. We have, therefore, next to consider these two cases, which we shall at first suppose to belong to the first mode of the second and third kinds of action mentioned in chap. 2, p. 21.

In the first of the two cases just alluded to, the passage of the voice is entirely cut off, and the effect of the modification is only felt during the opening and closing of the mouth; thus in the words ebb, add, egg, the voice is in this manner entirely cut off as the mouth closes, and in the words bee, day, go, the voice is only heard while the mouth is being opened. The instant that the mouth closes, the sound ceases; or so nearly at that instant, that the period of time during which it really lasts (a very small fraction of a second) may, for the present, be disregarded.(1) But if the contact is not complete, as in the second case, mentioned

(1.) It is, however, an important fact, that the sound is not cut off absolutely instantaneously; and we shall have occasion to take this into consideration hereafter.

above, the duration of the voice will be heard as a faint buzzing sound; as in the words have, seethe, whiz, judge.

We have, then, in all these cases, a modification of the voice brought in a manner analogous to that considered in the last chapter. Now, it has never been customary to have a sign for the simple unmodified voice, because the voice cannot be produced but in some shape, as it were, caused by the opening of the mouth or nostrils, so that we cannot hear voice except in a modified state, as long as it must pass through the necessarily modifying organs, the mouth, &c. Thus we may talk of matter generally, and of a shapeless mass, but we can never take cognizance of matter except when modified by certain conditions; nor can we see a mass that has not a definite (although irregular) form. It would appear, therefore, that we must suppose that the signs employed for the purpose of regulating the arrangement of the parts of the mouth, &c., necessarily imply the putting on of the voice. In ancient languages, this seems to have been universally the case for all the three kinds of actions mentioned in Chap. 2. It was, however, soon found that the parts of the mouth might be arranged in more complicated manners than were expressed by the individual signs of the alphabet, and that this complication arose from compounding these several arrangements into one, so that the new compound arrangement might be very well expressed by making two or three of these marks in succession, and placing another mark to shew that the voice is not to be heard till after the last ajustment had been completed. This is the method we find used in the Hebrew, Arabic, and Sanscrit alphabets, in which each letter is supposed to represent a certain sound, produced by arranging the parts of the mouth, &c., in a certain position, and setting on the voice, unless a mark of quiescence—called shəwa' (=shəvā) by the Hebrews, and jazmü (=jezmon, or jezm) by the Arabs-were annexed, being written either under (as in Hebrew and Sanscrit) or over the consonant (as in Arabic).

In the later alphabets, which are but modifications of the older ones, another method has been chosen, which is, in many respects, more convenient. In the vowel signs, the position of the parts of the mouth, and the setting on of the voice, are, as before, indicated by a single character, an additional mark (h, or н) being prefixed when the vowel sound is preceded by the breath; while, in the second and third kinds, the sign for the first mode only represents the modification itself, whereas, that for the second mode indicates both the modification and the putting on of the breath,(2) This must be further explained. We may produce a sound

(2.) This is a mere matter of convenience in writing and printing. It would

not only by the voice, but by the explosion of the breath, or its condensation in the mouth. In producing this sound the larynx has no share, and the result is, therefore, termed whispering, and not speaking. The consonants, accompanied by this whispering effect, belong, as already indicated, to the second mode of action, spoken of in Chap. 2. If the reader will pronounce the words his, hiss, and endeavour to lengthen their sounds as much as possible, he will form a clear idea of the essential difference between a spoken and a whispered consonant.

The reader will now have a general notion of the nature of consonants; and, as a future chapter is devoted to the details, we might conclude here, were there not a question, which has been so often agitated, that it may be, perhaps, worth while to point out the ambiguity which has occasioned the dispute. "? Can a consonant sound by itself," is the question to which we allude. The word consonant (from con, with, and sonans, sounding) would seem to favor the idea that it could be sounded alone, but that it is generally (or, as some say, always) sound with some other letter.

Now, we have just stated that the true consonant consists in the ajustment of the parts of the mouth, which mere ajustment can, of course, produce no sounds, and, therefore, the set of symbols (also generally called consonants, but more correctly termed consonant signs) which represent these ajustments do not represent any sound. This ajustment may, however, be accompanied by the voice or the breath; in the first case, the voice is also produced under a modified form, and therefore represented by a vowel sign; but, in the latter case, both the breath and have been more philosophical to have proceeded thus:-Let there be a mark for the voice simply, say '; let there be a sign for each arrangement of the parts of the mouth, &c., as A, E, I, O, U-B, D, G, &c.; and let there be a mark to imply the breath, as. Then we might symbolize thus: A' is a simple vowel; A" an aspirated vowel; B' a simple consonant, followed by the voice; B' a simple consonant, followed by the breath; B A' a spoken consonant, followed by a vowel; B' A' a whispered consonant, followed by a simple vowel; B' A" a whispered consonant, followed by an aspirated vowel-a combination which does not occur in English, but is common in Sanscrit, and is heard in the Irish brogue. Thus, if A', E', I', O', U', be the sounds represented by the italic letters in the words mar, may, me, mow, moo, respectively, and B, D, G, the ajustments indicated by the italic letters in bee, do, go, respectively, we shall have BA'=baa; DE'=day; GO'=go; while B'E'=pay; D'I' tea; GI' key, &c.; E"=hay; I"=he; E"D'=hate, &c. From these instances, the conveniences of the mode of symbolisation, in which A', E', I', O', U', are represented by simple signs, as a, e, i, o, u, and B', D', Gʻ, by the simple signs, p, t, k, while E'', &c., is denoted by prefixing some sign to the mark chosen to represent E' (thus, нe), will be very evident, although the law of symbolisation appears, at first sight, to be somewhat complicated.

the ajustment are represented by the same sign; that is, a single symbol represents both the ajustment and the contemporaneous emission of breath, and it, therefore, does represent a sound. Now, it happens unfortunately that the word consonant is, in common language, also applied to these signs which do represent a sound, although not a vocal sound; and hence it is both true and false that consonants can sound of themselves. If we discriminate the classes, calling the former symbols "spoken consonant” signs, and the latter "whispered consonant signs," then the real answer is, that the first do not, and the second do, represent a sound, and, therefore, in common parlance, the first cannot, and the second can, sound by itself.

This is the view of the case which we think it best to take; others suppose the element of the voice to be inherent in the spoken consonants, so that they, also, can be pronounced; but we cannot agree with them in this conclusion. Since, then, s, sh, th, f, &c., can be pronounced independently, of course, sp, sht, tht, ft, ps, psh, &c., constitute real sounds ; but these sounds are not vocal, and are, therefore, not generally recognized as constituting syllables. Dr. Orpen observes: (3) "It is commonly said, that a consonant cannot be pronounced without a vowel, and, in fact, its name was given, probably, from this theory, but it is altogether a mistake; for the consonants, p, t, k, have no vocal sound at all, and yet can be pronounced with an s before or after them without any vowel, or any vocal sound; as in ps, ts, ks; sp, st, sk; whose whole compound sound is an affection of the breath alone, and not of the voice. Again, b, d, g, have a slight vocal sound in their essence, but that sound is not a vowel, and these letters can be united to z, without any vowel; as in bz, dz, gz; zb, zd, zg. Thus, too, m, n, ng, have a vocal sound essential to them, but it is emitted through the nose, and, therefore, cannot possibly be a vowel sound ;(4) and, besides, they can be united with s or z without any vowel ; as ms, mz; ns, nz; ngs, ngz; or in sm, zm; sn, zn; sng, zng.(5) Besides, f, th, s, sh, lh, rh, kh, have no vocal sound, being mere affections of the breath, yet they can be united with each other, and with other consonants, in various ways, without any vowel; as in fs, ths, &c., sf, sth, &c. The corresponding vocal consonants, too, v, dh, z, zh, l, r, gh, have a vocal sound essential to them, but that sound is not a vowel, and they

(3.) The Pestalozzian Primer, pp. 127—129.

(4.) If Dr. Orpen means to imply that there can be no nasal vowels, he is certainly in error.

(5.) We question those cases in which z occurs; and we believe that, in the other cases, the sounds are what we shall represent by a, a, n, being the whispered consonants, corresponding to what we denote by m, n, N.