quality of sound, and in each case the longer in sound is the closer of the two.

The broad a of all is called by our authors the "German a :" an unfortunately chosen title, as the German has no such sound.

The "short o" of on, not, and the like, is stated to be the shortened form of o in orb. This is not quite correct; it is more nearly the shortened form of the open ä (in far). It is, in truth, the nearest approach which our language makes to a proper short a, of the true original quality of sound, and differs but very little from the German short a, as in hat, Mann. So far as it differs from this, it verges toward broad â, but the approach is by no means close. The tendency which has thus shifted the short o from its proper place in the vowel series up close to ä seems to belong to the Latin side of our language; for the French and Italian exhibit it strikingly, while the German is altogether free from it.

The u of burn receives the appellation of the "natural vowel:" another striking misnomer, in our opinion. The history of alphabets shows it rather to be the most unnatural of the vowels: it is everywhere of very late development; it is altogether wanting in the great majority of languages; it is nowhere but in English, so far as known to us, admitted to the full rank and privileges of a vowel, furnished with a short and a long value, and suffered to appear in accented as well as unaccented syllables. It is properly the indefinite or uncharacterized vowel, being as nearly as possible the mere grunt, the breath intonated without fairly opening the mouth, a swallowed äin short, the most indolent and ungraceful of all our vowel sounds, especially in its shortened form, as the u of but.

The tendency so generally prevailing in our country to shorten the long o of certain words, as of home, whole, none, etc.-dropping off the vanishing sound of u inherent in our English ō, and perhaps also giving it a little more open quality, as in the case of the other short vowels-is noticed by our authors, and condemned as an unauthorized provincialism. No exception, certainly, can be taken to this sentence; and yet we cannot help saying that if the tendency may not be

approved by the orthoepist, it may at least be regarded with interest by the student of the language. For since we converted all our short o's into short ä's (as in not), or the grunting vowel (as in son), we are in the anomalous position of entirely lacking so important a member of our vowel system. It seems a little as if we were now feeling the want, and trying to supply it by cutting short a portion of our long o's, whereof we have abundance; and the endeavor does not require to be altogether frowned upon as vicious. We should like very

much to know how far the tendency extends, and whether it does not show itself, at least sporadically, in England also; for our own part, if it should continue to spread and prevail, and should establish itself finally within certain definite limits, not threatening to swallow up the whole mass of o's, we should be willing to vote for its recognition, and for the adoption of the new o as a legitimate member into our family of sounds.

Many languages have shown a tendency to mix their long u(oo)-sound with an (ee). Thus the ancient Greek and i modern French have corrupted all their original u's into a new and peculiar sound, which is an actual amalgamation of i and u the tongue, in its pronunciation, being fixed to pronounce the former, and the lips at the same time to pronounce the latter of the two vowels. The ancient Moso-Gothic and the modern English prefix an i to the u, forming the compound yu, as in union, use, acute, mew, and the like. If the long u-whether represented by u itself or by certain digraphsstand at the beginning of a word, or if it be preceded anywhere by a palatal or labial letter (k, g, p, b, f, v, m), the prefixion takes place without fail. So, also, after the lingual letters t, d, n, 1, th, in all cases where the accent lies on the preceding syllable; thus we say pen-yu-ry, as-sid-yu-ous, valyu, as distinctly as pyur (pure), myut (mute), kyu-riús (curious). All orthoepists are agreed, however, that it is not so after an r; we pronounce rude and rued precisely like rood: and our authors very properly remark that the usage is the same after sh and ch (as in sure, sugar, chew); to which we would add zh and j (in azure, jewel), as belonging obviously in the same category with their corresponding surds, sh and ch.

As to the u which follows t, d, l, n in an accented syllable, common practice is somewhat divided. If we are not mistaken, very few in such a case utter a full and distinct y before the oo-sound, saying lyut (lute), penyurious, assidyuity; the semivowel is so far slighted, and half-suppressed, as to form a kind of indistinct and almost indefinable prefix to the o0. But many persons, even among the educated-and not excepting, as we can testify from careful observation, some distinguished speakers-omit the prefix altogether in such positions, and pronounce a clear and unmixed oo, knowing no middle ground between it and a full and unmistakable yoo. So satisfactory a physical explanation can be given of this difference of usage as regards and the other dental letters referred to, that we are inclined to recognize it as a legitimate tendency of the language to get rid of the prefix in the cases defined-in accented syllables beginning with t, d, l, and n-and to suspect that, but for the adverse influence of the orthoepists, that result would be in no long time attained.

Our authors give in their adhesion to the party, apparently a growing one among modern orthoepists, of those who hold the sound represented by our wh (as in when, whit, etc.) not to be compound in its nature, made up of a u with a prefixed h, but to be a simple and indivisible utterance. They cite to the following effect a trusted authority: "We doubt not that, if a man will observe carefully for himself how and with what difference he pronounces wit and whit, he will be satisfied that the h is really pronounced neither before nor after the w, but in a sort of constant combination with it." A convenient way this, surely, of proving a point. It is, to our apprehension, as if one were to say: "If a man will carefully observe a train of cars issuing from a station, he will satisfy himself that the engine neither precedes nor follows the baggage-car, but that each of the two is inside and outside of the other." We can only answer that we are satisfied of precisely the contrary. We seem to see clearly that putting an h before wit makes it whit, and that whit can be produced in no other way; that as 00-en, by the slighting of the oo, makes wen, so hoo-en, in just the same way, makes when. There is even a palpable ab

surdity in the other view as just stated: how can one utter at the same time, with the same throat, an h, which is an emission of unintonated breath, an aspirate, a surd, and a w, which is pronounced with intonated breath, being a sonant or vocal letter? Can a fountain send forth at the same time sweet waters and bitter?

As little can we agree with the unitarian views set forth in the Manual touching the character of the sound represented in our language by ch (in church). That this is a simple and uncompounded sound we are to believe upon the strength of the following argument, derived from the same authority as that cited above: "It is produced by placing a certain portion of the tongue near the tip, but not the tip itself, against a certain part of the palate, and, after pressure, suddenly withdrawing it with a violent emission of breath. It has no t-sound in its composition, for neither the tip of the tongue nor the teeth are used in its production. Neither does it end in an sh-sound; for, in that case, it could be prolonged ad libitum, which the true English ch cannot be. Moreover, it does not begin with any one sound and end with another, but is the same simple sound throughout its whole extent." To which we reply, as before: not so, but the contrary. The process first described would give no ch, but a t. We say ▲ t, because the sound so produced is not our usual t; the latter is in truth uttered with the tip of the tongue, only not against the teeth, but just behind them. At may be formed anywhere on the roof of the mouth forward of the soft palate, by any part of the tongue which can be applied to it: it will vary in quality according to the point at which the contact is made; and some languages have two, three, or even four t's, produced at different points in the region indicated; the Semitic tongues possess that very t which we use only as the first element in our ch. It is brought forth by a complete contact of the two organs at just the point where a very slight opening, "with a violent emission of breath" through it, gives rise to sh. If, then, instead of "after pressure suddenly withdrawing" the tongue, you merely open the organs a little, a sh ensues, and the t and the sh together form ch. It can be prolonged ad

libitum without the smallest difficulty; only, instead of prolonging the whole compound, of course, we prolong only the final constituent of it, the sh. The j, which is merely the sonant or vocal ch, is composed in like manner of a d followed by zh. The two cannot accurately be written with tsh and dzh, as is so often done, because the initial t and d are not precisely the sounds which we write with those characters; yet the inaccuracy is but an insignificant one.

We have noted but one other point of any consequence, with regard to which we feel called upon to differ from the authors of the Manual, and this concerns the alleged double pronunciation of the r. This letter, we are taught, is vibrated or trilled when it precedes a vowel, but has in other situations a softer sound, to be designated as the "smooth r:" such is the r in fir, urn, and the like; it is attempted to distinguish the rough r from that of urn by descriptions in which we read of streams of breath directed against the palate, of vibrations of the uvula and of the back part of the tongue, etc.mere words without meaning, if we do not misjudge them. We can discover no ground for the distinction sought to be established, save in the fact that in our ordinary careless pronunciation we are apt not to make the audible at all after a vowel. Is it not a mistake to call our r in roam, florid, a trilled sound? Trilled it may be by an effort, and trilled it often is in affected or in laboriously distinct or energetic speech; but in customary pronunciation there is, we believe, no vibration. An may be uttered between the tip of the tongue and the roof of the mouth, at any point, from close behind the teeth to as far back as the tongue can reach: when quite near the teeth, the sound is harsher, and a vibration or a buzz can hardly be avoided; but, the tongue once reverted within the dome of the palate, the utterance is smooth and easy, and a vibration can only with difficulty be produced. Of the latter kind is our English 7, if we do not misapprehend its character, both when followed and when preceded by a vowel. In such words as fire, pour, there, here, pure, etc., we add a vanishing sound of u in urn or up to the principal vowel of the syllable, as introductory to the r, and then often