rhythm and in the quality of its tones. Rhythm has two branches: time and stress, or quantity and accent. Both are familiar to music, but time more especially. Hence, that poetry which depends, for metrical effect, chiefly upon detailed time-relations (quantity) will come nearer to music than the poetry which depends chiefly on stress-relations (intensity, accent).

§ 2. QUANTITY.

Quantity deals with the relative length of a syllable ; that is, with the time required to utter it. The Greeks adopted quantity as principle of their metre, and based their verse upon the relation of long and short syllables. A syllable was long which contained a long vowel or a diphthong, or a final consonant coming before another consonant in the next syllable; a long syllable was equal to two short ones. For such poetry, the term "metre" is very appropriate: the verse was really measured. In the Germanic languages, and in nearly all modern poetry, accent is made the principle of verse: we weigh our syllables, we ask how much force, not how much time, they require. Meanwhile, we do not utterly refuse to recognize quantity as an element of verse, nor was classic poetry unfamiliar with accent. In the latter, an "ictus," or stress, fell upon the long syllable; in modern verse, while the main principle is the alternation of heavy and light syllables, we nevertheless admit quantity as a "regulative" element. It is a secondary factor of verse.

First, as to the principle of quantity in classic verse. Take the famous line of Vergil:—

"Quadrupedante putrem sonitu quatit ungula campum,”.

and a verse of Evangeline:

66

This is the forest primeval, but where are the hearts that beneath it,"

and at first sight we call each a dactylic hexameter verse. We give a scheme: :

In one sense, this scheme fits both verses; but there is a radical difference in the application. In the Latin, contrast of long (-) and short (~), a fixed relation of time within the foot as well as within the verse, gave exquisite pleasure to the sensitive ear. This timerelation was the chief metrical factor, although an "ictus" (') or stress undoubtedly marked the long syllables. In the English verse there is no fixed relation of quantity within the foot: "this" requires practically no more time than "is" or "the," and not as much as the metrically short but actually long pri- in "primeval." The time-intervals of the whole verse are marked off by the recurrence of the stress, just as in Latin by the recurrence of the long syllable. This is an important difference. We may say that in classic metres, quantity is the mistress, while quality (stress) plays a handmaid's part. The result was a harmony more musical than can be given by our verse, in which stress is chief metrical factor, and quantity has only a regulative office. Some writers say that modern verse does not recognize quantity at all. This is a mistake. "Long and short syllables," says Schipper in his Englische Metrik, "have no constant length, no constant relation, but they depend on their place in the

verse, and on the context; though they do not deter

mine the rhythm of verse, they still act as regulators of our metre in a very important degree." That is, while no precise rules prevail, the skilful poet avoids an excess of unaccented long syllables or accented short ones. It is not the proportion of long and short within the foot that we heed, but the proportion in the whole verse. Further, quantity is used to help the meaning -a sort of onomatopeia: as in

"The lowing herd wind slowly o'er the lea."

It is very important to hold apart this special, classical principle of quantity, or the time of separate syllables, from the general principle of time-intervals underlying all rhythm (cf. p. 134). Thus Tennyson's two verses: "Break-break - break"-and "On thy cold, gray stones, O Sea!" are rhythmically harmonious, since the time-intervals agree; as may be seen by any one who will tap off the accented syllables, allowing for the pauses in the first verse. But we can arrive at no metrical result by simply applying the test of quantity to the individual syllables. It is not the length of the word "break" (of course, elocutionary motives may prolong the sound at will) which makes it metrically equal to "on thy cold"; it is the heavy accent, followed by a pause.

§ 3. ACCENT.

Accent, then, is the chief factor of modern verse. But there are two kinds of accent which we must consider before we can fully grasp the difference between classical and modern metres: the word-accent and the verse-accent. (1) WORD-ACCENT. When a word has two syllables, one of these receives a marked increase

of tone as compared with the other. In words of more than two syllables, there is generally a secondary accent: i.e., one of the remaining syllables receives less tone, indeed, than the accented syllable, but more than the rest: cf. shepherd, shepherdess, shepherdèsses. Of course, there can be a third accent, if the word have syllables enough; for, as said above, speech tends to alternate accented with unaccented.

Of the same nature as the word-accent are, further, the syntactical and the rhetorical accent, which concern relations of words in a sentence. The accent lifts certain words into prominence, leaving others without special stress of tone, and without the added distinctness of articulation which often accompanies accent.

These two accents-of the word and of the sentence -are of great importance in modern verse; but in the classic metres, which had more of a musical character than our own, they exercised less influence. Especially is this the case with word-accent; and this we must look at more closely, in order to see what difference. there is between ancient and modern languages in their methods of selecting, in a given word, the syllable to be accented. This applies, of course, to prose as well as to poetry. (1) The Grammatical Accent. This is the principle in Sanskrit, and, to a certain extent, in Greek. Taking a given word, we find its accented syllable shifting with different grammatical forms of the word. In Sanskrit this word-accent is not even confined, as it is in Greek, to the last three syllables. Thus we have a Movable Accent. (2) The Rhythmical Accent. The word-accent tends to fall upon a long syllable, as in the Latin. In Greek, the accent was

indifferent to the quantity of the syllable on which it fell thus the Greek chimaira became Latin chiméra. (3) The Logical Accent. - A brilliant piece of research by Carl Verner has proved the existence of a movable accent in the oldest forms of the Germanic languages. This has left its mark in a few sound-changes with which we are not here concerned. But it is certain that at a very early period, before the date of any Germanic literature known to us, this movable accent was given up, and the word-accent became a fixed one. It chose and clung to a certain syllable, and this was the syllable which gave meaning to the word. Hence the term "logical accent." In all original English words, and in many words derived from foreign sources, we bring out with additional stress the syllable which bears the real weight of the word, the root-syllable. Instead of the shifting Greek accent which changed from a nominative to a genitive of the same word (ánthropos: anthropou), we have such persistence as sheep, shepherd, shepherdess, shepherdesses.

(11) VERSE-ACCENT. We have seen that verse is now marked off by the regular recurrence of a stress or accent falling on certain syllables; and that even in classic metres a stress fell upon the long syllables. We naturally ask how this verse-accent agrees with the word-accent just described. Looking, first, at the different ways in which we could make verses, we find the simplest plan to be a mere counting of syllables, with absolute ignoring of word-accent. Each syllable would be a verse-accent. Thus, if we slowly count off "one - two three-four," then repeat the words with the same slowness, accenting each like the rest, we shall