serious poetry, it was lawful for them to produce results not wholly unlike that in the third rhyme of the following, the classic quality of which some of us hitherto may not have recognized:

For he might have been a Roosian,
A French, or Turk, or Proosian,
Or perhaps I-tal-i-an.

But in spite of all temptations

To belong to other nations,

He remains an Englishman.

-Pinafore: Gilbert.

Our poets, on the contrary, have gone back to the primitive methods, antedating those of Greece, and base the rhythms of their verse on the accents of speech. The result, as compared with the language of our prose, is more natural than that reached by the other method; and in its way is fully as artistic. Nor, in other regards, is English inferior to the classic tongues in its capabilities for artistic treatment. Owing to an extensive use of terminations in nouns, articles, pronouns, adjectives, and verbs, in order to indicate different grammatical relationships, the Greeks and Romans could change the order of words in a sentence without changing its meaning. In their language, "The dog ate the wolf," with slightly varied terminations, could read, "The wolf ate the dog.' For this reason, they could alter their phraseology, in order to accommodate it to the requirements of metre, as is not possible for us; and so far they had an advantage over us. Nevertheless, for some reason, when they came to put their words into verse, as every schoolboy who tries to scan knows, they produced a language which, like the present French poetic diction, sounded unlike

that of conversation. Even supposing, with some scholars, that in reading they did not scan their verses as we do now, nor even chant them invariably, as some infer was the case, their poetic language was not the same as their spoken language. Aristotle tells us, when mentioning things which it is legitimate for the poet to do, that he can invent new words, that he can expand old ones, either by lengthening vowels or by adding syllables, that he can contract them by shortening vowels or omitting syllables, and that he can alter them in various other ways. Spenser and others since him have applied similar methods to English poetic diction; but, at present, such changes, except in rare instances, are not considered admissible, and this because they are recognized to be unnecessary. The fact that they are not admissible in our language, and were admissible in the classic languages, proves that, in one regard at least, our language is superior to them as a medium of metre. The following is a typical English stanza. In it there are no changes from ordinary prose in the arrangement, spelling, or punctuation of any of the words:

He will hold thee, when his passion shall have spent its novel force,
Something better than his dog, a little dearer than his horse.
-Locksley Hall: Tennyson.

In this chapter we have been considering rhythm as related to certain general underlying principles, an acquaintance with which, as has been intimated, is all that is absolutely necessary for either reading or writing poetry. But, for a full understanding of the subject, the formal systems of metre and versification into which, in our language as in others, these principles have been developed, ought also to be examined. This will be done in the chapters following.

CHAPTER III.

ART-METHODS AS DEVELOPING MEASURE AND VERSE.

The Art Methods, especially Repetition, as Causing Groups of Syllables in Measures-Double and Triple Measures-Initial, Terminal, Median, Compound, and Double Initial and Terminal—Significance of Each Measure-Art-Methods as Causing Groups of Measures in Lines-Hebrew Parallelism, and Greek—The Couplet-The Cæsura-Lines of One, Two, Three, Four, Five, Six, and More Measures-Examples of Them-The Iambic Tetrameter-The Iambic Pentameter, Heroic Measure, Blank Verse-The Classic Hexameter-English HexameterChildren of the Lord's Supper-Another Example-A Translation from the Iliad-The Alexandrine.

IT will be noticed that, according to the chart on page 3, the methods already mentioned are all those that are absolutely necessary for the production of rhythm, the methods further developed from these being more particularly connected with harmony. At the same time, even these latter methods are only more subtle manifestations of the former, and certain traces of them are apparent even in rhythm. This is especially true of repetition, and the methods immediately connected with it. The artistic tendency to comparison needs only to be intensified, as applied to the form, and it will cause accented syllables in all cases to be separated by exactly the same number of unaccented syllables; and will also cause exactly the same number of both accents and syllables to be placed in each line. When this has been done,-even before it has been done as we have noticed in the poetry

already quoted-each accented syllable, together with one or more unaccented, seems to constitute one group; and a certain number of these groups to constitute one line. As a result, the line can be regularly measured by the number of the groups into which it is divided. For this reason they are termed measures, and, owing to a supposed correspondence of movement between the use of one measure after another, and that of the feet in walking, they are also termed feet.

In general, we may divide all possible measures into two classes, namely, those that are double and those that are triple. The first are made up of feet of two syllables, every other of which is accented, e. g. :

When the hours of | day are | numbered.

It also includes feet of four syllables, only one of which receives a strong accent; though the second from it may receive a subordinate accent. The general effect, therefore, of this measure, which is sometimes termed quadruple, is that of a doubled double measure, e. g.:

Roses are in blossom and the | rills are filled with | water-cresses. Triple measures contain three syllables, e. g.:

Cannon to right of them, | cannon to | left of them.

But besides being distinguished from one another by the number of syllables composing them, measures differ according to the syllable in them-whether the first, second, third, or fourth-that receives the accent. This method of difference in connection with the other just noticed leads us to find six, or, in case we consider the quadruple measures other than modifications of the

Here

double, eight kinds of measures (see page 103). they are with names indicative of the methods of forming them, in connection with which are given also the terms of Greek origin ordinarily assigned to them. But as these terms apply to arrangements of quantity rather than of accent, they frequently fail to describe accurately the English measures. Hence the use here of the new

terms.

Initial or initial double measure is accented on the first syllable, and corresponds, if composed of one long syllable followed by one short, to the Greek trochee or choree; if of two long, to the Greek spondee.

When the hours of | dày are | numbered.

Terminal or terminal double measure is accented on the second syllable, and corresponds, if composed of one short followed by one long syllable, to the Greek iambus.

Among | thy fàn | cies, tèll | me this.

Initial triple measure, if composed of one long followed by two short syllables, is the same as the Greek dactyl.

Out of the cities and | into the | villages.

Median or medial triple measure, i. e., triple measure with the accent on the middle syllable, if composed of one short, one long, and one short syllable, is the same as the Greek amphibrach.

There came to the shore a | poor exile | of Èrin.

Terminal triple measure, if composed of two short syllables followed by a long one, is the same as the Greek anapæst.

If our landlord supply | us with beef and with fish.