the line must not be constrained to begin with an accent, as in the metres which we call Trochaic. It remained to inquire what should be its compass; and a series of trials showed, that it was best to compose the line of four beats added to three. Many passages of Homer can be got into an "Alexandrine," that is, into three and three; but I found that, first, this could not be kept up systematically, without becoming too terse; whereas the genius of Homer is to be loose and expansive: secondly, my metre could not be right, unless it would render also the polished hexameter of epigrams and epitaphs; but while "four and three" had compass enough for this, the "three and three" often failed entirely; next, I found that many even of the Homeric lines by no compression could be brought into the Alexandrine, and that beauty and effect was sometimes largely lost if it was impossible to render line by line; lastly, no long trial made it certain to me that the monotony of the Alexandrine is unendurable in a long poem, since the first part of the line has no facility of various subdivision. Such were the general arguments which forced me to believe four beats and three beats to give the elementary solution of my problem.
But, beside this, I held it as an axiom that rhyme must be abandoned. Even to Chapman, with his Homeric genius, and a metre fundamentally good, it was impossible to let the Englishman know what Homer had said, and not obtrude on him what was Chapman's own: for, the exigencies of rhyme positively forbid faithfulness. Yet on abandoning rhyme, to which our ears are accustomed in the popular ballad, I found an unpleasant void, until I gave a double ending to the verse, i. e., one (unaccented) syllable more than our Common Metre allows. Having attained this result by an exhaustive process of argument and experiment, I found with pleasure that I had exactly alighted on the metre which the modern Greeks adopt
