Exception must therefore be taken, it seems to me, to another contention of Mr. Robertson's; namely, that "there is no time-unit." "Our feet," he says, "are a pure convention, and the sole rhythmic fact is the fluctuant relativity of long and short, or stress and slur." I am glad to be able to believe that the fundamental rhythmic fact is something more definite than this.[52] But the latest writer on the subject, Mr. Mark Liddell, in his _Introduction to the Scientific Study of Poetry_, joins Mr. Robertson in finding no feet in English verse. Nay, he represents the metrist who makes use of the old conventions of rhythmical measurement as one who will "flounder ceaselessly amid the scattered timbers of iambuses, spondees, dactyls, tribrachs, never reaching the firm ground of truth"! Mr. Liddell points out that we do not pronounce English words, even in verse, with mechanically regular alternations of stress; and he rejects the explanation that there is nevertheless a typical form in the poet's and the reader's mind, on the ground that it is "a strange state of affairs that the æsthetically imperfect should produce a greater pleasure than the æsthetically perfect." Strange, perhaps, but as familiar as sunrise, if by "æsthetically perfect" we mean absolutely regular. Here we need to recur to the analogy of music, where the phenomenon in question is so obvious. Let any one attempt to follow a symphony with a metronome in his hand, and he will soon discover that if the metronome represents the æsthetically perfect rhythm, the orchestra represents a more pleasurable imperfection. Its accelerations and retardations carry on a continual conflict with the typical time of the music, yet that typical time is not only printed on every sheet, but is in the mind of every player. It is precisely so with verse. It is true, of course, that the variations from regularity of rhythm are more numerous and conspicuous in verse than in music. The reason is obvious: the sounds of verse have constantly to effect a compromise between the typical rhythm to which they are set and the irregular stress-and time-variations of human speech, while music has no such complicated task. Lyrical verse, being closest to music, keeps the typical rhythm freest from interruption; and it is worthy of remark that Mr. Liddell's study of English rhythms is based very largely on those of a non-lyrical character, although he himself points out (p. 271) that these are the least regular. The drama is dominated, most of all forms of verse, by the necessity of representing natural human speech; hence it is not the place to look for the fundamental laws of verse at their purest.