¹ See “Verse-Sentence Patterns in English Poetry,” Philological Quarterly, vii, 283.

² The Tragedies of Shakespeare, 1925; The Histories and Poems of Shakespeare, 1925; The Comedies of Shakespeare, 1927: Oxford University Press.

³ For instruction in the actual use of the method a much fuller explanation would be required.

⁴ The complete table may be found in Neilson and Thorndike's Facts A bout Shakespeare, 1915, p. 71. The parts I have quoted from this condensed and convenient table were originally derived from G. König, “Der Vers in Shakesperes Dramen,” Quellen und Forschungen, lxi (1888) and J. K. Ingram, “On the ‘Weak Endings’ of Shakespere …,” Transactions of the New Shakespere Society (1874), Series i, Part ii.

⁵ Here the run-on lines correspond to my divergence types and the end-stopped lines to my parallel types.

⁶ It should be noted that I have not asserted that Othello is seventy times further on the scale of development than is Love's Labour's Lost, but only that my method reveals a difference seventy times greater than that revealed by the test of the run-on line.

⁷ A. C. Bradley (Shakespearian Tragedy, p. 474) states that, in computing the proportion of speeches ending within the line, he does not count speeches which cover less than a single line. In the Neilson and Thorndike table, however, these part-line speeches must have been counted. Of speech-endings in connection with speeches continued from a previous line, 1 have been able to find only 21 in the whole play of Love's Labour's Lost. Since there are more than three hundred verse speeches in the third act alone, it is quite evident that the 10 per cent of speeches ending within the line must include the types of speech-ending found in my illustrative passage. But even if I should use Professor Bradley's corrected method, the speech-ending test would here be unsatisfactory. There would be, to be sure, no evidence of a premature outcropping of Shakespeare's later style, but there would be no positive evidence of his early style. Furthermore, an examination of the 21 speech-endings which Professor Bradley's method would record reveals only two divergence line-types. The others are C and — C lines. On the other hand, in Act v, Scene 2, lines 709 to 780, inclusive, there are 7 of these major-divergence types, an unusually high proportion for a passage in this play, and no speeches of any sort ending within the line.

⁸ The L line is one which is either one construction or is composed of fragments connected by the bond of restrictive modification. In either case itover-runs the end of the line.

⁹ I am confining my explanations of line-markings to the types here used. An explanation of other types may be found in my previous article, but I have subsequently subjected several of these other types to considerable revision, and I shall probably make still further sub-divisions before I apply them to the problems of chronology and collaboration involved in Shakespeare's early plays. The L and C types, however, have already been explained in this article, and the — C and C — types may easily be understood from the explanations of the —C— and —C / types.

¹⁰ Here it is evident that an independent-clause group runs into the line from previous lines and comes to an end with the word “burden.” At this point another group begins and runs out of the line without impediment.

¹¹ Here we find essentially the same relation between line and sentence structure as in line 6 except that the phrase “like a cipher” partly interrupts the free flow of thought from the introductory “and therefore” to the main subject and verb “I multiply” at the end of the following line. A very little reflection should make it evident that the two lines are essentially alike. In contrast to the prevailing style of the early dramas, where the author maintains control over line and sentence by making their main joints coincide, he here seems to write his sentence almost as though he had forgotten all about the line, and yet fulfills the line requirements with the off-hand ease of a supreme master of metrics.

¹² Here the line is divided between a portion of an independent clause and a portion of a non-restrictive dependent clause. In classifying c lines, I regard this clause relationship as essentially equivalent to the juncture of non-restrictive dependent clauses. An illustration of the latter is found in the following line.

¹³ The number of divergence types is not excessive for a typical passage in The Winter's Tale, but a typical passage would yield a few of the parallel types and less of the intermediate types which are not involved in the present study.

¹⁴ It will be seen that my figures indicate only 8 plays of a later date of composition than that of All's Well. If The Two Noble Kinsmen proves to be, in part at least, a Shakespearian play, I shall doubtless recognize 9 plays of a later date. In the chronological position which I ascribe to the play I am in substantial agreement with Chambers, Malone, and Fleay. On the other hand, Drake recognizes 20 plays of a later date, Delius 19, Furnivall and Dowden 16. See The New Shakspere Society's Transactions, Series i, Part i, p. 10; and Edward Dowden's Shakespeare, A Critical Study of His Mind and Art, pages x and xvii.

¹⁵ As Mr. Fleay stated in 1874 (Transactions of the New Shakespere Society, Series i, Part i) it has long been generally admitted by scholars that the first two acts of Pericles are not by Shakespeare. My analysis, however, affords what I believe to be the most convincing demonstration of two different styles in the play. In ascribing parts of the Chorus to Shakespeare, I seem to be in disagreement with the majority of scholars.

¹⁶ Opinions concerning the authorship of Henry VIII are too numerous to refer to in this brief article. It happens that my division of the play is substantially the same as that made by James Spedding in his article entitled “On the Several Shares of Shakespeare and Fletcher in the Play of Henry VIII” (The New Shakespere Society's Transactions, Series i, Part i, p. 1), originally published in The Gentleman's Magazine for August, 1850, under the title: “Who Wrote Shakespere's Henry VIII?”

¹⁷ I have already made a study, by 500-word samples, of a considerable number of plays by Chapman, Fletcher, and Massinger. So far, I have found only one play, The Maid of Honor by Massinger, with a quotient at all comparable to that of the high-divergence portions of Henry VIII. Unfortunately the date of composition for this play does not seem to be known. Arthur Symons states (Philip Massinger, i, 230) that the play has been variously assigned to “before 1622,” to 1628, and to 1631. If Massinger's stylistic development was at all similar to Shakespeare's, it is highly improbable that the younger dramatist could have written in the style of the high-divergence portions of Henry VIII before 1613, the year in which we have good reason to believe that Henry VIII was acted.