28. Macbeth cannot, in view of its obvious allusions to James I., be of earlier date than 1603. The style and some trifling allusions point to about 1605 or 1606, and a hint for the theme may have been given by Matthew Gwynne's entertainment of the Tres Sibyllae, with which James was welcomed to Oxford on Aug. 27, 1605. The play was revived in 16io and Simon Forman saw it at the Globe on April zo. The only extant text, that of the First Folio, bears traces of shortening, and has been interpolated with additional rhymed dialogues for the witches by a second hand, probably that of Thomas Middleton. But the extent of Middle ton's contribution has been exaggerated; it is probably confined to act iii. sc. 5, and a few lines in act. iv. sc. 1. A ballad of Macdobeth is mentioned in the Stationers' records during 1596, but is not known. It is not likely that Shakespeare had consulted any Scottish history other than that included in Raphael Holinshed's Chronicle; he may have gathered witchlore from Reginald Scot's Discoverie of Witchcraft (1584) or King James's own Demonologie (1599) 29. The entry of King Lear in the Stationers' Register on Nov. 26, 1607, records the performance of the play at court on Dec. 26, 1606. This suggests 16o5 or 1606 as the date of production, and this is confirmed by the publication in 1605 of the older play, The True Chronicle History of King Leir, which Shakespeare used as his source. Two Quartos of King Lear were published in 1608, and contain a text rather longer, but in other respects less accurate, than that of the First Folio. The material of the play consists of fragments of Celtic myth, which found their way into history through Geoffrey of Monmouth. It was accessible to Shakespeare in Holinshed and in Spenser's Faerie Queene, as well as in the old play.
31. In the case of Coriolanus the external evidence available is even scantier, and all that can be said is that its closest affinities are to Antony and Cleopatra, which in all probability it directly followed in order of composition. Both plays, like Julius Caesar, are based upon the Lives of Plutarch as translated by Sir Thomas North.
32. There is no external evidence as to the date of Timon of Athens, but it may safely be grouped on the strength of its in ternal characteristics with the plays just named, and there is a clear gulf between it and those that follow. It may be placed pro visionally in 1607, although some critics put it next after Lear. The extraordinary incoherencies of its action and inequalities of its style have prevented modern scholars from accepting it as a finished production of Shakespeare, but there agreement ceases.
It is sometimes and perhaps most reasonably regarded as an in complete draft for an intended play; sometimes as a Shakespear ian fragment worked over by a second hand either for the stage or for printing in the First Folio ; sometimes, but not very plau sibly, as an old play by an inferior writer which Shakespeare had partly remodelled. It does not seem to have had any relations to an extant academic play of Timon which remained in manuscript until 1842. The sources are partly in Plutarch's Life of Marcus Antonius, partly in Lucian's dialogue of Timon or Misanthropos, and partly in William Paynter's Palace of Pleasure (1566).
33. Similar difficulties, equally unsolved, cling about Pericles. It was entered in the Stationers' Register on May 20, 1608, and published in 1609 as "the late and much admired play" acted by the king's men at the Globe. The title-page bears Shakespeare's name, but the play was not included in the First Folio, and was only added to Shakespeare's collected works in the Third Folio, in company with others which, although they also had been print ed under his name or initials in quarto form, are certainly not his. In 1608 was published a prose story, The Painful Adventures of Pericles Prince of Tyre. This claims to be the history of the play as it was presented by the king's players, and is described in a dedication by George Wilkins as "a poore infant of my braine." The production of the play is therefore to be put in 1608 or a little earlier. It can hardly be doubted on internal evidence that Shake speare is the author of the verse-scenes in the last three acts, with the exception of the doggerel choruses. It is probable, although it has been doubted, that he was also the author of the prose scenes in those acts. To the first two acts he can at most only have contributed a touch or two. It seems reasonable to suppose that the non-Shakespearian part of the play is by Wilkins, by whom other dramatic work was produced about 1607. The prose story quotes a line or two from Shakespeare's contribution, and it follows that this must have been made by 1608. The close resemblances of the style to that of Shakespeare's latest plays make it impossible to place it much earlier. But whether Shake speare and Wilkins collaborated in the play, or Shakespeare par tially rewrote Wilkins, or Wilkins completed Shakespeare, must be regarded as yet undetermined. Unless there was an earlier Shakespearian version now lost, Dryden's statement that "Shake speare's own Muse her Pericles first bore" must be held to be an error. The story is an ancient one which exists in many ver sions. In all of these except the play, the name of the hero is Apollonius of Tyre. The play is directly based upon a version in Gower's Confessio Amantis, and the use of Gower as a "pre senter" is thereby explained. But another version in Laurence Twine's Patterne of Painefull Adventures (c. 1576), of which a new edition appeared in 1607, may also have been consulted.