Page:The Library, volume 5, series 3.djvu/384

 370 PROBLEMS OF THE ENGLISH there is no reason why more than one stanza should not have been used in the original composition of a single play, an author would not change from one to another without some rational cause. It follows that wherever a change of stanza occurs without discoverable reason we are justified in supposing that we have not got the play in its original form. This canon has been commonly assumed by critics, and I do not think, if it is reasonably applied, that anyone is likely to quarrel with the results. We will now see what sort of evidence may be expefted from each of the three chief sources of information which I mentioned before. The manuscript is written on paper this happens to be fortunate and the size is quarto. 1 Almost all the leaves have been detached and mounted on guards, but a set of late signatures, in conjunction with the water-marks, enables us to reconstruct the original quires with all but absolute certainty. With the exception of one play the whole original text is in a single hand. 2 This is a good plain hand of the second half of the fifteenth century, showing marked East-Anglian peculiarities ; near the middle of the manuscript occurs the date 1468, and there is no reason to doubt that this is actually the date of writing. The play of the Assumption, which immediately precedes ' Dooms- day,' is in a different hand, the home of which is less clear. Halliwell (p. 418) assigned this hand to 1 That is to say, each leaf is one quarter of a sheet. Strictly speaking, even a paper manuscript has no format. 2 There are a few incidental additions which are not in the same hand as the text.