Page:The Library, volume 5, series 3.djvu/191

 MIRACLE CYCLES. 179 the discussions so far is that the collective manu- scripts fall into two main groups, one of which is represented by the youngest manuscript H alone. It is also clear that within the larger group a specially close relation unites W and K, the two manuscripts purporting to be by the same scribe. The position of P has not been seriously discussed, and indeed the importance of this, by far the earliest text we possess, for the textual criticism of the cycle does not seem to have been recognized. When Deimling wrote its existence was not generally known, but it was actually accessible to Hemingway in a printed edition, and that he should have ventured to form an opinion upon the relative merits of the cyclic manuscripts without taking its evidence into consideration is inexcusable. In the five collective manuscripts we clearly have texts of the whole cycle as officially recog- nized. There is nothing in any of them to suggest that they were compiled, like the York ' register,' by transcribing a number of separate play-books in the hands of the various guilds. Where diver- gencies of tradition appear they seem to affect the whole cycle, not merely individual plays. Now we have already, in my previous lecture, seen reason to believe that the ' original ' of the Chester cycle was an official copy in the possession of the corporation. That 'original' was presumably from time to time renewed, alterations being incor- porated in a fresh copy. From some such our present manuscripts must be descended. But their differences prove that they were not all transcribed from the same ' original,' but that they represent