Page:The Library, volume 5, series 3.djvu/209

 MIRACLE CYCLES. 197 Of these 58 are agreements of W K against the rest, that is of variants in F, and 20 agreements of P H, that is variants in /3. 1 This leaves about 30 readings which, though not necessarily contradict- ing our scheme, are worth attention. The most important and numerous group is that to which we must look to determine the positions of B and D. Four agreements of B D against the rest suggest a common source for these two manuscripts. On examination, however, three of these prove to be easily explained coincidences, and the fourth, though less obvious, and therefore to be allowed some weight, is also capable of explanation. 2 The evi- dence in favour of a B D source is, therefore, not strong. On the other hand, there are four readings in which PHD agree against B W K and contra- dicl: our scheme, and eight in which P H B agree against D W K and support it. Now, on examina- tion the four anomalous cases all appear capable of ready explanation as coincidences due to obvious emendations or to alterations of obsolete or un- expected readings of the original. 3 The soundness 1 There is one case common to these two classes, namely an instance of P H : B : D : W K. 2 The instances of D B are : 278 P (&c.) This theffys comyng to abyd (D B theefe his). 476 P (&c.) Goddis peple that stondis vs bye (D B stande K standeth). 522 P (&c.) For thrughe his myght and his maistrye (D B maiestie). 626 P (&c.) Reigne no longer thowe ne maye (D B lenger nowe thou). 3 The instances of P H D are : 97 P (&c.) Men buryed in graue as ye may see (B W K graues). 167 P (&c.) Iff I be Crist nowe leuys ye (B ye me W K you me). 1 78 P (&c.) And we shall knelingon oure knen (B W K kneele vpon). 534 P (&c.) Conuertis to me most myghty (B W K mightelie).