and full stops, which are not always inserted in the right places. And I have broken up the chapters into convenient paragraphs.

§ 8. A very curious point about this piece is the fact which I was the first to observe, viz. that the initial letters of the various chapters were certainly intended to form an acrostic. Unfortunately, Thynne did not perceive this design, and has certainly begun some of the chapters either with the wrong letter or at a wrong place. The sense shews that the first letter of Book I. ch. viii. should be E, not O (see the note); and, with this

correction, the initial letters of the First Book yield the words—MARGARETE OF.

In Book II, Thynne begins Chapters XI and XII at wrong places, viz. with the word 'Certayn' (p. 86, l. 133), and the word 'Trewly' (p. 89, l. 82). He thus produces the words—VIRTW HAVE MCTRCI. It is obvious that the last word ought to be MERCI, which can be obtained by beginning Chapter XI with the word 'Every,' which suits the sense quite as well.

For the chapters of Book III, we are again dependent on Thynne. If we accept his arrangement as it stands, the letters yielded are—ON THSKNVI; and the three books combined give us the sentence:—MARGARETE OF VIRTW, HAVE MERCI ON THSKNVI. Here 'Margarete of virtw' means 'Margaret endued with divine virtue'; and the author appeals either to the Grace of God, or to the Church. The last word ought to give us the author's name; but in that case the letters require rearrangement before the riddle can be read with certainty.

After advancing so far towards the solution of the mystery, I was here landed in a difficulty which I was unable to solve. But Mr. H. Bradley, by a happy inspiration, hit upon the idea that the text might have suffered dislocation; and was soon in a position to prove that no less than six leaves of the MS. must have been out of place, to the great detriment of the sense and confusion of the argument. He very happily restored the right order, and most obligingly communicated to me the result. I at once cancelled the latter part of the treatise (from p. 113 to the end), and reprinted this portion in the right order, according to the sense. With this correction, the unmeaning THSKNVI is resolved into the two words THIN USK, i.e. 'thine Usk'; a result the more remarkable because Mr. Bradley had previously hit upon Usk as being the probable author. For the autobiographical details exactly coincide, in every particular, with all that is known of the career of Thomas Usk, according to Walsingham, the Rolls of Parliament, and the continuation of Higden's Polychronicon by John Malverne (ed. Lumby, vol. ix. pp. 45-6, 134, 150, 169); cf. Lingard, ed. 1874, iii. 163-7.

The date of the composition of this piece can now be determined without much error. Usk was executed on March 4, 1388, and we find him referring to past events that happened towards the end of 1384 or later. The most likely date is about 1387.

I here append an exact account of the order of the text as it appears in Thynne; every break in the text being denoted, in the present volume, by a dark asterisk.

Thynne's text is in a correct order from p. 1 to p. 118, l. 56:—any mouable tyme there (Th. fol. 354, col. 2, l. 11)^[5].

(1) Next comes, in Thynne, the passage beginning at p. 135, l. 94:—Fole, haue I not seyd—and ending at p. 143, l. 46:—syth god is the greatest loue and the (Th. fol. 356, back, col. 1, l. 5).

(2) Next, in Thynne, the passage beginning at p. 131, l. 97:—ne ought to loke thynges with resonnyng—and ending at p. 132, l. 161, at the end of a chapter (Th. fol. 356, back, col. 2, last line).

(3) Next, in Thynne, the passage beginning at p. 124, l. 8:—Now trewly, lady—and ending at p. 128, at the end of the chapter (Th. fol. 357, last line).

(4) Next, in Thynne, the passage beginning at p. 132, new chapter:—Uery trouth (quod she)—and ending at p. 135, l. 94:—that shal bringe out frute that (Th. fol. 358, back, col. 1, l. 25).

(5) Next, in Thynne, the passage beginning at p. 118, l. 56:—is nothyng preterit ne passed—and ending at p. 124, l. 7:—euer to onbyde (Th. fol. 360, col. 1, l. 24).

(6) Next, in Thynne, the passage beginning at p. 128, new chapter:—Nowe, lady (quod I) that tree to set—and ending at p. 131, l. 97:—vse ye (Th. fol. 360, back, col. 2, l. 9).

(7) Lastly, the text reverts to the true order, at p. 143, l. 46, with the words:—greatest wisdom (Th. fol. 360, back, col. 2, l. 9. as before). See The Athenæum, no. 3615, Feb. 6, 1897.

It is not difficult to account for this somewhat confusing dislocation. It is clear that the original MS. was written on quires of the usual size, containing 8 folios apiece. The first 10 quires, which we may call a, b, c, d, e, f, g, h, i, and k, were in the right order. The