¹ In the following 21 MSS the conclusion of the Clerk's Tale follows this arrangement: Dl, Lc, Ld², Fi, Gl, Ii, Mg, Sl¹, Nl, Bw, Ry², Ha², Mm, Pw, Ra², Ph³, Ra⁴, Tc², Ld¹, Ll¹, and To. Also Py, though it contains the mention of the Wife of Bath (E 1170–1176), agrees with these twenty-one MSS in the order of stanzas in the Envoy. I take this opportunity to express my thanks to Professor Manly for his courtesy in answering my queries in regard to the readings of several of these MSS in the matter of the Clerk's Envoy and also the couplet in the Merchant's Tale (E 1305–6). Since this paper was put in type I have also checked the list with that given in Sir Wm. McCormick's Manuscripts of Chaucers' Canterbury Tales (Oxf., 1933), p. xxxi.

² I avail myself of the list of MSS containing the Host's stanza, carefully compiled by Tatlock (The Harl. MS. 7334 and Revision of the Cant. Tales, Ch. Soc. 2nd Ser. 41, p. 24 note), citing the MSS for the reader's convenience according to the abbreviations adopted by Manly and Robinson: ten of Manly's Class i, viz. El, Gg, Dd, Ds¹, Ad¹, Cn, Bo², En¹, En³, Ma; five of Manly's Class iia, viz. Ha³, Ln, Ne, Py, Tc²; three of Manly's Class iic, viz. Bw, Ha², Ry²; three of Manly's “Very irregular MSS,” viz. Ch, Hg, Se. It occurs also in the single-tale Naples MS.

³ The Eight-text Ed. of the Cant. Tales, Ch. Soc. 2nd Ser. 43, p. 25.

⁴ Tatlock, Development and Chron., pp. 200–201.

⁵ H. B. Hinckley observes: “For though the Merchant, in his Headlink, begins with words of great bitterness about women, the misogyny of his Tale itself is not consistently maintained” (“The Debate on Marriage in the Cant. Tales,” PMLA, xxxii, 300).

⁶ Cant. Tales of Geoff. Chaucer, p. 596.

⁷ Mr. C. Robert Kase, in his “Observations on the Shifting Positions of Groups G and DE in the MSS of the Cant. Tales” (Three Chaucer Studies, New York, 1932, iii, p. 69), discusses the absence of the Merchant's Prologue in the MSS of the “inedited” type (his “Class A”). He supposes that the archetypal scribe was unable to find a copy of this prologue in his exemplar. By drawing what seems to be the natural inference from this fact: viz., that the Prologue of the Merchant's Tale was composed later, he would have simplified his explanation and reinforced his general conclusions. Mr. Kase's study of the manuscript evidence as to the order of the Tales has been most helpful to me in preparing this paper.

⁸ The most convenient tabulation of the arrangement of the Cant. Tales in existing MSS is that presented by Mr. Kase (Three Chaucer Studies, iii, 32–36). I note that in addition to these 22 MSS the sequence E² D is also found in the fragmentary Paris MS.

⁹ Mod. Phil., viii, 165–186; 305–334.

¹⁰ For a discussion of the date of Deschamps' composition of the Miroir see Lowes, p. 165, 166 and the footnotes attached. Deschamps refers in the Miroir to events in the year 1385. Raynaud, the editor of the Miroir, finds reason for believing that he stopped work on it in 1389. In any case, Chaucer could hardly have become acquainted with Deschamps' poem until the early 90's, that is, during the later period of the Canterbury Tales.

¹¹ For the Merchant's Tale I quote the lines as numbered in parentheses in Skeat's edition, instead of the Group E numbers, in order to show more clearly the position which they occupy in this Tale.

¹² Chaps. 17, 18, 20, 21, 35, 37–39 and 43.

¹³ For a list of these thirteen MSS see Manly's ed. of the Cant. Tales, p. 576.

¹⁴ Another unmistakable point of contact between the Merchant's Tale and the Wife's Prologue is the couplet,

which is to be compared with the Wife's declaration (D. 111–112),

Lowes regards these lines in the Merchant's Tale as “directly reminiscent of the Wife of Bath's Prologue” (p. 175), and this may be true, although it would be possible to consider either of the two passages as a reminiscence of the other. But this instance of interrelation is distinctly different from the one discussed above. For in this case the lines in question occur in all MSS of the Wife's Prologue, whereas in the other they are plainly a later insertion. Probably, therefore, they occurred in the text of the Wife's Prologue in its original form, which, according to our hypothesis, preceded the composition of the Merchant's Tale. Moreover, though the context in the Merchant's Tale (lines 197–222) shows clear dependence on Deschamps, there is no suggestion for these lines in the Miroir.

¹⁵ Chaucer Studies, iii, 435.

¹⁶ Oxford Chaucer, v, 359.

¹⁷ Develop. and Chron. of Chaucer's Work, p. 205.

¹⁸ “A Conjecture on the Wife of Bath's Prologue,” J.E.G.Ph., xxiv, 512 ff.

¹⁹ The 12-line Pardoner-Shipman link found in a considerable group of MSS. (cf. Manly, Cant. Tales by Geoff. Chaucer, p. 85) is universally regarded as spurious.

²⁰ The Seiden MS. was seized upon eagerly by earlier editors because it alone linked B¹ and B². But the arrangement of the Tales presented in this MS (A E¹ D E² F¹ B¹ B² G C F² H I) is sufficient to deprive it of any authority.

²¹ The Eight-text Ed. of the Cant. Tales, etc., Ch. Soc. 2nd Ser. 43, p. 30.

²² A. Brusendorff, The Chaucer Tradition, p. 65.

²³ En² is defective owing to the loss of leaves after E 1166.

²⁴ Cf. C. Robert Kase, op. cit., pp. 31 ff.

²⁵ These 20 MSS are: Bo¹, Bw, Dl, En², Ha², Ha³, He, Lc, Ld², Ln, Mg, Mm, Ne, Nl, Ph², Pw, Py, Ry¹, Ry² and Sl¹.

²⁶ For the text of this Clerk-Franklin link see Manly, p. 84, and Robinson's ed., p. 1009. One of the chief objections to this Clerk-Franklin link is presented by the riming of do with words with an o deriving from O.E. ä. But Manly's remark that this combination of rimes “belongs to the Fifteenth century” is not quite accurate since it is of frequent occurrence in the Confessio Amantis, and in one instance in the Monk's Tale we find thereto:mo:wo:go (B 3510–3515).