^{Note 1 in page 30} M.L.N., lv (1940), 613–619.

^{Note 2 in page 30} For a more detailed statement of the situation and a list of the MSS composing this group, see my paper, “The Evolution of the Canterbury Marriage Group” (PMLA, xlviii, 1041 ff.)

^{Note 3 in page 31} The same opinion is expressed also in Vol. i, pp. 286 and 458.

^{Note 4 in page 34} Editors have laid too much stress on the topographical allusions in determining the order of the tales. Manly justly remarks: “Chaucer did not lay out a general plan for the whole journey to Canterbury and return and assign each block of tales to its proper place in this plan.... When he changed his intentions with regard to the use of a particular tale, he did not always remove all traces of its previous use” (ii. 490).

^{Note 5 in page 35} “The Bearings of the Shipman's Prologue,” JEGPh., xxxiii, 352–372.

^{Note 6 in page 35} “The Cant. Tales in 1400,” PMLA, l, 115–116.

^{Note 7 in page 35} Manly's definition of the b* MSS is as follows: “Throughout CT, Group b is associated with a variable number of irregular MSS which because of their continually fluctuating combinations cannot be assigned to any constant group” (ii. 79).

^{Note 8 in page 35} The single exception is Ha⁴, which ends abruptly at line 1185, leaving an unfinished couplet. Ha⁴, according to Manly, “is associated with the large composite group of MSS, forming with some member or members of this an independent subgroup, usually apart from the main line of tradition” (i. 221). In the order of tales (B¹DEF) Ha⁴ agrees with the MSS of group a as opposed to groups b, c, or d.

^{Note 9 in page 36} “Si jura l'âme de son père que chierement comparer il feroit cil outrage à tous ceulz de le cité,” Chroniques de J. Froissart, Livre Premier §663, ed. Simèon Luce (Soc. de l'Histoire de France 188), vii, 243.

^{Note 10 in page 37} Develop. and Chronol., pp. 147–148.

^{Note 11 in page 38} E 2419–40, F 1–8. Earlier editors designated these lines as the Merchant's Epilogue and the Squire's Prologue. But McCormick, Tatlock, and Manly rightly insist that they comprise a single unit, and indicate their function as a Merchant-Squire link.

^{Note 12 in page 38} The following MSS lack the ML endlink: El, Hg, Gg, the five MSS of group a and fourteen other MSS as follows: Ad¹, Ad³, Bo ¹, Ch, En³, Ha⁶, Hk, Ld¹, Ma, Nl, Ps, and To. I do not include the mutilated Ad² which has only 31–507 and 702–822 of the MLT, nor Mg which once contained the endlink but has lost B 744–1190.

^{Note 13 in page 38} There are nineteen of these: El, (Gg), the five MSS of group a and twelve other MSS as follows: Ad¹, Ad³, Bo², Ch, En³, Ha⁴, Ha⁶, Ld¹, Ps, Ry¹, Se, and To. Gg once contained this link but is now defective at this point. The two MSS of Bo ¹, which appear in the former list, shifted from an exemplar in which the tales were arranged according to type a to one of type d and thus missed both Merchant and Squire tales.

^{Note 14 in page 38} Ry¹ according to Manly “cannot be regarded as anything more than a conglomeration of tales” (i. 479) and Se “is so thoroughly contaminated that it is often nearly impossible to distinguish between its genetic relationships and those due to correction ... and is merely a bad 15 c. edition of no textual authority” (i. 496).

^{Note 15 in page 39} This shift in Chaucer's arrangement naturally resulted in some confusion on the part of the scribes. The significance of these shifts in the position of D and E was ably discussed in 1932 by Dr. C. R. Kase in his study, “Observations on the Shifting Positions of Groups G and DE in the MSS of the Cant. Tales,” Three Chaucer Studies (Oxford Press, New York). His conclusions have not thus far been refuted. Additional evidence in support of his general thesis is supplied by the materials made available in the Manly-Rickert volumes. It seems desirable to restate the essential basis of his argument, amending it by recognizing Chaucer's shift in the Marriage Group tales.

^{Note 16 in page 39} Gg also once contained this link, but no longer has it on account of the loss of leaves.

^{Note 17 in page 39} Group c is distinguished in general from the others by its lack of links. But it does show the ML endlink.

^{Note 18 in page 41} In three MSS (Ry² Bw and Ha²) these two 7-line stanzas are preceded by the Host stanza (E 1212^a–g) with the heading, “Here endith the Clerk of Oxenford And here begynneth the prolog of the ffrankeleyn.” The first scribe of Ln seems also to have intended this arrangement, for on f. 91^a he copied the Host stanza headed “þe prolog of the Frankelens tale” and then left space for 16 lines (See Manly i. 334).

^{Note 19 in page 41} Three group-d MSS (Bw, Ld², Ry²) have a spurious link of sixteen lines following E 2418 introducing the Wife of Bath (text in Manly-Rickert vi, 3), but with this we are not here concerned.

^{Note 20 in page 42} To^b inserts the Hengwrt version of F 1–8 between the Clerk's tale and the Merchant's Prologue (E 1213–44). In this case the resourceful scribe altered “Frankeleyn” to “Merchaunt.”

^{Note 21 in page 42} Preserved in Bw, Dl, Ha², Lc, Ld², Mg, Nl, Ry², Sl¹. It occurs also in Fi, but is there misplaced between E² and F². Also it stood originally in En², but the text has disappeared through the loss of leaves in the MS (see Manly i. 137).

^{Note 22 in page 42} To object that “the rhymes of 11, 2, 4, and 5 indicate that the combination belongs to the fifteenth century” (Manly, Cant. Tales of Geoffrey Chaucer, 1928, p. 85) is a serious misstatement of fact. Skeat's explicit statement on this point is strictly correct: “When the long o is absolutely final, as in go, do, Chaucer considers these as permissible rimes, and pairs them together freely” (Oxf. Chaucer, vi, xxxv).

^{Note 23 in page 43} “Three Notes on the Text of the Cant. Tales,” MLN, lvi (1941), 169.

^{Note 24 in page 43} See my observations, PMLA, xlviii, 1044–45, and A. C. Baugh, Mod. Phil. xxxv, 24–26.

^{Note 25 in page 46} This essentially is the reading in eighteen MSS: El, group a, Ad¹, Ad³, Ha⁴, Ha⁵, Bo², En³, Ch, Ps, To^a, Se, Ld¹, Ry¹.

^{Note 26 in page 49} ii. 43. Miss Rickert makes the same observation: “b is more frequently associated with a but only occasionally with c and never with d alone” (ii. 486.)

^{Note 27 in page 49} J. S. P. Tatlock, “The Canterbury Tales in 1400” (PMLA, l, 101).

^{Note 28 in page 50} Ibid., p. 131. Manly's Statement on this point is more reserved: “Not only are the prevalent patterns of arrangements in the groups abcd not the work of Chaucer; there is no single MS or small group the order of which can be ascribed to him” (ii. 476).