^{page 244 note 1} In his masterly Catalogue of English Mediaeval Rolls of Arms (CEMRA) (Oxford, 1950)Google Scholar(Aspilogia, i) pp. xiv-xv, Sir Anthony Wagner has provided the following classification: 1. Illustrative rolls; 2. Occasional rolls; 3. General rolls; 4. Local rolls; and 5. Ordinaries. However, for the thirteenth century, only the Matthew Paris Shields in England and the Chansonnier du Roi in France belong in the first category, and Cooke's Ordinary, c. 1340, is the earliest known list of its type. For the French rolls, see Adam-Even, Paul, ‘Catalogue des armoriaux francais imprimes’, Nouvelle Revue Hiraldique, N.S., i (1946), 19–29Google Scholar.

^{page 244 note 2} Rolls of Arms, Henry III. The Matthew Paris Shields, c. 1244-59, ed. Tremlett, T. D..Google ScholarGlover's Roll, c. 1253-8 and Walford's Roll, c. 1273, ed. London, H. S. (London, 1967)Google Scholar(Aspilogia, ii), pp. 91, 100. Cf. Perceval, C. S., ‘Two Rolls of Arms of the Reign of King Edward the First’, Archaeologia, xxxix (1864), 397Google Scholar:‘the shields were filled up pretty much as the arms occurred to the memory of, or were suggested to, the compiler’. While they were perhaps not the only authors of rolls of arms, it is reasonable to suppose that heralds were engaged in this work by the end of the thirteenth century; see Wagner, A. R., Heralds and Heraldry in the Middle Ages, 2nd ed. (Oxford, 1956), pp. 54–5Google Scholar.

^{page 244 note 3} Aspilogia, ii, p. 91Google Scholar. Perceval, p. 397, discusses a few ‘attempts at arrangement according to artistic effect, not inconsistent however with the absence of more methodical composition’,

^{page 244 note 4} Wagner, , Heralds and Heraldry, p. 50Google Scholar: ‘As time goes on, we meet with rolls which seem to be no more than compilations from older collections of various dates with perhaps a few contemporary coats added.’ Wagner specifically mentions the Parliamentary Roll, c. 1312 (CEMRA, pp. 42-50), as having ‘formed the foundation of others’ (p. 52). Perceval, p. 391, noted that of the 677 items in St. George's Roll (E, edited by him, pp. 418-40), about 350 occur in Charles's Roll (F, pp. 399-417, 486 shields); see also CEMRA, p. 19. Denholm-Young, N., History and Heraldry 1254 to 1310. A Study of the Historical Value of the Rolls of Arms (Oxford, 1965), p. 92Google Scholar, n. I, suggests that ‘both probably derived from a painted thirteenth-century original’.

^{page 245 note 1} Greenstreet, James and Russell, Charles, ‘Dering Roll’, Jewiti's Reliquary, xvi (1875), 135–40, 237-40Google Scholar; xvii (1876), 11-16, 209-12; xviii (1877), 23-8, 89-92, 171-5 (324 shields). The Fitz-william Version of the Heralds'Roll was edited by Greenstreet, James as ‘Planchés Roll’, The Genealogist, N.S., iii (1886), 148–55, 240-4Google Scholar; iv (1887), 17-22, 197-203; v (1888), 173-9. Concerning trie latter roll, Wagner says: ‘The English portions have much in common with the Dering Roll’ (CEMRA, p. 11; see also, p. 14: ‘[A] and the Fitzwilliam Roll have many coats in common but the exact relation between them has not yet been established’).

^{page 245 note 2} Greenstreet, James, ‘The Original Camden Roll of Arms’, J.B.A.A., xxxviii (1882), 309–28Google Scholar.

^{page 245 note 3} CEMRA, pp. 9-14. The Earl of Bedford's Version and Everard Green's Version are later compilations.

^{page 245 note 4} College of Arms MS. B 29, pp. 20-7 (unpublished).

^{page 245 note 5} Cambridge, Fitzwilliam Museum MS. 297, ‘early or middle 15th century’ (CEMRA, p. 11); College of Arms MS. Vincent 165, ff. 131-52b, copied in 1590; College of Arms, Muniment Room, Box 15, Roll 24, 16th-century copy; Society of Antiquaries MS. 664, vol. ii, ff. 1-12b, Roll II, a facsimile made c. 1640 (CEMRA, p. II).

^{page 245 note 6} CEMRA, p. 10.

^{page 245 note 7} FW 104, Henrie de Hastings (HE 81, Rauf de Hastinge); FW 176, Rafe de Hastange (HE 162, Roberd de Hastang). Major variants: FW 65 (HE 49), 101 (78), 117 (97), 152 (148), 176 (162), 289 (188). Except for FW 289, which does not appear in either A or D, the preceding items are all noted below. Other differences consist of variations in tincture or number of charges, and omission or addition of a secondary charge. I am indebted to Sir Anthony Wagner, Garter Principal King of Arms, for providing me with a copy of his personal transcript of HE which made a comparison of FW and the latter version possible.

^{page 246 note 1} CEMRA, p. 9. Cf. Denholm-Young, pp. 45-6: ‘The existing Heralds’ Roll… was conflated with much extraneous material to form, in the Fitz-william Version, a roll of seventeen membranes containing some 697 [sic, for 696] shields painted in colour.… the Heralds’ Roll itself appears to have been a compact and well-ordered document, but the ‘versions’ do not seem to have been inspired by any principle other than that of mere agglomeration’

^{page 246 note 2} CEMRA, p. 9. ‘A date c. 1270-80 appears probable for the compilation. The dating of the foreign portions raises some difficulties. Dr. Paul Adam-Even inclines to the view that it comprises two sections, one compiled c. 1260, the other rather later. It has been suggested that the latter section may be as late as the early 14th century. If this were established the dating of the compilation of the roll as a whole might have to be revised’ (p. II).

^{page 246 note 3} CEMRA, p. II.

^{page 247 note 1} Greenstreet and Russell, p. 137; CEMRA, p. 14. Wagner adds: ‘The shields in the last 6 rows, Nos. 289–324, belong wholly or for the most part to northern France and the Low Countries, deriving probably from the same source as the foreign coats i n the Fitzwilliam Roll.’

^{page 247 note 2} History and Heraldry, pp. 64-89.

^{page 247 note 3} A 1, 15, 48, 79, 212, 220, 221, 235, 256, 269, 307, and 311. The defective shields A 212, 256, and 307 may correspond to FW 124, 342, and 481, respectively. For A 311, see the note to FW 376. A 269 resembles D 185.

^{page 248 note 1} In the captions above the shields, the titles Chastelein, Cunte, Due, Emperur, Prince, Rey and Seynt are found, Sire being used for twenty-two foreign coats and for Alfonso. In the blazon, on the other hand, the names of the English lords and knights are generally preceded by Munsire, the title Sire being regularly used for foreigners. However, the Englishmen D 53* (an asterisk indicates the order of items in the blazon which does not always correspond to that of the shields) and D 104* have no titles, and D 57*, a foreigner, is designated as Munsire). From D 71* to the end of the blazons, all names of lords and knights, whether English or foreign, are preceded by Munsire, except the foreigner D 80*.

^{page 247 note 2} D 24, 54, 73, 75, 77, 84, 86, 87, 91, 92, 97-100, 103, 104, 123, 154, 159, 185, 186, 187, 191,193, 195, 210, 220, 231, 232, 234, 235, 237, 239, 242-6, 248, 250-3, 256, 261, 262, 268-70. Five shields (D 73, 92, 97, 99, 104) are completely effaced and may have been derived from FW; the absence of a corresponding blazon for these items as well as for D 84, 91, 98, 100, 103, 158, and 185, all of which are uncaptioned and partially effaced, suggests that these shields were found in this state by the author of the verbal blazon. However, see note to FW 28. Greenstreet, D, p. 310, cites the omitted blazon corresponding to D 9 as proof that the verbal descriptions were composed ‘subsequent t o the assertion of Edward's claim to suzerainty over Scotland, namely close upon A.D. 1286’. Firsthand knowledge cf the arms the Camden copyist added to his list is not to be ruled out, of course, but he also invented two coats (D 11 and 24; see the notes to FW 26 and 375) on his own, it would appear.

^{page 247 note 3} See the notes to the Camden Roll in my edition of Eight Thirteenth-Century Rolls of Arms in French and Anglo-Norman Blazon (University Park, Pa. and London, 1973).Google Scholar The rolls in this volume are: the Bigot Roll; Glover's Roll, St. George's Version (two copies); Walford's roll, Charles’ Version (one copy) and Leland's Version (two copies); the Camden Roll; the Chifflet-Prinet Roll; the Falkirk Roll, Thevet's Version; the Nativity Roll; and the Siege of Caerlaverock.

^{page 249 note 1} D 89 and 126 refer to FW 606 and 637, respectively, and, consequently, belong there, too, but there are significant differences in the arms in question (see below).

^{page 249 note 2} FW 39, Earl of Gloucester (d. 1262); 44, Earl of Salisbury (d. 1257) 547, Simon de Montfort, Earl of Leicester (d. 1265); 48, Earl of Chester (d. 1237); 54, Earl of Aumale (d. 1260); 56, Earl of Winchester (d. 1244); 59, Earl of Devon (d. 1262). See also FW 89, the Scottish earl, Adam de Kilconquhar (d. 1270); 102, Baldwin Wake (d. 1263); 201, Elias Giffard (d. 1248). Every one of these items, except the last, was carried over by D who even repeats FW 59 (D 134, 214). On the other hand, A copies only FW 102 and 201, leaving out the deceased earls entirely. However, he begins his roll with A 1, Richard, fiz le Rey, that is, Richard Fitz Roy or de Douvres, bastard son of King John, who died in late 1245 or early 1246 (B 82). On the possible significance of this addition in A, con- suit Denholm-Young, pp. 12, 64-5, 74. Many of the same earls appear in the Parliamentary Roll (N, edited by Barron, Oswald, ‘The Parliamentary Roll of Arms’, The Genealogist, N.S., xi (1885), 108–16, 174-82, 238-44Google Scholar; xii (1886), 59-62, 133-6, 206, 2fi, 268-82), items 1035-43, with the caption'Ces sount les nons e les armes abatues de grandseignors’. On this practice, see Denholm-Young, pp.46, n. 1; 92, n. 3; 97.

^{page 249 note 3} See above, p. 246, n. 2.

^{page 249 note 4} Stevenson, J. H., Heraldry in Scotland (Glasgow, 1914), i, 60Google Scholar.

^{page 249 note 5} The Complete Peerage (London, 1910-1959), ii, 202Google Scholar.

^{page 249 note 6} The Complete Peerage, vi, 345Google Scholar; Aspilogia, ii, p. 144Google Scholar.

^{page 249 note 7} The Complete Peerage, v, 672Google Scholar; Aspilogia, ii, p. 180Google Scholar. Richard succeeded his brother, not his father (correct CEMRA, p. 14).

^{page 249 note 8} CEMRA, p. 17; Aspilogia, ii, p. 126Google Scholar.

^{page 249 note 9} The Complete Peerage, iii, 417Google Scholar; Aspilogia, ii, p. 154Google Scholar.

^{page 249 note 10} The Complete Peerage, iv, 139Google Scholar; Denholm-Young, p. 65, n. 1.

^{page 250 note 1} The Complete Peerage, xii (2), 932–35Google Scholar; CEMRA, p. 14; Aspilogia, ii, pp. 131, 186Google Scholar; Denholm-Young, p. 65, n. 1.

^{page 250 note 2} Moor, Revd. C., Knights of Edward I (London, 1929-1932) (Publications of the Harleian Society, Vols. 80–84), v, 2Google Scholar; Aspilogia, ii, p. 200Google Scholar.

^{page 250 note 3} Moor, i, 294; Aspilogia, ii, p. 126Google Scholar.

^{page 250 note 4} Aspilogia, ii, pp. 127, 197Google Scholar.

^{page 250 note 5} CEMRA, pp. 11, 14; Denholm-Young, p. 65, n. 1.

^{page 250 note 6} A is dated c. 1275 in the title and running title, CEMRA, pp. 14-16, but c. 1270-80 at the end of the discussion of the date, p. 14, mentioning A 67 (William Peyforer), 172 (Richard Fitzjohn), and 177 (Alan la Zouche). In addition to A 67 and 172, Denholm-Young, p. 65, n. 1, cites A 103 (Roger la Warre) and 159 (John Balun, who died i n 1275), concluding: ‘The Dering Roll could not have been completed before 1277-8’ (see also pp. 11-12). However, since A used FW, the terminus a quo is now moved forward to 1279. A terminus ad quern is established by A 251, Anselm Basset, who died on the Wednesday after Michaelmas, 1280 (Moor, i, 51). The Camden Roll is dated c. 1280 in CEMRA, p. 17, on the basis of D 149 (Patrick de Chaworth) and 160 (Bartholomew de Sudeley). Bearing in mind once again that D could not have been executed before 1279, the following items serve to confirm Wagner's terminus ad quem: D 140, John de Burgh, died on or before 25 February 1280; 187, Robert Mucegros, died 21 December 1280; 256, Henry, Count of Luxem-burg, died in 1281.

^{page 250 note 7} See my article entitled ‘The Relationship Between the Heralds’ Roll, Grimaldi's Roll, and the Dean Tract’, forthcoming in The Coat of Arms.

^{page 250 note 8} There are variants in the copies of any medieval roll of arms, but A presents a particularly difficult problem in this respect. For purposes of comparison here and unless otherwise noted, it was assumed that the ‘correct’ shield in the Dering Roll was the one closest to FW.

^{page 252 note 1} The following symbols and abbreviations appear in the remaining part of this article: B (Glover's Roll), C (Walford's Roll), and Matthew Paris (the Matthew Paris Shields), editions in Aspilogia, ii; BA, the Bigot Roll, edition in Adam-Even, Paul, ‘Un armorial francais du milieu du XIIIe siecle. Le role d'armes Bigot—1254’, Archives Héraldiques Suisses, lxiii (1949), 15–22, 68-75, 115-21Google Scholar; CC, Le Roman du Castelain de Couci el de la Dame de Fayel par Jakemes, edd. Matzke, John E. and Delbouille, Maurice (Paris, 1936)Google Scholar [Société des Anciens Textes Franfats]; CP, the Chifflet-Prinet Roll, edition in Max Prinet, ‘Armorial de France composé à la fin du XIII^e siècle ou au commencement du XIV^e’, Le Moyen Age, xxxi (1920), 1–49Google Scholar; CPA, the Chifflet-Prinet Roll, Additions, see Paul Adam-Even, ‘Role d'armes de l'ost de Flandre (juin 1297)’, Archivum Heral-dicum, lxxiii (1959), 2–7Google Scholar; Blazon, Early, Brault, Gerard J., Early Blazon. Heraldic Terminology in the Twelfth and Thirteenth Centuries with Special Reference to Arthurian Literature (Oxford, 1972)Google Scholar; G, Segar's Roll, edition in Greenstreet, James, ‘The “Segar” Roll of Arms as an Ordinary’, The Genealogist, iv (1880), 50-8, 90–7Google Scholar; H, the Falkirk Roll, edition in Greenstreet, James, ‘Falkirk Roll.’, Jewitt's Reliquary, xv (1875), 27–32, 68-74Google Scholar; J, Guillim's Roll, edition in Greenstreet, James, ‘“Guillim's” Roll of Arms as an Ordinary’, The Genealogist, i (1877), 323-7, 355–62Google Scholar; K, the Siege of Caerlaverock, The Roll of Arms of the Princes, Barons and Knights who Attended King Edward I to the Siege of Caerlaverock, in 1300, ed. Wright, Thomas (London, 1864)Google Scholar; M, the Nativity Roll, edition in Greenstreet, James, ‘Nativity Roll’, Jewitts Reliquary, xv (1875), 228–30Google Scholar; Watson, , Watson, G. W., ‘Notes on the Foreign Coats in Planché's Roll’, The Genealogist, N.S., vi (1889), 153-8, 222–7Google Scholar; vii (1890), 36-9, 152-5; viii (1891), 210-18; WB, the Wijnbergen Roll, edition in Adam-Even, Paul and Jéquier, Léon, ‘Un armorial francais du XIII^e siècle, l'armorial Wijn-bergen’, Archives Hiraldiques Suisses, lxv (1951), 49-62, 101–10Google Scholar; lxvi (1952), 28-36, 64-8, 103-111 lxviii (1954), 55-80.

^{page 254 note 1} FW 6 and 7 are both captioned ‘Le Roie St. Edwarde’, the first being intended for St. Edmund (Early Blazon, p. 44).

^{page 255 note 1} The author of Grimaldi's Roll (c. 1350) appears to have followed D here (item 39, Le Due de Baver, barruly azure and argent, a bend gules; edition in ‘Copy of a Roll of Arms (of the Reign of Edw. Ill) in the Possession of Stacey Grimaldi, Esq. F.S.A.’, Collectanea Topographica et Genealogica, ii (1835), 320–8)Google Scholar.

^{page 255 note 1} Cf. the Dean Tract, line 73: ‘Sire Gaweyn porte l'escu de sable frettee d'or’ (edition in Dean, Ruth J., ‘An Early Treatise on Heraldry in Anglo-Norman’, Romance Studies in Memory of Edward Billings Ham, ed. Holmes, Urban T. (Hayward, Cal., 1967)Google Scholar [California State College Publications, 2], pp. 21-9).

^{page 255 note 3} Lejeune, Rita and Stiennon, Jacques, La Légende de Roland dans I'art du Moyen Age (Brussels, 1967), i, 319Google Scholar.

^{page 255 note 4} Lejeune and Stiennon, i, 318; Early Blazon, pp. 53-4. In HE 44, the tincture of the field is erroneously given as argent.

^{page 255 note 5} Benoît de Sainte-Maure had earlier (c. 1174) ascribed a shield with lioncels on a field of azure to William the Conqueror; see Early Blazon, p. 21.