^{page 75 note 1} The basic study of twelfth-century literary heraldry is de Lesdain, Louis Bouly, ‘Études héraldiques sur le xii^e siècle’, Annuaire du Conseil Héraldique de France, xx (1907), 185–244Google Scholar. Consult also Adam-Even, Paul, ‘Les usages héraldiques au milieu du xii^e siècle d'après le Roman de Troie de Benoît de Sainte Maure et la littérature contemporaine’, Archivum Heraldicum, lxxvii (1963), 18–29Google Scholar. Several illustrative passages are given by Sir Anthony Wagner in Appendix A of Heralds and Heraldry in the Middle Ages (London, 1939)Google Scholar. For the thirteenth century, see de Marsy, Le comte, ‘Le langage héraldique au xiii^e siècle dans les poèmes d'Adenet le Roi’, Mémoires de la Société des Antiquaires de France, 5^e série, xlii (1881), 169–212Google Scholar, and especially Prinet's, Max penetrating studies: ‘Les armoiries dans le roman du Châtelain de Coucy’, Romania, xlvi (1920), 161–79Google Scholar, and ‘Le langage héraldique dans le Tournoiement AntéChrist’, Bibliothèque de l'École des Chartes, lxxxiii (1922), 43–53Google Scholar.

^{page 75 note 2} Adam-Even, Paul, ‘Études d'héraldique médiévale: un armorial français du milieu du xiii^e siècle, le rôle d'armes Bigot—1254’, Archives Héraldiques Suisses, lxiii (1949), 15–22Google Scholar, 68–75, and 115–21. London, H. Stanford, ‘Glover's Roll’, in Rolls of Arms. Henry III (London, 1967)Google Scholar, [Aspilogia, gen. ed. Sir Anthony Wagner, ii], pp. 89–96 and 103–66.

^{page 75 note 3} Gerard J. Brault is the first scholar to recognize the importance of going beyond mere lexicography in studying blazon. In ‘The Emergence of the Heraldic Phrase in the Thirteenth Century’, The Coat of Arms, viii (1964–5), 186–92Google Scholar, Brault stresses a syntactical and structural approach to blazon, which he has maintained in more recent studies.

^{page 75 note 4} Early Blazon. Heraldic Terminology in the Twelfth and Thirteenth Centuries with Special Reference to Arthurian Literature (Oxford: Clarendon Press, 1972)Google Scholar.

^{page 75 note 5} A brief view is found in Hope, W. H. St. John, A Grammar of English Heraldry, 2nd edn., rev. by Anthony R. Wagner (Cambridge, 1953)Google Scholar. See also Prinet, Max, ‘Les usages héraldiques au xiv^e siècle d'après les chroniques de Froissart’, Annuaire-Bulletin de la Société de l'Histoire de France, liv (1916), 3–16Google Scholar. For a more thorough study, consult Barstow, Allen M., ‘A Lexicographical Study of Heraldic Terms in Anglo-Norman Rolls of Arms: 1300–1350’, Dissertation Abstracts International, xxxi (1970), 2866–aGoogle Scholar (University of Pennsylvania).

^{page 76 note 1} Ten Anglo-Norman rolls are listed by SirWagner, Anthony, A Catalogue of English Mediaeval Rolls of Arms [CEMRA] (London, 1950)Google Scholar, [Aspilogia, i], pp. 24–37. To these may be added the French Chifflet-Prinet Roll and the Tournoiement as Dames de Paris.

^{page 76 note 2} Bodleian Library, Rolls 4 (formerly 14A) [hereafter abbreviated as AS]. The manuscript consists of six leaves of vellum sewn end to end and rolled, containing recto a copy of the Ashmolean Roll of Arms (with twenty-four sixteenth-century entries appended), verso the Ashmolean Tract (a fifteenth-century treatise on heraldry in English). The manuscript is described by W. H. Black in his Catalogue of the Ashmolean Manuscripts (1845), p. 7, and by Wagner, CEMRA, pp. 57–8. The beginning of the manuscript is partially torn and mutilated, but it is possible to decipher some of the first five entries, and most of nos. 6–15. There is no heading or introduction. There are two sixteenth-century tricks of this manuscript (College of Arms MS. Vincent 164, fols. 119^v–134^v and Queen's College, Oxford, MS. 158, pp. 403–33). Although these tricked copies contain no text, they are useful in interpreting the meaning of some mutilated or badly copied entries in AS. The copyist has made many errors, but he writes in a clear hand, and does not seem to have consciously altered the original text. Noel Denholm-Young points out that the copyist has altered the title in some eleven entries to indicate the status of the family in the early fifteenth century (The Country Gentry in the Fourteenth Century with Special Reference to the Heraldic Rolls of Arms (London: Oxford University Press, 1969), pp. 97–9)Google Scholar.

^{page 76 note 3} The lords are termed le sire de (nos. 33–59, 61–79, 81–90, 94–105, 109–10, 118–19, 200–1, 271–2). Nos. 91–3 are called ba[ron], while there is no title for nos. 112–14, 121–4, 126–7 140, and 475. The rest (364 in all) are simple knights, with the title mons' de. Denholm-Young (loc. cit.) names two knights among the lords and nine lords among the knights; note that his numbering system does not include the kings and earls, and that his totals are twelve earls, forty-eight lords, and 406 knights. He thinks that ‘for the most part it seems to have been made at or near York and to contain persons who were there or passed through on their way to the north’ when the seat of government had been moved to York for the duration of Edward's Scottish campaigns.

^{page 77 note 1} Unlike several continental rolls, there is little confusion in AS between vair (veir/ver) and vert, always spelled with a t.

^{page 77 note 2} AS 18, 187 and 188.

^{page 77 note 3} AS 5, 299 and 372.

^{page 77 note 4} Reference letters for rolls cited here are those listed by Wagner in CEMRA, p. 176. B = Glover's Roll (London ed.); C = Walford's Roll (London ed.); CG = Cotgrave's Ordinary (Nicolas ed.); D = Camden Roll (Greenstreet ed.); FW = Heralds Roll (Cambridge, Fitzwilliam Museum MS. 297); G = Segar Roll (Greenstreet ed.); H = Falkirk Roll (Greenstreet ed.); K = Caerlaverock Poem (Nicolas ed.); LMS = Lord Marshal's Roll Old, Part II (London, Society of Antiquaries MS. 664, vol. i, fols. 20–20^v); M = Nativity Roll (Greenstreet ed.); MP = Matthew Paris shields (Tremlett, ed. in Aspilogia ii, pp. 1–86Google Scholar); N = Parliamentary Roll (Nicolas ed.); O = Boroughbridge Roll (Greenstreet ed.); P = Grimaldi's Roll (Grimaldi ed.); ST = Stirling Roll; WNS = Sir William le Neve's Second Roll (Society of Antiquaries MS. 664, vol. i, fols. 25–26^v). Approximate dates are as follows: MP—1244; B—1253; C—1275; FW—1275; D—1280; G—1282; H—1298; K—1300; M—1300; ST—1304; LMS—1310; N—1312; O—1322; WNS—1325; CG—1340; P—1350.

^{page 77 note 5} See Humphery-Smith, C. R., ‘Purpure’, The Coat of Arms, iv (1956–7), 19–20Google Scholar, and Brault, Early Blazon, p. 263.

^{page 77 note 6} Lacy: FW 593, G 33, H 1, K 1, WNS 15, N 4, and P 49; Malemeis: M 46, N 296, AS 372, CG 80 and 352.

^{page 77 note 7} MP. II 72, B. 16 (= II. B 4), M 64, ST 69, N 750, 999, 1024, 1078; O 180, AS 5, 299; CG 80 and 352.

^{page 77 note 8} These terms tend to appear in literary heraldry rather than in rolls of arms (see Early Blazon, passim).

^{page 78 note 1} From 1300 (Galloway Roll) to 1334 (Second Dunstable Roll) the number of ‘pieces’ is announced little more than 10 per cent of the time.

^{page 78 note 2} Brault cites ten examples, of which half occur in a literary context. The earliest of the five examples from rolls of arms is 1300 (Chifflet—Prinet Roll 33: fessié de … et de … a 6 pieces). The other four are from later versions of thirteenth-century rolls (C.II.B 9, H.II 16, 36, and 60).

^{page 78 note 3} Pommelé occurs earlier in Chifflet—Prinet 131 with the meaning ‘botonny’ in describing the cross of Tonlouse (une fausse crois … pommelé).

^{page 78 note 4} Semé de … (or a … semé) is used in both the continental rolls, Bigot and Chifflet—Prinet, but there are only a few scattered examples in Anglo-Norman rolls (see Early Blazon, s.v. semé).

^{page 78 note 5} Orle de … occurs for the first time with this meaning c. 1255 in B.II 23 and 186, but it does not reappear until the Boroughbridge Roll (1322).

^{page 79 note 1} The AS herald is traditional in avoiding the use of endenté with any ordinary consisting of conjoined lozenges. The Parliamentary Roll is an unreliable guide—it is the only medieval roll which consistently violates this tradition, with some fourteen examples of fesse endentee.

^{page 79 note 2} Estachié de … et de … occurs in thirteenth-century continental heraldry, cf. Early Blazon.

^{page 79 note 3} Walford's Roll, the Nativity Roll, and the Raine—Dunn Roll are the only other medieval rolls in which this term occurs.

^{page 79 note 4} Crois recercelee is used earlier in C 160 and 177.

^{page 80 note 1} There are a few earlier examples of an eagle armed: B 204, beke et les pees; LMS 1, bek et jambes; WNS 81, 82, 98, enarmé.

^{page 80 note 2} Denholm-Young draws attention to the ‘peculiarity’ of the spelling lipard for lupard in AS as proof of a poor copyist (Country Gentry, p. 98, n.). This is inaccurate, since lipard occurs regularly in the Boroughbridge and Carlisle Rolls, whereas lupard appears only in the Parliamentary Roll, which should not be used as a reliable guide to fourteenth-century blazon. The most frequent spelling is as in modern English.

^{page 80 note 3} The only exception is in C.I.A 1 and 68 (egle espany) and 3 (egle displayé).

^{page 80 note 4} Eglencel is not mentioned in the standard Old French dictionaries (Godefroy, Tobler-Lom-matzsch, REW, FEW). The diminutives of aigle in Modern French blazon are aiglon, aiglet, and aiglette (d'Haucourt-Durivault, , Le Blason (Paris, 1949), p. 82Google Scholar).

^{page 80 note 5} OF cornille, from CL CORNICULA (FEW, ii, 1190). Brault cites an example of corneille in Cleomadés, see Early Blazon, p. 150. Corbins appear earlier in D 62, H 27, N 938 and 939, and WNS 78.

^{page 81 note 1} Fiché first appears in all three of the rolls blazoned c. 1334 (Carlisle, Second Dunstable, and AS). Botoné and paté had been applied to crosslets before this time, the former in WNS and the latter in N.

^{page 81 note 2} Compare CG 359: pieces.

^{page 81 note 3} Ed. by Marcos Jiménez de la Espada (Madrid, 1877). The entry for Rome is N 25.

^{page 81 note 4} Mascle appears twice in B.III.B 8 and 21. In no. 21 it describes the lozenges of Kent (Burgh), but in no. 8 mascles refers to the voided lozenges of Winchester. I believe that the latter blazon may be spurious, since it does not appear in any of the other versions of Glover's Roll. In medieval rolls mascle is used to blazon the lozenges of Croupes, Rivers, Guise, Gorges, Fleming, Charles, Frevill, and Withacre, none of which is voided.

^{page 81 note 5} There are few exceptions, and they are of doubtful authenticity. The lozenges of Kent are called losenges in B.III 21 (but not in any other version of Glover's Roll). The seven mascles of William de Ferrers are blazoned losengez in H. I. B 61 (but losenges persees in the older Wrest Park version), while in B.III.B 8 the Winchester mascles are incorrectly called mascles (see n. 34).

^{page 82 note 1} One can often ascertain the position of a charge by application of the rule against placing metal on metal or tincture on tincture.

^{page 83 note 1} See Brault, , ‘Ancien français de l'un en l'autre’, Romania, lxxxviii (1967), 84–91CrossRefGoogle Scholar.

^{page 83 note 2} There are no examples of arms marshalled on a fess in this period.

^{page 83 note 3} Foreigners with arms marshalled by quartering in fourteenth-century Anglo-Norman rolls include the Kings of Spain, Sicily, Cyprus, Bohemia, the Duke of Burgundy, Wolfrod/Gistelles, and Montagu.