William de Gommeth, the owner of Ferté-Alais, were fleurs de lis, taking all the circumstances together, including this, that in this William, and his presumed brother Geoffrey, we have just the two names of two presumed brothers, living fifty years afterwards, viz., William and Geoffry Alis, it would scarcely be too hasty to conclude at once, that William de Gommeth was an Alis, but took the name he bore from his estate. But it would

1 Notice has been taken in Part I. of the heraldic bearing, a leopard's head jessant a fleur de lis: it is remarkable that this is rarely met with in French blazonry, as is also the leopard's head or face alone. And in England, during the early Norman period, the fleur de lis seems to have been set aside by those baronial families who were entitled to use it, and some other bearing to which they had a right used instead; this does not seem, however, to have been the case in Scotland. A political reason may be assigned for this studious avoidance in France and England of these ensigns in England the use of the fleurs de lis, the royal emblems of France, would savour too much of affection and vassallage to the king of that country; in France the use of leopards-the arms of the Conqueror and his successors-or leopards' faces, would be a compliment to the Dukes of Normandy, not very pleasing to the sovereigns of France. This may explain the restrictions which William the Conqueror is said to have placed on the use of coat armour. Those families who were entitled to use the leopard's head as well as the fleur de lis, seem to have disguised the latter by the combination of both, unwilling to relinquish so ancient and distinguished an emblem as the fleur de lis, probably considered, at that time, an indication of descent from the royal race of France.

As the heraldic conjunction in question was used by the Ellises of England, and kindred families, it is desirable to inquire whence they got the leopard's face, because, in so doing, we may fully expect to elucidate their ancestry. Accordingly, we shall find that such inquiry leads to the acquaintance of a group of families, who seem to have had a common origin with, and to be synonymous in meaning with, that of Gommeth in

the text.

It will be useful, first, to show that the wolf's head, leopard's head or face, and lion's head or face, were used indiscriminately in heraldry, or were certainly often confounded, as it is well known that the lion passant was originally a leopard passant. The heraldic device of Hugh Lupus, Earl of Chester, is well known-a wolf's head, from which undoubtedly he got his name, and not from his disposition, as is supposed; for William Gouel de Perceval, son of Ascelin Gouel, a powerful baron, and a contemporary and probably relative of Hugh Lupus, was called Lupellus, unquestionably to distinguish him; and there is a seal of his extant, impressed with a wolf, his descendants taking the surname of Lovell, one of the coats of which family is a chevron between three wolves' heads erased, or (as sometimes corrupted) lions' heads erased. In the family of Weston, or Wiston, of Wiston, in Sussex, we find one branch using lions' heads (the Sussex); another (in Devonshire), leopards' faces; and a "Monsieur Thomas de Weston," at the tournament at Dunstable, 7 Edward III., azure three leopards' heads jessant de lis or: thus showing a synonymous mean

be well first to examine the pedigree of the Rocheforts, for we shall find that that family shows many indications of descent

ing in these bearings; and the family of Gower have used a chevron between three leopards, and also three lions' heads or faces; whilst the family of Gouiz have used lions' heads as well as wolves' heads erased.

Gouiz was one of the surnames of the father and ancestors of Hugh Lupus, as D'Avranches was his and his father's. The best pedigree of this family is in D'Anisy's Domesday. Rollo, the first Duke of Normandy, had a brother, whose grandson, Amfridus le Danois, Viscount of Exmes, till 798, was the first surnamed de Goz or de Gouiz. Amfridus, or Humphrey de Tilly, or Tilleul, grandson of the preceding Amfridus, was governor of Hastings in 1068, but resigned, or was deprived of that post, and returned to Normandy. In Desbois' Dict. de la Noblesse, he is said to have been progenitor of the eminent family of Tilly, in France and England, whose arms were a fleur de lis, through Robert de Ruddlan, his son; but he left no legitimate issue, and his brothers were monks. The true descent, however, of the Tillys, seems to have been given in Mr. Wiffen's Memoirs of the House of Russell (i. 119), who derives them from Roger Tilly, of Veroles, a brother of Humphrey. (It may be appropriately mentioned here, that Mr. Wiffen's pedigree of the ancestry of Hugh Lupus, omits an entire generation, viz., the first Humphrey.)

The families of Hastings and Despenser, there is little doubt, were descended from a brother or nephew of Humphrey, the governor of Hastings, and who probably succeeded him in his office. Their family names and arms strongly confirm this probability. A Robert and Ralph de Hastings are mentioned in the Domesday Survey; and Humphrey or Amfridus de Hastings was the Domesday owner of Leamington-Hastings, co. Warwick, and his descendants bore a chevron between three leopards' faces. Humphrey Vis de Leu, a tenant in capite in Berks, mentioned in that Survey, was probably the same person; his descendants bore three leopards' faces or wolves' heads. The family of Dene, in Sussex and Kent, in which occurs the Christian name of Amfridus, who bore leopards' faces, were, there are well-founded reasons for believing, a branch of the family of Hastings. Lastly, the families of Despenser and Hastings (the chief and most distinguished line) bore, as is indirectly proved, originally fleurs de lis and leopards' faces for their arms. Robert de Hastings was dispensator or steward to William the Conqueror, and lord of Fillongley, co. Warwickshire. The arms of Fyllingley were, sable a chevron engrailed between three leopards' heads or. Temp. Henry III., Geoffrey De Dutton, of Dutton, in Cheshire, sealed with a maunch, the hand holding a fleur de lis. The pedigree of this family is obscure, but it is certain this Geoffrey was a Despenser. The arms of the family of Hastings, as generally borne, were a maunch, without the fleur de lis, but, for simplicity or other reasons, to abandon a part of a coat of arms, was not uncommon: however, we find them both borne by the family of Mohun, and by that of Conyers; of the latter there is a seal with these arms surrounded by cross crosslets, as early as the time of King Stephen; and if the seal of a Hastings or Despenser could be found at that early period, no doubt, as in the case of Geoffry de Dutton, the fleur de lis would be found along with the mauuch. Again, one of the coats of Spencer is three fleurs de

from a race of evident common origin with the Alises and other synonymous families, to be treated of hereafter.

lis on a bend, and the same occurs in another, viz., quarterly in the second and third, a fret over all, on a bend three fleurs de lis.

The first Amfridus de Gouiz doubtless derived that name from his estate so called, which estate he as doubtless obtained from his mother Gerlotta, daughter of Theobald Count of Blois. William, the first known Count of Blois, who died 834, was descended from Childebrand, brother of Charles Martel, and was succeeded by his collateral descendants, the ancestors of Hugh Capet. The second race began with Thibaud, brother of Rollo Duke of Normandy, who, having married Richilde, sister of Robert and Eudes, kings of France, was made Count of Blois, Tours, and Chartres. The fleurs de lis of the early Counts of Blois were probably used by the Norman Counts, as well as their own ensigns, the leopards passant, and, it would seem, the leopard's face or wolf's head; for these last were, there is little doubt, in use by the ancestors of Hugh Lupus, the brothers of Rollo. The origin of the baronial family of Plaiz is not known. One of their coats was a lion or leopard passant; another a fleur de lis. If we consider Plaiz a synonyme for Blois, their arms are accounted for, and they justify the synonyme. The arms of the baronetical family of Blois are a bend vaire between two fleurs de lis. The Counts of Blois were also Counts of Champagne. The family of Campion, or de Campania, was probably a branch of these latter; a seal of Sir Robert de Campania of Kent, in the thirteenth century, is a coat vaire (Hasted's Kent). Vaire would seem, therefore, to have been another of the bearings of this race. Three pales vair were borne by Stephen Count of Blois and King of England, and also by the house of Chastillon, before they became Counts of Blois; and they were probably sprung from an early cadet. Now, the three ensigns in question-the leopard, &c., the fleur de lis, and vaire-were all borne by Gouiz, or synonymous families. Robert de Gouviz sealed, in the twelfth century, with vaire over all a bendlet (D'Anisy, Seals, plate 8, No. 16). He seems to have taken his name from the parish of Gouviz, or Gouvix, in Normandy. In 1181, Raoul de Gouvis was witness to a charter of donation to the Abbey of Barberry. In the eleventh century, William and Richard Gouiz were benefactors to the Abbey of St. Stephen, at Caen (Transactions of the Society of Antiquaries of Normandy, 1836, p. 229). In 1166, Richard de Guviz held five knights' fees in Oxfordshire (Liber Niger). There is little doubt that Gower is synonymous with Goviz, and also Jay, Joy, and Gay, who bore three leopards' heads jessant de lis. The arms of the family of Gael or Guader, Earls of Norfolk, were party per pale a bend vaire. This family took their name from, or gave it to, their castle of Gael, in Brittany. The family of Gouel, or Ivery, came from that province, of whom Robert and William Gouel occur in 1184; they subsequently took the name of Lovel, as we have seen, from Lupellus, the soubriquet of one of them, who used a wolf passant, or wolf's head. These two families were, doubtless, of common origin with the preceding. According to the Dictionnaire Généalogique (Paris, 1757), GovE of Anjou bore or, a lion (? léopard) gules, surmounted by a fleur de lis azure: GoVIE, seigneur de Montgiron, in Normandy, party per or and azure, three fleurs de lis gules; and DE Gouz, écuyer, seigneur de Chateau-Thouars,

As Guy the Crusader (who is called also le Rouge) and his brother William were both styled de Rochefort, it is clear that

in Normandy, the same: COUE, seigneur de Brossai, argent, a fess sable between three fleurs de lis gules. Hommet and Hommell are known to be synonymous: we can have no difficulty, therefore, in believing Gouel, Gael, Couet, and Gouet, to be identical, at least originally. In 1166, William Gouet held half a knight's fee in Norfolk. As we shall see presently, a William Goet married a heiress of Alluye or Ales, who was succeeded by four successive Williams. Another connection of these two families appears in the name Gué-Alais (? near Avranches), of which Henry de St. Germains was lord in the seventeenth century.—(Hozier's Arms de France, Reg. 4, Ped. of Quesnoy.)

The question now arises, whether Gommet or Hommett were originally the same as Gouet, Gael, &c. Guader and the latter were used synonymously, and they are more dissimilar than Gommet and Gouet. The genealogy and heraldry of the question strongly countenance the identity. But, taking this for granted, what was the original form and meaning of these names? Can it be doubted that the name and family of Guelph, or Wolf, is the fons et origo of all the preceding families, and corresponding armorial ensigns?

The descendants of Guelph, a Bavarian count, who lived in 820, became Dukes of Bavaria. The arms of this kingdom, lozengy, are known to be one of the most ancient in Europe. A Grimoaldus was Duke of Bavaria in 665, and his grandson, of the same name, in 728. Grimaldus, Prince of Monaco in 920, is considered ancestor of the family of Grimaldi, who bear the arms of the principality of Monaco, viz., lozengy. This was the armorial ensign of the Counts of Angouleme, descended of the Dukes of Aquitaine, who bore also this coat, and another, gules a leopard passant or, as borne by William Duke of Aquitaine, a Crusader, in 1101: Robert, Duke of Normandy, whom he accompanied to the Holy Land, bearing gules two leopards passant or. Now, the first Duke of Aquitaine lived 668, and was named Loup, who was succeeded by Eudes, a descendant of Lothaire II., King of France. From Eudes descended two Dukes of Gascony, named Loup. Can there be a doubt then that the leopard passant of the Dukes of Aquitaine was derived from Duke Loup? The arms of the family of St. Lo, or Loup, are, or a leopard gules on a chief indented, azure, three fleur de lis of the first. Guy was a prevalent Christian name in this family; the arms of Guy, sire de Laval, of the same family, in 1101, were gules a leopard passant or; and the arms of Pierre de Lyoberd, in 1218, or a leopard gules. The arms of Chanteloup were, a wolf or a leopard passant, whilst the coat of Robert de Chanteloup, a Crusader in 1096, was lozengy or and sable, thus doubly indicating a connection with the preceding families. William, brother of the first Amfridus de Gouiz, and grandson of the brother of Rollo, was ancestor of the Crispins, Barons of Bec-Crispin. This family bore lozengy for their arms. Rollo was son of Rognald Count of Merc, whose son, Theodoric, inherited that title. Thibaud, or Theobald, another son, we have seen, was Count of Blois. These two names were prevalent in the Carlovingian race of kings. When Rognald, who had a contemporary of the same name Duke of France, was expelled the court of Harold Harfager, the King of Norway, his mother

that title and estate must have been inherited from an ancestor, and could not have been acquired jure uxoris by Guy, who is styled generally Comte, and sometimes only Sire de Rochefort. There can be little doubt that that ancestor was William de Gommeth. There were various families named Rochefort, many of them, doubtless, of common origin. No particulars can be found of any Comtes de Rochefort previous to Guy; but in the year 1619 a family of this name was so ennobled, which is styled "an illustrious and ancient house, that took its name from a town and chatellenie in the diocese of Dole, in Franche-Comté." Another family were Comtes de Rochefort-en-Vivarais, one of the most distinguished houses du Forez, where it flourished from 1200, being established in the Vivarais 1648. This family bore three fleurs de lis, with a chief charged with a demi-lion rampant, and may have been the representative of the early Comtes de Rochefort. Du Chesne calls Thibaud, the grandfather of Guy the Crusader, ancestor of the Comtes de Rochefort-en-Iveline. Guy himself, his eldest son Guy, who died without issue, were styled Comtes, and his sister and coheiress Comtesse de Rochefort, which title she carried to her husband, the Count of Evreux and Montfort, whose son's wife, the daughter of Robert, Earl of Leicester, is witness to a charter of his, and described as "Amicia filia mea Comitissa de Rochefort" (Dug. Mon. vi. 1079). Guy, son of Simon Count of Evreux, and who died 1228, was lord of FertéAlais. It is probable, upon the whole, that Guy the Crusader was at first only Sire, and afterwards became Comte de Rochefort.

The earliest notice we have of the name of Rochefort is that of Antoine de Rochefort, who, with his wife Margaret, heiress of Ally, founded, in 1001, the priory of Bonnet or Rochefort, in the diocese of St. Flour.3 Geoffry de Rochefort is witness, in 1068, to a charter; and in 1047, Fulcard de Rochefort, along with Elias de Chalisio and Elias de Charnaco, were witnesses to a charter.5 4 Phillipe I., Bertrand de Rochefort was witness and

interceded on his behalf, and told the king he was banishing a man of noble race, warning him it was dangerous to attack the wolf, &c. (Thierry's Norman Conquest.) This latter phrase of course had only a figurative meaning-might it not allude to his family emblem? and, when describing him of "noble race," might not that expression mean his descent from the French kings-in fact, do not all the preceding heraldic and genealogical circumstances point to such an origin, instead of the fabulous descent from fictitious Scandinavian jarls?

1 Courcelles' Dictionnaire de la Noblesse, iv. 121.

2 Hist. de la Maison de Montmorenci.

3 No notice of this priory is to be found in the Gallia Christiana. 4 Gall. Christ. ii. App. 272.

5 Ibid., p. 481.