return to the old spot proved totally unsuccessful. This notion was the foundation of a host of medieval legends, chiefly connected with churches, which are often found built in unpleasant or inconvenient situations, to which they are supposed to have been condemned by the fairies. It would not be difficult to make a long list of such legends of churches still preserved in various parts of England. The church of Broughton, in Shropshire, stands in a hollow; but this, they tell us, was not the site originally chosen for it, as the foundations were laid on the top of the bank; but, during the night, two white oxen carried away the foundations and as much of the wall as had been raised, and placed them in their present position below. Next day, the builders returned to the spot they had chosen on the bank, but the two white oxen came again in the night and again carried the walls and foundations into the hollow, and this was repeated, night after night, until the builders were compelled to abandon their first design, and the church was built where it stood till recently. (I believe that this church has been recently removed.) Sometimes, in these legends, the fairies have yielded to an influence which has been alluded to in an earlier part of this paper, and have been transformed into demons. There is a legend of this class in a parish a short distance beyond the southern boundary of our county, in that of Hereford, which furnishes an apt illustration of these remarks. In the days when all the people in these parts were still heathens, a good Christian man came from far off and taught them a better creed, and after a while he set them to build a church at the village of Leinthall; but one morning, immediately after the work was begun, all the materials were found carefully laid down in a field about three miles off, far from any house. They were immediately carried back, with much ceremony, and the building was commenced again in the old spot, but again they were removed; and the devil continued so obstinate in his resolution to prevent the erection of the church in a spot where it could be conveniently attended by the villagers, that, after repeated attempts, the good preacher was obliged to give in, and the church was built in the lonely place where it now stands, and where few of the people of Leinthall could go to it.

However, I will leave the fairies and the witches of Shropshire to some other occasion, when I hope to become better acquainted with them. I have endeavoured to bring together a few of our legendary stories in order to explain to you their true nature and the real importance to which they have a claim, and I shall rejoice if I have contributed in any degree to the preservation from oblivion of such as still exist in our county. The study of them is of much greater interest than has generally been supposed. They are a slender but still perceptible link between us and that remote past, a direct view of which is concealed from our eyes by the collective obscurity of so many ages. In them we have some traces of the popular feelings of our primeval forefathers, while by thus tracing and

K

comparing them, we learn how totally unhistoric their character is, and how dangerous it would be to adopt them, as some inquirers have done, as records of historical events, or as evidence for any historical appropriation of the objects to which they are attached. I am satisfied, from my own observations, and from the information of others, that our county of Salop still offers a far richer harvest of such legends than anybody has supposed, and, in conclusion, I will only express the hope that such of my hearers as enjoy the opportunity of collecting them, will not allow any of them to perish, for all such legends are now rapidly disappearing.

ON THE NORMAN EARLS OF SHREWSBURY.

BY

J. R. PLANCHÉ, ESQ., ROUGE CROIX, HON. SEC.

MONGST the old Norman earls of England, those of the beautiful county in which we have now the pleasure of being assembled occupy the foremost rank, and Shrewsbury has still the honour of giving title to the premier earl of the kingdom. Roger de Montgomery, the first who possessed the earldom, was Viscount of the Oximin (Exmes or Hiemois), the intimate friend and counsellor of the Conqueror, furnished a noble contingent to his fleet, and was rewarded for his great services with the earldom of Arundel as well as of Shrewsbury, and large grants of lands in other counties.

His sayings and doings occupy many pages in the early annals of this country: Odelirius, the father of Ordericus Vitalis, one of our most instructive and valuable historians, accompanied him to England, and was a member of his household. Ordericus, born at Atcham, and educated in this city, has minutely chronicled his marriages, his children, his deeds of valour and of piety, his death and burial; and yet such is the mist that hangs over the genealogical history of our ancient nobility, that the father of this great and powerful earl has only been recently identified, and of his mother we absolutely know nothing. Brooke, in his Catalogue, declared him to be the son of Hugh Montgomery and of Sibell his wife, fifth daughter of Herfastus the Dane, brother of Gunnora, wife to Richard I, Duke of Normandy. Vincent triumphantly quotes Guillaume de Jumièges in contradiction of this assertion, and insists that he was the son of Hugh Montgomery by Jocellina his wife, daughter of Turolf de Pontaudemer, by Weeva, sister of Gunnora, Duchess of Normandy,1 and so he continued to be considered, notwithstanding that many passages in Ordericus show this to be a mistake, until the French editors of the latter historian, and the late Mr. Stapleton, in his illustrations of the Norman Rolls of the Exchequer, clearly proved that the first earl of Shrewsbury was not the son of a Hugh de Montgomery by either lady, but of another Roger de Montgomery living in the time of Richard III, and his brother Robert, Dukes of Normandy 1026-1035, and who, in an early deed, describes himself, "Ego Rogerius quem dicunt Montgomeri." His son Roger, our Earl of Shrewsbury, in the act of foundation for the abbey of Troarn in the Oximin, acknowledges and distinguishes his father in the following words: "Ego Rogerius 1 But the words of Gemetecensis are only "Ipse autem Rogerius natus est ex quædam neptium Gunnoris comitissa." Lib. viii, cap. 35.

ex Normannis Normannus, magni autem Rogerii filius." "The old chronicler Robert du Mont had heard," observes Mr. Stapleton, "of the reputed descent from a niece of the Duchess Gunnora, wife of Richard I, Duke of Normandy, but the genealogy given is certainly erroneous in making her, as wife to Hugh de Montgomery, the immediate progenitrix of Roger the Viscount of the Oximin or Hiemois." So far we are indebted to the researches of Mr. Stapleton and the observations of the French editors of Ordericus, and great is our obligation to them, as it saves us from a false start in the first instance, and throws open a new field for further exploration.

To any one unaccustomed to the examination of such subjects, it would appear strange that modern historians and genealogists could have overlooked the obvious inference to be drawn from the very circumstantial account given of the assassination of Osbern, seneschal of Normandy, by Guillaume de Jumièges himself, who, in the second chapter of his seventh book, informs us that Osbern, the son of Herfast,1 brother of the Duchess Gunnora, had his throat cut by William, son of Roger de Montgomery, one night while sleeping in the duke's chamber at Vaudreuille; that Roger for his perfidy was exiled to Paris; and that five of his sons, Hugh, Robert, Roger, William, and Gilbert, continued their wicked careers in Normandy. Surely no statement can be much clearer than this, that there was a Roger de Montgomery living during the minority of Duke William, who had five sons,2 the third being named after him, and who, it is evident from subsequent passages in the same and other histories, was the Roger de Montgomery who ultimately became Earl of Shrewsbury. For of these five sons we can trace the destinies. Hugh, Robert and William, were slain; the latter by Barno de Glotis, a servant of Osbern Fitz Herfast, in revenge for the murder of his master. Roger was Viscount of the Oximin; and Gilbert, the fifth son, was unintentionally poisoned by his sister-in-law, as I shall presently have occasion to relate to you. Yet notwithstanding this evidence, Messrs. Owen and Blakeway, in their History of Shrewsbury, 4to., London, 1825, confound the earl with his father, and speak of William and Gilbert as his sons instead of his brothers, vol. i, pp. 47-49; while in the pedigree they have given, vol. i, p. 64, they make them his uncles!

The deeds and charters quoted by Mr. Stapleton corroborate the most important parts of this genealogy. Of the five sons of Roger the first de Montgomery, Hugh was

2

1 We must be careful not to confound this Osbern or Osborne, the son of Herfast the Dane, brother of the Duchess Gunnora, with another "Osborne, the son of Herfast, a native of the district of Caux," who was made Abbot of St. Evrault, and of whom there is a long account in Ordericus's Ecclesiastical History. And probably a daughter, who was the mother of Ameria, niece to Roger Earl of Shrewsbury, and bestowed in marriage by him on Warin the Bald, his Viscount or Sherrif of Shrewsbury. Ameria may, however, have been the child of one of the earl's brothers, as it is scarcely probable that all four died unmarried or without issue.

apparently the eldest, as at the foot of one of his charters in the time of Duke Robert is "signum Hugonis filii ejus," and it is therefore highly probable that the father of the first Roger might have been named Hugh, and the husband of one of the nieces of Gunnora, and the confusion have arisen from that circumstance. The story told by Guillaume de Jumièges, though clear enough as regards the family of Montgomery, is obscure in other respects. William de Montgomery is named as the murderer of Osbern the son of Herfast, who, if there be any foundation for the statement of Brooke, must have been his near kinsman; and Roger de Montgomery, the father of the criminal, is banished apparently for the crime, which would seem to imply that it was committed at his instigation. However this may be, it appears to have been the result of a personal, if not a family quarrel; and it is remarkable that no mention is made of the murder by Ordericus, whose father, as a member of Earl Roger's household, could scarcely have been ignorant of the fact. He nevertheless records that Osborne, the steward of Normandy, and William and Hugh, the two sons of Roger de Montgomery, and many other powerful knights, made war on each other in turn, causing great distress and confusion in the country, which was now (i. e., during the minority of Duke William) deprived of its natural protectors,1 having in a previous chapter simply mentioned Osborne as one of the many nobles who fell in these mutual quarrels.2

Let us now test the probability of the descent from a niece of Gunnora to which Mr. Stapleton alludes. The genealogy of the dukes of Normandy from Rollo is in all the collateral portions exceedingly confused, and the chronology of the duchy itself beset with difficulties. Next to Charlemagne, the Duchess Gunnora appears to have been the favourite starting point for our Norman genealogists. If there is any insuperable obstacle in the way of hooking their line on to the Emperor of the West, they eagerly hitch it up, no matter how, to some loose end of the family of that fortunate fair one, for whose romantic history we are indebted to the pages of Guillaume de Jumièges. As it is short as well as romantic, and so very old that it may be quite new to many present, I will venture to tell it in the fewest words possible. One of the foresters of Richard I, Duke of Normandy, was blest with a most beautiful wife, of Danish blood as it would appear, named Sanfrie, the report of whose charms inspired the duke with a vehement desire to ascertain the truth of it by personal observation. He, therefore, had a hunting party in that direction, and stopped, as a matter of course, at the house of the forester for rest and refreshment. The beautiful Sanfrie received her sovereign as was her duty, and the duke was so captivated that he commanded her husband to resign her to him. As resistance could avail nothing, the lady, who had as much wit as beauty, contrived to substitute her sister for herself, and the duke, luckily for all parties, was not 'Ord. Vit., book v, ch. ix.

2 Ibid., book 1, chap. xxxiv.