Edward the Elder.* Even more startling was his avowed belief that the very late tower at Conisborough was built, and in use, in Pagan times,' and his discovery in it of a niche for an 'idol'! Guildford, he thought, was probably constructed in the time of the Heptarchy; and Restormel was even of more early date than Conisborough itself.'† But so recently as 1852, a paper was read by Mr. Duesbury before the British Archæological Association, contending that Rochester keep (which was built after 1126) was 'a striking case' of early construction, its Anglo-Saxon origin being a matter of mathematical certainty. Roman origin has been claimed in England for Exeter and the Tower of London; in France, for Provins and other keeps; and Colchester, which the learned Stukeley believed to be a Roman granary, has been strenuously represented, in our own days, as originally a heathen temple. That It is necessary to realize the chaotic beliefs that had prevailed on the subject of our castles before we can justly estimate the work which Mr. Clark has done, and appreciate the change he has wrought in our knowledge of military architecture. his conclusions are not final, we may at once admit : indeed we shall have to correct them, on some points, ourselves. And in using his volumes it is well to remember that their historical side is the weaker, and distinctly inferior to the architectural. Nor is it only on this ground that a revised edition of this great work ought some day to be issued: the collection of scattered papers, 'written at very long intervals of time,' leads, as the author confesses in his preface, to occasional iteration,' and, we may venture to add, confusion; while the want of an index involves the most tedious investigation, if we would ascertain all that the author has said on any given point. Nor can we be sure, even when we have done so, that we have Mr. Clark's final verdict. In these volumes, for instance, the gloomy keep of Newcastle-past which the traveller is whirled by the East Coast express-is always assigned to 1080, a date which places it among the earliest examples of the 'tower': yet the very year this work appeared, Mr. Clark was assuring the Newcastle Congress that it was a distinctly late type, of the days of Henry II. Such a change not only destroys all corollaries of his previous conclusion, but makes us fear that he is hasty at times in his estimate of the structural evidence. So also do certain quite contradictory statements in his great paper on the Tower of London. The importance of the 'fixed point' in all such researches as this cannot be over-estimated: if erroneous, there is no limit to the mistakes and contradictions to which it may lead. To take a somewhat different instance, Mr. Clark asserts that 'the burh at Wigmore' was built by Edward the Elder in 921, and unsuccessfully assailed by the Danes the same year (i. 5, 22, 42; ii. 526-7). Now a glance at Florence is sufficient to show that Edward built this 'burh' not in the extreme west of England, but in the east, while warring against the Danes. This is no mere correction of fact: it may be of the utmost importance in its bearing on the earliest Norman castles. For their problem, we shall see, must be studied in Herefordshire, the cockpit, in the eleventh century, of England and Wales. Again, in his study on Rochester Castle, Mr. Clark holds that Henry II.' (but he confuses him with Henry I.) 'alludes to the castle in his confirmation charter,' and deduces from this that 'the arx, or citadel, was then in progress' (which would place its construction after 1174). But when we find that this charter only refers to the old formal trinoda necessitas, the deduction falls to the ground. Lastly, we may mention that Mr. Clark, like others, estimated too highly the accuracy, while rightly admiring the enthusiasm, of Professor Freeman. He declares him, in his preface, to be 'a master of mediæval architecture,' and dedicated his work to him, as an unrivalled authority on English military buildings. Mr. Freeman, in return, described Mr. Clark as the great master of military architecture'; and their close connexion will compel us, occasionally, to consider the views of both writers. 6 Perhaps the most fascinating feature in the study of English strongholds is their gradual change and development in the hands of successive races. It is almost startling to find that at the very end we stand where we stood at first, and are now reverting, in our entrenched camps, to the earliest of all defences, the earthwork. As Mr. Clark begins his studies with Post-Roman and English Earthworks,' we have not his guidance for those British' camps which still crown in massive grandeur many a lonely height. We have lately had from Mr. Hardy's pen a study of one of the vastest and most impressive of these works, that of Badbury. Far more widely known, however, are the marvellous defences of Old Sarum; a very noteworthy place, in some respects the most noteworthy in Britain' (ii. 448). Associated by some with the great Reform Bill, or 'the name of the elder Pitt,' Old Sarum for those who have seen it will remain the greatest of hill-citadels: The Norman fortress, the city, the cathedral church have all vanished; ... and here, as at Stonehenge, the memory of the Briton is once more predominant.' Mr. Clark holds that Badbury Badbury is the work 'most worthy to be compared with it; but the central hill at Old Sarum, 500 feet across at the summit, has no parallel there. Mr. Clark leans to the view that this was an English addition, but he has not, we think, allowed for the vast scale of the defences. The earthen rampart of this inner hill makes it of itself a British camp which may fairly be compared with that at Exeter, which was utilised by the Romans, and was eventually turned into a Norman castle. Place Exeter within Badbury, and you obtain the double defences of Old Sarum. Dover Castle,' the key of the kingdom,' is probably the most perfect example of continuous development that we have. A British hill-camp, following the figure of the hill,' was occupied-but not adapted-by the Romans; it was afterwards, in days before the Conquest, so strengthened that Guy could write : 'Est ibi mons altus, strictum mare, litus opacum. · Further strengthened by the Conqueror, it received from Henry II. its stately and familiar keep. Fresh towers and walls were continually added to its defences, till it eventually became a concentric fortress, more or less of the Edwardian type. In recent times the ancient earthworks have been scarped, extended, retrenched, and tunnelled, barracks and magazines have been built, the keep has been converted into storerooms and water tanks, and in its basement are two powdermagazines.' Thus its use as a place of arms down to the present day has somewhat obscured and confused its successive stages of development. Roman stations, like British camps, have been turned to accounts, in their own way, by English and Normans in succession. There is not much danger of our confusing them; for while the latter stood on hills, and followed in their lines, as a rule, those of the ground, the former were normally rectangular, besides being slighter in construction. The Romans did not restrict themselves to the hills, nor did their gateways at all resemble the tortuous and curiously guarded entrances to a British camp. Two admirable instances of a castle on a Roman site are Pevensey and Porchester. Mr. Clark holds that the former is, in some respects, the most interesting place in the south of England'; while of our remaining Roman fortresses, 'none are to be compared for completeness of preservation, and but few for extent of area, with Porchester.' At Pevensey the massive walls and towers of the Roman castrum were destined to contain an English stronghold, afterwards adapted as a Norman castle, and finally developed, early in the fourteenth century, into an Edwardian fortress. Porchester, of which one side was, as at Pevensey, washed by the waves, was similarly occupied by the Normans, who eventually raised within it one of their rectangular keeps, whilst Pevensey received the other type. At both places the strong Roman gatehouses were, with some alterations, made use of by the Normans. The third style of fortification in which our castles had their origin has an interest peculiar to itself. This is assigned to the dark period following the departure of the Romans. The natives seem to have adopted the rectangular lines of their conquerors, but to have carried them out in earth, as if incapable of constructing them in stone. To this period are assigned the works at Wareham and at Tamworth, and, with some probability, those at Wallingford and Cardiff. To them may be added the earthen extensions of the Roman defences at Lincoln and at York. We have an impression that the massive earthworks of Lincoln and Colchester castles may also belong to this period, though in this view Mr. Clark, we gather, would with us. not agree It is not, however, to such defences, but to those of the Anglo-Saxon period that Mr. Clark has devoted special attention. He claims to have been the first to set forth' the right explanation of those moated mounds, which formed, he holds, their distinctive feature. Neither British in date, nor sepulchral in origin, they were, he argues, 'thrown up, in England, in the ninth and tenth centuries,' and were introduced by the Northmen,' when they penetrated into the interior. That they may be traced to the Danish wars we think highly probable, especially from their being so often placed on the banks of rivers; but we wish that Mr. Clark had made it clearer to whom he really assigns them. In his chapter on the subject the 'Northmen' reappear as the English people, that is the Northern settlers generally, as distinguished from the Britons and the Romans'; and he traces back these strongholds to the Anglo-Saxon settlement. So too he treats the burh-' a moated mound with a table top and a base court, also moated, either appended to one side of it, or within which it stands -as the typical residence of an English lord,' or thegn. Now, either these mounds were an English institution before the Danish' invasions, or they were not. For our part, we think that their prevalence in Normandy, on which Mr. Clark rightly insists, points to their Scandinavian origin; and the Irish evidence, which is not within his province, would strengthen this conclusion. Giraldus speaks of the forts and ditches erected by the invaders in the ninth century as yet to be traced in his own day, and on the Barrow, at St. Mullin's, in Carlow, there is, we believe, still a fine Danish mote'; while the 'Thingmote,' by the Steine, at Dublin, was also a work of the Northmen. We imagine that the English adopted from their foes this peculiar method of defence, just as they seem, in turn, to have handed it on to the Welsh. Having done so, they placed the mound, with a timber stockade round its summit (as we see it represented in the Bayeux Tapestry) within Romano-British, Roman, and even British lines, till it needs all the acumen and patient care of Mr. Clark to disentangle the component parts of the stronghold. On these mounds, in later days, arose the 'shell-keeps' of the Normans; and in tracing their origin, in showing us how they were 'timbered,'-and how this enabled them to be burnt,'-above all in demonstrating their numbers and importance, Mr. Clark has made his chief addition to our knowledge of the habits of our forefathers and of early fortification. His theory as to these moated mounds is based on a careful study of the Anglo-Saxon Chronicle. No one can read the stirring story of these Danish raids that scourged our land in the latter part of the ninth century, and the earlier years of the tenth, without being struck by the peculiar tactics uniformly employed by the invaders. Their base of operations was essentially the sea, and when they landed their first thought was to throw up works to which one might apply the words of William of Poitiers on the landing of the Conqueror himself: "Quæ sibi receptaculo, navibus propugnaculo, forent.' As they began to push their way, in their long shallow keels, up the creeks and the rivers into the heart of the country, they steadily adhered to the same precaution, and wrought a work' at every point where they took up a position. From it they darted forth-on horses where they had them-to scour and pillage the country; to it they withdrew when the people rose, knowing that its shelter would enable a small band of warriors to defy a far more numerous force of raw levies. It was only in the days of Edward the Elder that the English adapted themselves systematically to these tactics, and began to oppose burh to burh, garrison to garrison. While Edward pushed his posts eastwards, his sister, the Lady of the Mercians, fortified her land towards the north, either throwing up a burh on a new site, as at Warwick, or adding it to pre-existing defences, as |