The monumental work of Mr. T. G. Clark, _Mediæval Military Architecture_, has had no formidable rival since its appearance, but unfortunately it must now be read with care since much of the matter is obsolete. The distinction between the Saxon _burh_ and the primitive type of castle thrown up by the early Norman invaders was not apparent at the time the work appeared, and consequently many scores of castellated works are assigned to incorrect periods. This had the effect of making the chronology of the Rectangular Keep incorrect. Unhappily _The History of the Art of War_ by Oman followed Clark's lead and with, of course, the same result. Mr. J. H. Round in his _Geoffrey de Mandeville_ appears to have been one of the first, if not the first, to differentiate between the _turris_ and the _castellum_ (_i.e._ the Keep and the Ward) of medieval writers, who were proverbially loose with respect to their employment of technical terms. Excellent work also in this respect has been carried out by Mrs. E. Armitage, who, by the process of practically investigating in detail some of the defences mentioned in Domesday Book, has been able to definitely assign the Motte and Bailey type to the early Norman Period. In the recently issued _Victoria History of the Counties of England_ the effect of these discoveries is discernible in those parts relating to castellation, which very carefully correct the errors prevailing in former standard and in local topographical works. With regard to Earthworks, the invaluable investigations carried out by "The Committee upon Ancient Earthworks and Fortified Enclosures," acting in co-operation with the Society of Antiquaries, has resulted in a flood of light being thrown upon these interesting remains, so that the old allocation to British, Roman, and Danish influence, so arbitrarily insisted upon in former times according to the contour of the earthwork in question, no longer subsists, or only as far as circumstances justify the nomenclature. No generally available work is to hand dealing with these subjects in a non-technical manner, and it may be hoped that this endeavour will help to fill the interregnum between the work of Clark and a future equally monumental tome. The thanks of the Author are herewith gratefully tendered to the Congress of Archæological Societies of 1903 for permission to make use of the plans of Earthworks issued in their "Scheme for Recording Ancient Defensive Earthworks and Fortified Enclosures," and also to Mr. Cecil C. Brewer for the plans of various floors in Hedingham Keep. CHARLES H. ASHDOWN. ST. ALBANS. CONTENTS CHAPTER I PAGE NATURAL FORTRESSES STRENGTHENED 1 CHAPTER II FORTIFIED HILL-TOPS 13 CHAPTER III SIMPLE ARTIFICIAL ENCLOSURES 33 CHAPTER IV THE MOTTE AND BAILEY CASTLE 48