It is to documents, then, that we must look, and, unfortunately for this action, our principal document is not contemporary. It is from the pen of Froissart, who was but nine years old when the battle was fought, and who wrote many years after its occurrence. Even so, his earlier version does not seem to be familiar to the public of this country, though it is certainly the more accurate. Froissart used a contemporary document proceeding from the pen of one "John the Fair," a canon of Liége. Of the lesser authorities some are contemporary: notably Baker of Swynford, and Villani, who died shortly after the battle. But the whole bulk of material at our disposal is pitifully small, and the greater part of what the reader will have set before him in what follows is the result of an expansion and criticism of the few details which writers of the period have bequeathed to us. When the documentary evidence, contemporary, or as nearly contemporary as possible, has been tabulated, the historian of a medieval battle next proceeds to consider what may be called the "limiting circumstances" within which the action developed, and these have much more than a negative value. As he proceeds to examine and to compare them, they illuminate many a doubtful point and expand many an obscure allusion. For instance, in the case of Crécy, we carefully consider the contours, upon the modern map, of a terrain which no considerable building operations or mining has disfigured. We mark the ascertainable point at which the Somme was crossed, and calculate the minimum time in which a host of the least size to which we can limit Edward's force could have marched from that to the various points mentioned in the approach to the battle-field. We ascertain the distance from the scene of action to the forest boundary. We argue from the original royal possession and subsequent conservation of that forest its permanent limits. We can even establish with some accuracy the direction of the wind, knowing how the armies marched, how the sun stood relative to the advancing force, and their impression of the storm that broke upon them. We calculate, within certain limits of error, the distance necessary for deployment. We argue from the known character of the armour and weapons employed certain details of the attack and defence. We mark what were certainly the ancient roads, and we measure the permanent obstacles afforded by the physical nature of the field. I give these few points as examples only. They are multiplied indefinitely as one's study proceeds, and in the result a fairly accurate description of so famous, though so ill attested, an action as this of Crécy can be reconstituted.