The scene of the battle is slightly undulating country, the variations of level being only a few feet. The chroniclers, to whom language for expressing minute differences was wanting, talk of hills and deep valleys, and have thereby misled writers who have not seen the ground, nor examined with attention a contoured map. South-eastwards from Poitiers runs the modern Limoges road, almost parallel to an ancient Roman road, which may have been still the working road of the fourteenth century. A small rivulet, the Miosson, flows at the bottom of a ravine, about 100 feet below the level of the battle-field, and joins the Clain just above Poitiers. The bottom is presumably muddy, and the quantity of water varies greatly with the season. But there is a ford (the Gué de l'Homme marked on the map) to which a narrow road, believed on good evidence to be ancient, leads from close to La Cardinerie. That farm itself is not so old as the battle, having taken the place of the hamlet of Maupertuis, which stood somewhere in the same neighbourhood, and is said to have been destroyed at the time of the battle. Maupertuis was[28] supplied with water from a pond, now almost filled up, which used to be known as "la mare aux Anglais," and out of which sundry relics of the battle have been taken. The overflow of this pond, and doubtless the surface drainage of the immediate neighbourhood, which in rainy weather might be considerable, passed down a very slight hollow running nearly north and south on the Poitiers side of La Cardinerie. As the soil is soft, and the slope very gentle till near the Miosson, the bottom of this hollow may well have been boggy. It is a good illustration of the exaggerated impression conveyed by the defective vocabulary of the chroniclers, that this depression of a very few feet is the place best answering to the _profunda vallis_, and the _torrens_ of Baker of Swinbrook, the chronicler whose narrative of the battle has a far greater air of precision in details than any other.