That crossing place was, of course, finally discovered opposite the steep gravel bank upon which the oldest part of the City of London is built. The land has been so often turned and returned in at least twenty centuries of building that it is not easy to-day to reconstruct the original conditions of that crossing; and, unless we look at all the evidence, slight as it is, it is easy to fall into errors upon it. Thus several writers upon this subject have often spoken as though the gravel-topped knoll upon which the original London stood was the sole factor in establishing the crossing, and I have myself fallen into the error of believing that the approach from the south could only be made over a long artificial causeway. A further consideration of the evidence, and especially of that concerning surface soils to the south of the crossing, has convinced me (subject to yet further evidence which may appear) that the opportunity for a crossing near the site of London Bridge was almost as tempting from the south side as it was from the north. It is true that no considerable rise of land is to be discovered on the Southwark side until we have gone some distance from the river, and the contour lines do not, therefore, suggest an easy crossing at this spot. But much more important than the lie of the land was the nature of the surface over which travel must proceed. The rocks across which a road is driven are not of the first importance in primitive times, though they become important, of course, when the road is expected to bear very great loads, or when it is so thoroughly metalled that the presence of good stone in its neighbourhood has to be considered. What is important to a primitive track is the immediate soil under foot, and if that be fairly hard and dry it can be quite shallow and yet sufficient for the purposes of travel.