Obviously special methods have had to be developed to deal with these problems. In particular what is called 'open area' excavation is now recognised as being the only way of gaining a complete picture. This means the large-scale stripping of the remains of each period in its entirety before going down to the next level. The technique was first developed in Denmark but the English site which has led
the way in developing 'open area' excavation is the deserted village on the chalk uplands of the North Riding in Yorkshire, Wharram Percy. I Here since 1952 the Deserted Medieval Village Research Group has been engaged in excavations.2 Aerial photographs and ground observations very clearly pinpointed house-sites, roads and field systems. First the area to be excavated was divided into 1.52 m squares, and a contour survey was made. The turf was then stripped and the soil removed to expose a rubbly level. This is material from the collapsed walls of the buildings. In the second stage, all the small stones which were obviously not part of walls were removed, leaving in place all stones larger than 15 em across. These might be parts of walls ill siw or they might be simply tumble or odd boulders lying about. They had to be recorded in detail so that if it was found at a later stage that they had been removed in error, the missing parts could be reconstructed. Gradually the tumble was then removed to expose the wall lincs undcrncath. It became apparent at this stage at Wharram Percy that there was a complex series of buildings, and that the earlier ones were often on different alignments. As the buildings of each period were recorded they were removed exposing the earlier remains beneath. These were built of timber, and the post-holes and slots, even more difficult to trace than the flimsy stone walls, could only be seen when the whole site was cleared to this level. Baulks were left which recorded sections across the site at each level. When these were recorded they were removed to expose the whole level. The next level was then peeled off leaving again a baulk exposing the section. In this way it was possible to make a vertical
record of the stratigraphy which did not get in the way of the main purpose which was to expose a large area, a level at a time. The final technique used was to record the finds. To begin with this was done by plotting all the finds three dimensionally but it was subsequently realised that it was only necessary for small finds and significant potsherds to be bagged according to the level and 1.52 m grid square. Only the datable finds are fully recorded three dimensionally.