ABSTRACT
Not so long ago I was working on a small trial excavation in the village of Castor in eastern England; it involved cutting a narrow trench into a beautiful old orchard garden backing on to a churchyard in order to find evidence for a Roman palace which once occupied almost the whole village. In the end, we did find the remains of Roman buildings on a terrace, as well as a great deal of subsequent occupation which ceased sometime in the twelfth or thirteenth century when the area became part of the churchyard. The process of excavation involved using a machine to strip off the garden soil, followed by hand digging with mattocks, spades, shovels and trowels. In the process, we sought to identify separate deposits marked by differences in their composition, deposits such as slopewash, floor layers, pit fills, walls and so on. Each of these was described on separate record sheets accompanied by measured drawings to scale, and identified by a unique number; any artefacts or other remains such as animal bones or shells were bagged and labelled according to the deposit they came from. Critical to the whole process was understanding both what any deposit represented and what its relationship was to other deposits, i.e. earlier, later or contemporary.
