THE DATING OF THE SITES - EARLY CHALCOLITHIC AND BRONZE AGE
Obviously, such datings must be used with caution. A risk pertaining to many of the areas surveyed was that the sherd sample indicated not the full occupation of the site but only its later phases. This was particularly relevant in the case of mounds on an alluvial plain which often became partially submerged by silty deposits concealing all trace of their earliest levels. Other problems arise from selection and recognition. Accepting that sampling is more often than not weighted in favour of decorated sherds which are more readily identifiable than plain wares, a change in ceramic fashion resulting in the use of undecorated pottery might well go un recognised in surface collecting, unless the forms were particularly distinctive, and a corresponding period of occupation could pass unnoticed. It is just possible that such a change could provide an explanation for the apparently widespread abandonment of sites in the second half of the third millennium reflected in Table 2.