Palynology (on both peat cores and turves from ramparts) has been used both to support and to reject the proposal that the presence of the Roman army and the pax Romanorum affected the landscape (e.g. Whittington and Edwards 1993; Tipping and Tisdall 2005; Hanson 1996 contra Dumayne and Barber 1994). It is now accepted that any such changes are difficult to recognise and that much more work on closely-dated sequences is needed in order to be able to take this forward. It has been argued that there should be an effect on indigenous settlements, perhaps a growth in their number and size, but such sites are difficult to date without modern excavation. Furthermore, a possible counterbalance, the forced export of men for service elsewhere in the Roman army, is equally difficult to recognise in the archaeological record. In short, it is difficult as yet to see in any aspect of the archaeological evidence an impact of the Roman army on the Scottish landscape, though this should only encourage further work. Tipping and Tisdall (2005) have provided a thorough overview of the landscape context of the Antonine Wall, but other work has, generally speaking, been more piecemeal.
More (targeted) pollen analysis to create a Scotland-wide view of the environment during the late pre-Roman Iron Age, Roman and post-Roman periods.
Work focused on the environs of specific forts, where suitable samples can be identified, would allow identification of local variation in conditions. In favourable circumstances, closely-spaced sampling, analysis of several closely-adjacent cores, and selection of different locations to give local as well as regional pictures, may be able to provide a detailed appraisal of environmental changes and their causes.
Further studies from north-east Scotland are needed to test the conclusions reached by Whittington and Edwards (1993) on a limited range of samples