Recent studies within Devon have recognised the limitations of using upland pollen diagrams to infer vegetation change and land use in lowland landscapes. These studies provide a detailed record of local landscape histories within a lowland context during the mid-to-Iate Holocene. However, there remain considerable tracts of the lowland landscape for which little is known of the vegetation history and land use. This thesis presents the results from a palynological study (undertaken as part of the Community Landscapes Project) of three distinct physical landscapes in lowland Devon: the Clyst valley, the Blackdown Hills and the Hartland Peninsula. The study is significant because it has increased understanding of the lowland landscape and in particular has created local landscape histories in areas which have very little or no published material. The use of small pollen sites has allowed land use and vegetation history to be reconstructed on a local scale and has provided a comparison between other local environmental histories to suggest regional trends in land use. The study employs an holistic approach to land use and vegetation history using archaeological and historical sources (where possible) to provide a cultural context to the pollen diagrams and demonstrates that the use of archaeological and historical data can aid in the interpretation of lowland local pollen diagrams and in the creation of local landscape histories. The results suggest the early Holocene lowland landscape of Devon had more in common with the landscape of the southern lowlands of England than with upland areas of the South West. Multiple elm decline events occurred across the lowland landscape although the first appearance of cereal pollen appears to have occurred much later in the study areas. The nature and timing of mid-Holocene clearance was clearly asynchronous but once the agricultural landscape was established there appears to have been very little change in the vegetation history of all three study areas indicating the landscape oflowland Devon is remarkable in its continuity in the later Holocene. The Roman period is only visible in one of the study areas (Clyst valley) and there appears to be no evidence of Post Roman regeneration. There is also no indication of convertible husbandry during the Medieval period which is in direct contrast to the early Medieval landscapes elsewhere in Devon. The pollen profiles from this study have also suggested intra-regional variations in land use and vegetation history spaning both the prehistoric and historic periods and indicate that within large or distinct landscapes (such as the Blackdown Hills) whilst there is a general trend towards continuity there are also smaller pockets of environmentally and perhaps culturally distinct landscapes. Finally, this study also suggests care must be taken in undertaking palynological studies from lowland local sites because such sites may suffer from poor temporal resolution, complicated stratigraphy (both of which will affect radiocarbon dates) and may have been subject to peat cutting and dynamic hydrological processes throughout their history.