Investigations into Mesolithic and Neolithic activity and environmental impact on the Cambridgeshire fen-edge are described, consisting of stratigraphic and pollen analytical research at Peacock's Farm, and trial excavations at Peacock's Farm and Letter F Farm. At Peacock's Farm, the bulk of archaeological occupation was found to be of Mesolithic date, around 8000 BP uncalibrated; minimal signs of Earlier Neolithic activity were recovered. At Letter F Farm there was mainly Earlier Neolithic occupation, around 5000 BP, but there had also been Mesolithic activity. Radiocarbon dating at Peacock's Farm shows that the Mesolithic black band within the peats flanking the sand ridge, first described by Clark et al. (1935), covers a surprisingly long period: over 1700 years, mainly between approximately 8500 and 6800 BP. Dates from Mesolithic occupation areas on the sand ridge coincide with the first half of this period. Consideration of the stratigraphic results, radiocarbon dates and two pollen diagrams suggests that a channel was eroded through the Mesolithic black band at an early stage of its formation, probably not long before 7500 BP. In one area infilling of the channel apparently took place in a number of stages; in another it filled progressively with shell muds. A tentative reconstruction of the sedimentary, environmental and archaeological sequence is made. Before the Mesolithic occupation the landscape appears to have been densely forested both on the wetland and the elevated sands. Minor damage to this cover took place around 8500 BP, coinciding with the beginning of the Mesolithic occupation. This was followed, at about 8250 BP, by a substantial opening of the forest cover, when the site may have been more actively used than before, possibly as part of a new settlement pattern. Relatively open local conditions persisted for some 700–1500 years before the forest cover was re-established. Regeneration may have involved alder as a colonist, coinciding with the classical Boreal-Atlantic transition of Godwin. The relative importance of human impact and the occurrence of a period of dry climate are discussed. The balance of evidence, some admittedly circumstantial, points to a pronounced human impact on the local environment in the Mesolithic period. Burning may have been connected with short-stay visits in a settlement pattern spanning both wetland and dry areas; the context for this apparent lowland change might be sought in the insulation of the British Isles in the 9th millennium BP and increased territoriality from that date as reflected in microlith styles. By contrast in the Neolithic period there is very little pollen evidence of local environmental damage at Peacock's Farm. The Neolithic archaeological evidence from Peacock's and Letter F Farms suggests small short-stay visits only, as part of a regionally now more differentiated settlement pattern.