STUDIES IN THE VEGETATIONAL HISTORY OF MID WALES

SUMMARY The basal sediments of three pingo basins from the valleys of the Cledlyn and Cletwr in west Wales have been analysed palynologically. The basal pollen‐bearing deposits indicate that sedimentation began in all three basins at a time prior to the invasion of trees and before the initial expansion of Juniperus at the opening of the Flandrian. Radiocarbon dating confirms that the subsequent Juniperus peak is early Flandrian in age rather than belonging to the Allerød interstadial. The vegetational succession documented at the three sites allows a reconstruction of both regional and local developments during the transition from late‐Devensian to Flandrian times. These events broadly resemble those described for other parts of mid Wales, but with certain, notable differences. These sites in the Teifi Valley system have very low proportions of montane and arctic‐alpine types represented in their pollen records; this contrasts markedly with the site previously described from the uplands at Elan Valley. Also they do not give indication of the early Corylus rise which occurred just 20 km north at Gwarllyn. The likely explanation for this is that invasion and spread of Corylus was localized in what is now the Cardigan Bay area and pollen derived from this early expansion was carried to the exposed Mynnedd Bach area, but did not penetrate south‐eastwards into the Teifi Valley region. Evidence from these sites suggests that Pinus sylvestris and Alnus glutinosa were established locally in the area soon after the demise of Juniperus. In addition to providing data on vegetational history in this area, these studies also supply information concerning the origin and decay of pingos during late‐Devensian times. The fact that the basal sediments predate the Juniperus rise demonstrates that the melting of the pingo ice cores and the collapse of their domes occurred before temperatures were high enough to allow Juniperus to attain its maximum flowering potential. This in turn would indicate that pingo ice melt has a lower temperature threshold than Juniperus flowering and that the threshold was therefore crossed first. This being so, it is reasonable to deduce that the pingos could not have survived the late‐Devensian interstadial (Allerød), for during it, even in the uplands, the Juniperus flowering threshold was crossed. All of the pingos examined, therefore, appear to have formed and decayed during the post‐interstadial cold phase (Godwin's zone III).