## Abstract

Three-hundred-year histories of pH and total alkalinity (alk) have been inferred from diatom and chrysophyte remains in deep-water sediment cores from Mud Pond (pH 4.6, alk -23 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>) and Little Long Pond (pH 5.7, alk 4 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>), Maine and Haystack Pond (pH 4.8, alk -18 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>), Vermont. Three replicate cores were studied from each Mud Pond and Haystack Pond; one core from Little Long Pond; pH and alk inferences from diatoms were based on three different calibration equations: CLUSTER, DECORANA and CCA (CANOCO) (only CCA for chrysophytes). Replication of pH and alk inferences between cores was excellent. Different calibration approaches led to the same conclusions with minor exceptions. There were minor differences between chrysophyte- and diatom-based inferences, but both led to similar conclusions regarding acidification. These were: Mud Pond, ca. 1700-1925, pH 5.2-5.3, alk 0 to -15 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>; 1925-1970, acidification to pH <latex>$\simeq $</latex> 4.8 and alk -20 to -30 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>. Little Long Pond ca. 1700-1950, pH <latex>$\simeq $</latex> 5.9, alk 20-50 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>; 1950 ff., possible slight acidification to pH 5.7-5.8 Haystack Pond, ca. 1700-1925, pH 5.2-5.3, alk 0 to -10 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>; 1925-1970, acidification to pH <latex>$\simeq $</latex> 4.9 and alk -10 to -30 <latex>$\mu $</latex>eq l<latex>$^{-1}$</latex>. Correlation of lake acidification with great increases in sedimentary indicators of air pollution (carbonaceous particles, Pb, polycyclic aromatic hydrocarbons) and absence of correlated catchment disturbance point to anthropogenic acid deposition as the cause of lake acidification. Extreme acid sensitive lakes like these three are atypical for northern New England.