Extended R-value (ERV) models are essential tools for exploring and modeling quantitative relationships between pollen and vegetation percentages. Application of these models allows correction for the Fagerlind effect, and hence yields better estimates of calibration parameters for application to sedimentary pollen percentages. ERV models have not been widely applied, in part owing to the inavailability of suitably specialized computer programs. Development of the Fagerlnd program, an ERV program that will run on IBM-compatible personal computers, is aimed at rectifying that situation. Application of the program to a pollen-vegetation data set from southern New England illustrates the power of ERV models in studying pollen-vegetation relationships, and demonstrates that appropriate spatial scaling and distance-weighting of vegetation data are required for accurate estimation of calibration parameters.