Electrical conductance measurements are being used experimentally to diagnose caries. Current equipment, e.g. the electronic caries monitor (ECM), uses a probe to scan occlusal fissures. For full-mouth examination this method is rather time-consuming. A method with which only one measurement is needed for an entire (occlusal) surface would be preferable. However, the enlargement of the area being measured will influence the conductance. It was the purpose of this study to investigate the relationship between the electrical conductance of human teeth and the enamel electrode area, and to compare the range of results of surface measurements with those of the scanning method. Twenty-five sound extracted teeth were selected for the study. The reference electrode of the ECM was connected to the roots. The buccal surface was blotted dry, and a coloured dentifrice was syringed in increments onto the surface. After each increment the surface was photographed together with a metric reference, a conductance measurement was performed by holding the ECM probe tip in the dentifrice, and the ECM reading was recorded. For each tooth between 5 and 10 increments were applied and measured. The photographs were digitised and the electrode areas were calculated. A least squares curve fitting procedure yielded a linear relationship between conductance and electrode area (0.88 less than or equal to R(2) less than or equal to 1.0, mean R(2) = 0.97). For most teeth the threshold for dentinal caries as used for scanning ECM measurements (ECM reading = 6.00) was reached only when the electrode area exceeded 12 mm(2). For 6 teeth this conductance was already reached between 5 and 12 mm(2) Electrical conductance measurements of the occlusal surfaces of 18 teeth, with electrode areas as projected two-dimensionally of between 5 and 10 mm(2), showed only 3 measurements above the threshold. None of those 3 came from the group of 6 teeth mentioned before. It was concluded that the relationship between electrical conductance of teeth and electrode area is linear. It was also concluded that when electrical conductance measurements of occlusal surfaces are to be made, the threshold for dentinal caries will probably have to be shifted to a higher conductance level to limit false-positive results.