This work sought to evaluate the accuracy and reproducibility of a replication process that formed electroconductive replicas from addition cured silicone impressions for the purposes of monitoring tooth surface loss. Replicas were constructed by painting the surface of impressions with one of three high silver content electroconductive paints [Electrocure (EC), RS Silver Paint (SP) and Electrolube (EB)] and, once dry, backing this up with a cyanoacrylate based gel material (Zapit) and die stone. For each paint an impression was recorded of a gauge block of known step height (1270 microm) and five impressions of the palatal surface of a laboratory standard maxillary central incisor were also recorded. All impressions were taken using an addition cured silicone impression material (President). The resultant electroconductive replicas were mapped using a computer controlled probe. This data was analysed to determine both the replica's step height and, using surface matching techniques, the reproducibility of the replication process. The mean step heights recorded were; EC=1268.43 (s.d.=12.09), SP=1267.09 (s.d. 1.77) and EB 1299.58 (s.d. 14.47). Both EC and SP recorded the step height to within 3 microm. In the repeat replication of the palatal surface of an upper central incisor SP maintained a greater constancy of surface topography (99.6%) that was statistically superior to both EC (P < 0.05) and EB (P < 0.01). The SP was therefore the preferred paint for this technique. Bearing this in mind the replication technique was both accurate and reproducible but vigilance, as to the selection of paint for the process, should be exercised so as not to affect the good accuracy and reproducibility of the technique.