The potential of low viscosity silicone-based impression material for measuring stomatal aperture on intact leaves was investigated. It was found that the probability of successful replication of Commelina communis L. stomata depended on the width of the stomatal pores. The wider pores in a sample were more likely to be measured, so the resulting estimated mean was an overestimate of the true mean. This bias in estimates of stomatal aperture from impressions was studied and quantified. A transformation is described which was used to adjust apparent aperture measurements from impressions to give more accurate estimates of stomatal apertures on intact leaves of C. communis. Maximum absolute errors due to the bias usually occurred at apparent apertures on impressions of between 2 and 5 μm but were proportionally greatest below 3μm, where they often exceeded 40%. At a given aperture, the error depended upon the dispersion of the stomatal apertures about their mean and was greatest when this was large.