Two different methods of inoculum preparation for susceptibility testing were analysed. The first method was adjustment of inoculum size by haemocytometer counting. The second method was spectrophotometric adjustment at 530 nm. The reliability of both methods was assessed by colony counting. The overall agreement between colony counting and haemocytometer counts was 93.6%, and the intraclass coefficient correlation was 0.71 (P < 0.05). Pearson's correlation index between colony counts and optical density values was -0.059 (P > 0.05). Optical densities ranged between 0.01 and 1.2, showing less reproducibility than expected by the NCCLS M 38-P standard. Haemocytometer counting is a more reliable method of inoculum preparation for antifungal susceptibility testing.