Currently, there is considerable debate regarding the best in vitro method for testing antifungal combinations against Candida spp. In this study, we compared the results obtained by chequerboard dilution, time-kill studies and Etest for several antifungal combinations against Candida spp. Three Candida albicans isolates (fluconazole MICs of 1.0, 32 and >256 mg/L) and three non-albicans Candida isolates (C. glabrata, C. tropicalis and C. krusei) were tested in RPMI 1640 medium. By chequerboard testing, the majority of antifungal combinations were found to be indifferent. Notably, antagonism was identified by time-kill studies and by Etest for combinations of amphotericin B-fluconazole, but it was not detected by the chequerboard method. Pre-exposure of isolates to fluconazole did not affect results of the Etest or chequerboard method, but it did increase the frequency of antagonism noted by time-kill methods. This study indicates that chequerboard dilution testing in RPMI medium may not reliably detect the attenuation of amphotericin B activity. Of the three methods, Etest was the simplest to use and yielded reproducible results for testing antifungal combinations.