From a total of 40 Swedish diary A.I. bulls, the number of viable spermatozoa in frozen-thawed semen was estimated using a commercial luminometer (SVT) to determine ATP contents and by fluorescence microscopic examination of samples stained with the combined fluorophore probes Calcein AM and Ethidium homodimer (CAM/EthD-1). The bulls were of two breeds (24 Swedish Red and White and 16 Holstein) and two age classes (1-2 and 6-10 years old, prior to and after progeny testing, respectively). The SVT-instrument was able to estimate a frequency of viable spermatozoa (47.0 +/- 14.4%; means +/- SD) close to that recorded by the fluorophore probes (spermatozoa with intact plasmalemma and progressively motile, 53.6 +/- 11.5%) and by subjective visual assessment of sperm motility (53.4 +/- 6.0%). The percentage of motile spermatozoa assessed under phase-contrast was positively (r = 0.33, P < 0.01) correlated with both the total ATP content, the percentage of viable spermatozoa recalculated from the SVT reading (r = 0.26, P < 0.05) as well as the frequency of spermatozoa having intact membranes as assessed by CAM/EthD-1 (r = 0.45, P < 0.001). The latter was also significantly correlated (r = 0.59, P < 0.001) with the total ATP contents. No significant variation in ATP contents was present between bulls, breeds or age. A significant difference, however, was seen between bulls for sperm membrane integrity. The results indicate that the SVT instrument seems to provide a quick and easy estimation of the number of viable spermatozoa in frozen-thawed bull semen. The assessment of membrane integrity with fluorophore probes allows the estimation of sperm subpopulations within the sample and seems therefore to be more useful and objective to determine sperm viability than visual assessment of sperm motility.