The mechanism by which prostaglandin F2(α) terminates luteal function in the sheep is unclear even though it is used extensively in animal husbandry. At the time of luteal regression, a decrease in 3β-hydroxysteroid dehydrogenase (3β-HSD) activity is apparent in the corpus luteum, but is is not known whether the decrease in enzyme activity is the primary cause of structural luteolysis. The effect of trilostane, a 3β-HSD inhibitor, on luteal function and morphology has therefore been investigated. Intravenous injection of trilostane in the mid-luteal phase of the oestrous cycle caused a decrease in ovarian tissue progesterone content. A transient decrease in peripheral and utero-ovarian vein plasma progesterone was observed but there was no significant effect on the length of the luteal phase of the cycle. There was no significant change in plasma 13,14-dihydro-15-oxo-prostaglandin F2(α) during the period when plasma progesterone was depressed. Morphological examination of the corpora lutea revealed a decrease in the concentration of electron-dense granules without any other features of impending luteal regression. When plasma progesterone was reduced for more than 10 h by two injections of trilostane 4 h apart, there was again no subsequent effect on the length of the oestrous cycle or on the return to oestrus. Plasma progesterone returned to preinjection levels within 24 h of injection. This evidence suggests that competitive inhibition of 3β-HSD activity, per se, is ineffective in bringing about structural luteolysis.