Progesterone levels were estimated from sera of bonnet monkeys which were bled biweekly using a radioimmunoassay employing reagents and protocol as prescribed by the W.H.O. Matched Assay Reagent Programme (Sufi et al., 1983). Each sample was run in duplicates. Progesterone was extracted from serum (0.1 ml) by the addition of 2 ml of ice-cold ether in each tube and vortexing for 2 min. The tube was immersed in liquid nitrogen in order to flash freeze the serum phase and the unfrozen ether phase which contained the extracted steroid hormone was decanted into another tube. The ether was allowed to evaporate 0/N and 0.5 ml of steroid assay buffer (0.1 M PBS, pH 7 .3, 0. 1% thiomersal and 0.1% gelatin) was added to the tubes and the tubes were incubated at 40°C for 30 min. Steroid sticking to the walls of the tubes was recovered by vigorous vortexing. 100 J..LI of anti-progesterone Ab (at a dilution giving -50% binding of tritiated p.rogesterone in the absence of unlabelled competing progesterone) was then added to the tubes followed by addition of 0.1 ml of 3H-progesterone ( -10,000 cpm/tube). The mixture was incubated for atleast 16 hrs at 4oc. Unbound progesterone

was separated by addition of 0.2 ml of ice cold assay buffer containing 0.625% activated charcoal and 0.0625% dextran and incubated for 30 min at 4oc. This was followed by centrifugation at 2500 rpm for I 5 min at 4°C. The supernatant was carefully decanted into scintillation vials and 4 ml of scintillation fluid (0.4% 2,5 diphenoxazole; 0.01% POPOP [1-4 bis(5-phenyl-2-oxazolyl)benzene] in sulfur free toluene) was added and counted in a liquid scintillation beta counter (Beckman Instruments, California, USA). The amount of progesterone per ml of serum was calculated from a standard curve with known amounts of progesterone in each assay.