Radioimmunoassay systems are described which have been developed to quantitate two principle urinary metabolites of PGF2alpha; 9alpha,11alpha-dihydroxy-15-oxo-2,3,4,5-tetranorprostanoic acid (I) and 9alpha,11alpha-dihydroxy-15-oxo-2,3,4,5-tetranorprosta-1,20-dioic acid (II). Preparation of the required metabolites was achieved by total synthesis (I) or by bioconversion (isolation from urine of animals treated with 15-keto-PGF2alpha, II). These metabolites were used to prepare conjugates for immunization. Labelled metabolites, suitable as binding markers, were prepared by metabolism of 3-H-PGF2alpha in vitro (I) or in vivo (II). Specificity of the resulting antibodies was compared to an antibody to PGF2alpha and to 13,14-dihydro-15-keto PGF2alpha. Antisera of II had little or no affinity for 20-carbon precursors (PGF2alpha or 13,14-dihydro-15-keto PGF2alpha), but had nearly equal affinity for metabolite I. Antisera of I, however, had little or no affinity for antigen of II. Therefore, analysis of samples by both assay systems enables quantitation of these excretion products of PGF2alpha. Other assay parameters (binding, affinity, recovery, precision and the repeatability of the assays) were similar to those previously described for other RIA systems, and were considered satisfactory for quantitation of compounds in biological fluids. Quantitation of 24 hour urinary excretion of di-acid metabolite in humans was in close agreement with previously published values determined by physical-chemical means. Greater quantity of di-acid metabolite was excreted by human males (42.0 mug/24 hr) than by females sampled either during the follicular (20.0) or luteal phase (21.2) of the menstrual cycle. The total quantity of C-16 metabolites (as approximated by system II) excreted/kg body weight by the rhesus monkey was similar to that excreted by the human. However, the ratio of di-acid to monoacid was much nearer unity in the monkey than the human.