Prostaglandins (PGs) have been assumed to play a role in the biological activity of thymic hormones (TH). Indeed, it has been shown that type E-PGs are able to mimic the action of several TH. Moreover, indomethacin interferes in the rosette assay, which still represents the most commonly used bioassay for the evaluation of TH and, in particular thymulin levels, in biological fluids. Previously, our attempt to modulate PG production by different TH showed that none of the TH tested affect PGE2, 6-keto-PGF1 alpha, PGF2 alpha and TXB2 production by spleen cells from control and thymectomized (Tx) mice, while indomethacin was able to inhibit the spontaneous PG production. Here, we investigated a possible role for each endogenously produced PG in the experimental conditions of the rosette assay, in order to define: 1) whether or not there was a specificity of action of a given PG; and 2) to analyze the pattern of action between thymulin and the endogenously produced PGs. We demonstrated that PGE2 and 6-keto-PGF1 alpha are the PGs which are physiologically involved in the rosette assay, according to their levels of endogenous production, and that they are able to synergize with thymulin. This synergy was demonstrated in two ways: 1) by adding anti-PGE2 and anti-6-keto PGF1 alpha-antibodies, which prevent part of the thymulin effect, or 2) by simultaneous addition of PG and thymulin, at concentrations far lower than those which correspond to their thymulin-like effect. Moreover, PGE2 addition, at concentration close to that found to be endogenously produced, partially reversed the indomethacin-induced effect in the rosette assay. In conclusion, if PGs do not act as mediators of thymulin, they are able to synergize in one of its biological action.