An experiment was performed during the summer of 1982 to determine the effects of simulated acidic rain on seed yields on two commercial cultivars of soybeans grown according to standard agronomic practices. Plants were shielded from all ambient rainfalls automatically by two amount exclusion covers and exposed to simulated rainfalls m quantities equal to the average amount of rainfall that occurs at the site. Seed yields of cv. Amsoy exposed twice weekly to simulated rain of pH 4.1, 3.3 and 2.7 were, respectively, 3.0, 9.0, and 12.8% below yields of plants exposed to simulated rain of pH 5.6. A treatment-response function of seed yield vs rainfall pH for Amsoy was y= 10.20+0.587 x and had a correlation coefficient of 0.96 (y is seed mass per plant and x is the pH of the simulated rain). For cv, Williams, seed yields of plants exposed to simulated rainfalls of pH 5.6, 4.1, 3 3 and 2.7 were 11.5, 10.5, 11.4 and 11.4 g, respectively. A treatment-response function of seed yield vs rainfall pH for Williams was y= 11.13 + 0.016x which had a correlation coefficient of 0.038. Plants of Amsoy and Williams grown in plots adjacent to the exclusion shelters had mean seed yields of 11.4 and 9.8 g per plant, respectively. Seed yield per plant was dependent upon number of pods per plant because the number of seeds per pod did not vary among treatments for each cultivar.