Adult African clawed frogs (Xenopus laevis) were collected from a corn-growing region (CGR) and a non-corn-growing region (NCGR) with different exposure profiles for atrazine and related triazines. Physical, chemical, and biological parameters from the catchment areas were also measured. Frogs were surveyed for possible effects of exposure to triazine herbicides on plasma testosterone (T) and estradiol (E2) titers, gonadal aromatase activity, and gonad growth (GSI). Concentrations of both T and E2 varied among locations and were correlated to some accessory factors, such as pH, several ions, and metals. Greatest median plasma T concentrations (males: 19 ng/ml; females: 16 ng/ml) occurred in frogs inhabiting NCGR as compared to those from the CGR (males: 4 ng/ml; females: 1 ng/ml). Median E2 concentrations were also greater in frogs collected from the NCGR (males: 3 ng/ ml; females: 28 ng/ml) than those in frogs from the CGR (males: 2 ng/ml; females: 5 ng/ml). Because some exposure to agricultural chemicals at both regions occurred, as did simultaneous exposures to multiple chemicals, a regression analysis was employed. Negative correlations were observed between plasma T concentrations and concentrations of atrazine, deisopropylatrazine, deethylatrazine, and tertbuthylazine in females and between T and diaminochlorotriazine in males. Estradiol in females exhibited a significant negative correlation with atrazine and deethylatrazine. No correlations were observed between gonadal aromatase activity or GSI and any of the agricultural chemicals measured. Median aromatase activities in ovaries varied among sampling sites ranging from 7 to >3,000 times greater than those in males when measurable. Testicular aromatase activity was below the detection limit of the assay in male frogs at most of the sites. Although exposure to agricultural inputs did not affect aromatase activities, effects of atrazine or coapplied pesticides on sex steroid homeostasis cannot be excluded at this point.