Silicon dioxide, commonly referred to as silica, is present in plant cell walls and interstitial spaces, and is often found as a component of dietary fibers that have exhibited hypocholesterolemic activity in animals. The primary objective of this study was to determine the cholesterolemic effects of two different morphological forms of silicon in the diet of cholesterol-fed rats. Male Wistar rats were provided diets containing 1 g cholesterol/100 g diet, and 0.65 g Si/100 g diet as either a sodium salt (silicate group) or silicon dioxide, a synthetic silica polymer (silica group). Cellulose was used as a control (control group). The in vitro bile acid binding capacity of the SiO2 was also measured. After 44 d of diet administration, animals were deprived of food for 24 h and then killed. Plasma total, VLDL, and LDL cholesterol concentrations were 18%, 29%, and 26% lower, respectively, in the silica group than in the control group. However, liver cholesterol concentrations were not different among dietary treatments. During the initial 15 d of the study, average daily total fecal bile acids were 38% higher in the silica group than in the control group, but fecal bile acid outputs were not different for the remainder of the experiment. The silica polymer used in the feeding trial was found to adsorb 5 times more cholate than chenodeoxycholate, at pH 7.5 in vitro. In vivo, the potential for silica to enhance fecal cholic acid excretion, relative to chenodeoxycholic acid during the initial stage of the study, may have contributed to the hypocholesterolemic response to the silica diet.(ABSTRACT TRUNCATED AT 250 WORDS)