To assess the effects of guar gum on colonic microbial fermentation and cancer development, azoxymethane-treated rats were fed a partially hydrolyzed guar or control diet. Anaerobic fecal incubations were conducted at 8-wk intervals, either without added substrate or with cornstarch or hydrolyzed guar as substrates. Short-chain fatty acids in colonic contents and colonic carcinoma areas were measured at 27 wk. Fecal in vitro fermentation rates were higher for guar-fed rats than for control rats [three-way ANOVA (diet, time, in vitro substrates), P = 0.002]. Fecal in vitro butyrate production was greater for guar-fed rats than for control rats after 3-11 weeks of diet treatment (three-way ANOVA, P = 0.027). Butyrate concentrations of colonic contents at 27 wk were higher in guar-fed than in control rats and higher in the cecum than in the post-cecal colon (two-way ANOVA, P = 0.0001). A regression equation predicting colonic carcinoma area (r2 = 0.279) using propionate and butyrate concentrations of the contents of the post-cecal colon showed propionate as a positive predictor (P < 0.001) and butyrate as a negative predictor (P = 0.033). Our results show that patterns of short-chain fatty acid production may affect the results of fiber-carcinogenesis experiments. Dietary addition of hydrolyzed guar is associated with fecal fermentation low in propionate and high in butyrate; short-chain fatty acid concentrations are greater proximally than distally. These results suggest that butyrate protects against colonic neoplasia, whereas propionate enhances it, and demonstrate that colonic microbiota adapt to produce more butyrate if given time and the proper substrate.