The effect of the flavonoid quercetin and its conjugate rutin was investigated on (biomarkers of) colorectal cancer (CRC). Male F344 rats (n = 42/group) were fed 0, 0.1, 1, or 10 g quercetin/kg diet or 40 g rutin/kg diet. Two wk after initial administration of experimental diets, rats were given 2 weekly subcutaneous injections with 15 mg/kg body wt azoxymethane (AOM). At wk 38 post-AOM, quercetin dose dependently (P < 0.05) decreased the tumor incidence, multiplicity, and size, whereas tumor incidences were comparable in control (50%) and rutin (45%) groups. The number of aberrant crypt foci (ACF) in unsectioned colons at wk 8 did not correlate with the tumor incidence at wk 38. Moreover, at wk 8 post-AOM, the number and multiplicity of ACF with or without accumulation of β-catenin were not affected by the 10 g quercetin/kg diet. In contrast, another class of CRC-biomarkers, β-catenin accumulated crypts, contained less β-catenin than in controls (P < 0.05). After enzymatic deconjugation, the plasma concentration of 3′-O-methyl-quercetin and quercetin at wk 8 was inversely correlated with the tumor incidence at wk 38 (r = −0.95, P ≤ 0.05). Rats supplemented with 40 g rutin/kg diet had only 30% of the (3′-O-methyl-) quercetin concentration of 10 g quercetin/kg diet-fed rats (P < 0.001). In conclusion, quercetin, but not rutin, at a high dose reduced colorectal carcinogenesis in AOM-treated rats, which was not reflected by changes in ACF-parameters. The lack of protection by rutin is probably due to its low bioavailability.