Endostatin, one of the most potent negative regulators of angiogenesis, is naturally occurring as an inhibitor of angiogenesis capable of inhibiting tumor growth and their metastases. We aimed to investigate the in vivo activities of low dose of recombinant human endostatin on 1,2-dimethylhydrazine (DMH)-induced mice colon cancer. Thirty male Balb-c mice were injected with DMH (20 mg/kg/week) subcutaneously once a week for 12 weeks to induce colon cancer. Twelve weeks after the last DMH injection, 7 mu g rh-endostatin was injected every day for 6 weeks. The animals were killed after 30 weeks for histopathological examination. The weight of the animals, tumor inhibition rates, death rates and the distribution of the lesions in colon were evaluated after the mice were killed. The mean colonic lesions incidence in single tumor bearing mice was 11 +/- 4.0 in those treated with DMH and 8.1 +/- 3.7 in those treated with endostatin. When we look at the distribution of lesions in the colon, they occurred in the distal colon. At the end of our study, we noticed that the number of lesions decreased by 25% in the group of endostatin, considering the number of the lesions in the group of DMH. But there was no statistical difference between the mice treated with endostatin and those treated with DMH. It will be very significant to identify endostatin therapeutic effects as long as proper dose of endostatin is administrated at the proper time, duration and proper tumor model.