The relation ship of cause-result between low selen ium (Se) and Kashin-Beck disease (KBD) was probed by the prospective study of epidemiolog icalmethod with regarding low-Se as an exposure factor in this paper. 597 healthy children l ived in KBD areas with low, middle and high prevalence were divided in to the low-Se exposed group and the non-low-Se exposed group according to their Secontent in hair. The low-Se exposed group was divided in to three subgroups, such as Secontent in hair ≤ 110ng/g, 110ng/g<Secontent in hair≤150ng/g and 150ng/g<Secontent in hair ≤ 200ng/g, respectively. Six new cases of the total with KBD (incidence was 01574% person-year) were found in the low-Se exposed group during three years period of the investigation. No new case was found in the non low-Se exposed group. KBD incidence was notsign if icantly different between those two groups. Two new cases were found in children with Secontent in hair keptbelow 110ng/g during three years (inc idence: 1121% person-year). SMR in each group indicated that the new cases observed in the low-Se exposed group was remarkable lower than the new cases expected. It was not observed that the dose-respon serelation ship between low-Se and KBD, and was not supported that the low-Se was a predominant factor to cause KBD.