In July 2005, cases of hepatitis were reported from three villages in Nainital district, Uttarakhand, India. We investigated this cluster to identify the source and propose recommendations. A door-to-door search for cases of acute hepatitis was carried out in the three villages. We described the outbreak by time, place and person and conducted a cohort study to identify the source of infection. In addition, sera from cases were tested. We identified 205 cases among 1238 persons (attack rate: 16%, no deaths) between May and September 2005. Of the 23 sera tested, 21 were positive for IgM antibodies against hepatitis E virus. The attack rate was highest among 15-44 years old (19%). Cases began on May 3, 2005, peaked in July and decreased rapidly. The incidence was highest (23%) in one of the villages predominantly using water from an unprotected spring, which was distributed after stone bed filtration alone. In this village, the attack rate increased from 9% among those not using the spring, to 13.8% among those partly using it (RR [95% CI] 1.6 [0.8-3.4]), and to 29% among those exclusively using it (RR [95% CI] 3.4 [2.0-6.0]). Untreated drinking water from an unprotected spring may have been the source of this outbreak in a rural area. Sources of water supply must be protected and treated, including with chlorination. Reporting and investigation of smaller outbreaks in rural areas should be improved.