This article reports the findings of a 3-year epidemiological survey for dementia in an urban community-resident population in Mumbai (Bombay), India, wherein the prevalence of all types of dementia was determined. Method: The study was conducted in three stages. Stage 1: From a potential pool of 30,000 subjects aged 40 years or more, 24,488 (male = 11,875; female = 12,613) persons completed self-report or interviewer-rated protocols based on the Sandoz Clinical Assessment Geriatric Scale, but 5,512 (18.37%) persons refused to participate. Scores on the protocol had a possible range from 0 through 34. Stage 2: Persons with a score +2SD above the mean were selected in this stage where the persons were screened for cognitive functioning using a modified and translated version of the Mini-Mental State Examination. Individuals who scored below the 5th percentile were included in Stage 3 and underwent a detailed neurological, psychiatric, and neuropsychological evaluation as well as hematological, radiological, electrocardiographic, and electroencephalographic investigations. Diagnoses were made jointly by a neurologist, psychiatrist, and psychologist using the DSM-IV diagnostic criteria. Subjects were also rated on the Clinical Dementia Rating (CDR) scale and assessed for activities of daily living. Results: One hundred five subjects with dementia (CDR ≥ 0.5) were identified in this population of 24,488 persons. The prevalence rate for dementia in those aged 40 years and more was 0.43% and for persons aged 65 and above was 2.44%. Seventy-eight individuals had a CDR of ≥ 1 yielding an overall prevalence rate of 0.32%, and a prevalence rate of 1.81% for those aged 65 years and older. The overall prevalence rate for Alzheimer's disease (AD) in the population was 0.25%, and 1.5% for those aged 65 years and above. AD (n = 62;65%) was the most common cause of dementia followed by vascular dementia (n = 23;22%). There were more women (n = 38) than men (n = 24) in the AD group. Increasing age was associated with a higher prevalence of the dementia syndrome in general as well as AD specifically. Conclusion: In the population surveyed, the prevalence of AD and other dementias is less than that reported from developed countries but similar to results of other studies in India. Prevalence of the dementia syndrome increased with age and was not related to gender. AD was the most common dementia and the prevalence was higher in women than in men. Results are discussed with respect to shorter life expectancy, relocation of affected persons, and differences in the risk factors as compared to developed countries.