Objectives: To examine the long-term effects of midlife overweight for cognitive abilities. The evidence is growing strong that overweight in midlife is related to increased dementia risk. Few studies have addressed the question if overweight affects cognitive abilities among those who do not develop dementia. In two studies we examined cognitive performance in two cohorts of people (young-old and old) in relation to self-reported Body Mass Index (BMI) in midlife. Methods: The participants are from the Swedish Twin Registry who participated in longitudinal studies on aging and cognition, the SATSA study (young-old cohort, 50 years and older) and the OCTO-Twin study (old cohort, 80 years and older) . BMI was reported in 1963 and cognitive abilities were examined 20- to 30-years later with five measurement occasions at 3-year intervals (SATSA) respectively 2-year intervals (OCTO-Twin). The cognitive abilities examined included tests of long-term memory, short-term memory, speed, verbal ability, spatial ability and a composite score representing general cognitive ability. Results: Multilevel modeling adjusting for twinship, demographic factors, cardiovascular diseases, and diabetes, showed that higher BMI in midlife predicted lower test performance 30 years later. Significant associations were found in all cognitive abilities. Although we found a significant cognitive decline across the five measurement occassions in both cohorts at the follow-up assessments, a higher midlife BMI was not associated with steeper decline in the old cohort, with the exception of verbal ability. This was, however, found for a measure of general cognitive ability and spatial ability in the young-old cohort.Conclusions: Our results indicate that midlife overweight is related to lower overall cognitive function in old age. The fact that BMI-related effects in slopes were only noted in some abilities in the young-old cohort suggests that the negative effect of overweight has an early onset.