Purpose: The aim of this study was to investigate the tracking of television viewing (TV) time as an indicator of sedentary behavior among adults for a period of 25 yr. Methods: A random sample of 1601 subjects (740 men) age 18, 21, and 24 yr participated in the Cardiovascular Risk in Young Finns Study in 1986. TV time during leisure time was measured with a single self-report question at baseline and in 2001, 2007, and 2011. Tracking of TV time was analyzed using Spearman rank correlations and simplex models. Level and change of TV time were examined using linear growth modeling. Results: The 4- and 6-yr integrated TV time stability coefficients, adjusted for measurement errors, were Q0.60 in adulthood and quite similar for both men and women. The stability coefficients tended to decline as the time interval increased. The stability of the indirect estimation of TV time for a 25-yr period was moderately or highly significant for both genders in most age groups. Younger age, but not gender, was found to be associated with a higher initial level of TV time. Male gender and older age were found to be significantly associated with the slope of TV time. Conclusion: The stability of TV time is predominantly moderate to high during adulthood and varies somewhat by age and gender.