Objectives: The objective of this study is to examine the relationship between age and depression among people aged 65 and older. Method: The study uses three waves of longitudinal data (1991, 1996, 2001) from a community and institutional sample of Canadians aged 65 and older. The study uses generalized linear mixed-model techniques to estimate the trajectories of depressive symptoms and major depression in late life. Results: There is a linear increase in depressive symptoms after age 65, but this occurs in the context of medical comorbidity and is not an independent effect of aging. There is a significant u-shaped relationship between age and major depression, after adjusting for selected covariates. Discussion: The relationship between age and late-life depression is complex, and it depends on how the dependent variable is measured. Late-life depression develops through a different set of risk factors than it does in earlier stages of the life course. The “fourth age” appears to be a period of psychiatric morbidity.