Purpose: According to the "reserve perspective" on cognitive aging, individuals are born with or can develop resources that help them resist normal and disease-related cognitive changes that occur in aging. The reserve perspective is becoming more sophisticated, but gaps in knowledge persist. In the present research, we considered three understudied questions about reserve: Is reserve primarily static (unchangeable) throughout the life course or dynamic (changeable, in terms of increases or decreases)? Can reserve be increased at any point in life, or are there optimal time periods--such as early life, midlife, or late life--to increase it? Does participation in different types of leisure and occupational activities in early life and midlife have different effects depending on specific domains of late-life cognitive functioning? Here we link early cognitive and activity data--gathered from archival sources--with cognitive data from older adults to examine these issues. Design and Methods: 349 participants, all mid-1940s graduates of the same high school, underwent telephone cognitive screening. All participants provided access to adolescent IQ scores; we determined activity levels from yearbooks. We used path analysis to evaluate the complex relationships between early life, midlife, and late-life variables. Results: Adolescent IQ had strong direct effects on global cognitive functioning, episodic memory, verbal fluency, and processing speed. Participants' high school mental activities had direct effects on verbal fluency, but physical and social activities did not predict any cognitive measure. Education had direct effects on global cognitive functioning, episodic memory, and, most strongly, processing speed, but other midlife factors (notably, occupational demands) were not significant predictors of late-life cognition. There were weak indirect effects of adolescent IQ on global cognitive functioning, episodic memory, and processing speed, working through high school mental activities and education. Verbal fluency, in contrast, was affected by adolescent IQ through links with high school mental activities, but not education. Implications: Our study suggests that reserve is dynamic, but it is most amenable to change in early life. We conclude that an active, engaged lifestyle, emphasizing mental activity and educational pursuits in early life, can have a positive impact on cognitive functioning in late life. (Contains 4 tables and 2 figures.)