The abundant benefits of living a physically active life are well documented (Grundy et al., 1999; Westerterp, 1999; Epstein & Goldfield, 1999; Blair & Brodney, 1999; USPHS, 1996). Of particular interest is the relationship between physical activity and obesity. Americans are becoming obese at an epidemic rate (Mokdad, 1999; Must et al., 1999; Blair & Bouchard, 1999). To battle this epidemic, physical activity levels of children must be increased through physical activity intervention programs. Much of the research in this area has focused on increasing physical activity levels during physical education classes. However, Ernst and Pangrazi (1998) focused on involving the classroom teacher in implementing an activity intervention and indicated that classroom teachers successfully promoted increased activity levels in their students. The obvious difficulty with this type of intervention is that time must be used to implement such an intervention, and within today’s educational climate, many schools are unwilling to give up time devoted to “academic” pursuits. Aside from required physical education class time, recess seems to be time during the school day when children can participate in physical activity. However, increasing activity is not guaranteed during scheduled recess. There are only a few studies that have examined physical activity at recess, and fewer that have examined physical activity levels related to health at recess (Hovell, Brusick, Sharkey, & McClure, 1978; Kraft, 1989). The purpose of this study was to examine the effects of an in-school physical activity intervention on fourth and fifth grade students’ physical activity levels and compare intervention activity to activity gained at recess. One fifth and one fourth grade teacher and their students participated in a four-week physical activity intervention including structured daily 15-minute activity breaks. Twelve students (3 girls and 3 boys from each class) were randomly selected and coded using the System for Observing Fitness Instruction Time (SOFIT) to determine activity levels. Each student was coded a total of four times during the study, twice during regularly scheduled recess and twice during the intervention. Paired T-tests were used to determine significance in each of the SOFIT activity categories (sitting, standing, walking, very active). Results indicated that students spent significantly more time sitting and standing during recess t=5.40, p<.00, and significantly more time in activity during the structured activity intervention t=5.28, p<.00. These findings can help researchers and practitioners design appropriate and practical methods for increasing children's activity levels during the school day.