The aim of this longitudinal study was to explore temporal patterns of physical activity (PA) and adiposity in Spanish adolescents. Eighty healthy adolescents were followed over a 2-year period (42 girls and 38 boys). A PA score was estimated using the Physical Activity Questionnaire for Adolescents (PAQ-A). Adiposity was assessed by anthropometric measurements; body mass index (BMI) and fat mass percent (FMP) were calculated using standard equations. Sexual maturity was estimated by percentage of predicted adult stature. Dietary intake was assessed by a self-administered food-frequency questionnaire. Three assessments were performed: September 2011 (S1), 2012 (S2), and 2013 (S3). A repeated-measures ANOVA was conducted to examine temporal changes in PA and FMP and sex change in maturation categories (two factor mixed-design, 2x2x3). A stepwise linear regression was conducted in order to estimate the predictors of FMP change. Significant changes for FMP were found between S1, S2, and S3 (23.41±8.24 vs. 21.89±7.82 vs. 22.05±8.06, p<0.05; respectively); a significant interaction of FMP with sex was observed (F = 4.387, p<0.05 for S2-S3), but not for maturation. PA at S2 was significantly higher than S3 (2.58±0.72 vs. 2.29±0.73, p<0.001). An interaction between PA change and sex was statically significant (F = 4.889, p<0.05 for S2-S3). A reduction in PA was observed after the S2 period without changes in adiposity. In contrast, a significant reduction in FMP was seen between S1 and S2, while PA did not significantly change. There were no significant differences for nutritional variables between S1 and S3, and nutrition was not a determinant of the changes in PA or FMP. Our results suggest that body composition changes observed during adolescence are not driven by changes in PA. Moreover, the interaction analysis suggests that PA behavior is affected by sex, but is not modified by maturation.