The aim of this study is to quantify the genetic and environmental variation in biological parameters of the adolescent growth in stature in both sexes, using structural equation models. Standing height was measured in 99 twin pairs from the Leuven Longitudinal Twin Study. Subjects were seen at semi-annual intervals between 10 and 16 years and at 18 years. Preece-Baines model I was fitted to the individual data and biologically meaningful parameters were derived. Path models were fitted to these parameters and a gender heterogeneity analysis was performed. A model including additive genetic and specific environmental factors (AE-model), allowing for a difference in total variance or in genetic/environmental variance components in males and females best explains most of the growth characteristics. For the timing and velocity of the adolescent growth spurt, no gender heterogeneity was observed, and the genetic (0.89 to 0.93) and specific environmental (0.07 to 0.11) contributions were equal in both sexes. For stature at take-off, stature at peak height velocity and for adult stature, gender heterogeneity was observed and different models appeared to result in the best fit for boys and girls. For height at PHV and adult stature, the significant contribution of a common environmental (0.39 to 0.56) factor in boys was noteworthy. For percentage of adult stature and distances between relevant time points of the growth curve, a non-scalar effect was observed, indicating that different genes affect these characteristics in boys and girls. It is concluded that most characteristics of the growth spurt in stature are under strong genetic control. Gender heterogeneity is present for stature and for a number of derived growth characteristics.