Previous investigations have suggested that osteosarcoma may be associated with a taller stature, but the relationship between height and osteosarcoma remains controversial. Height at diagnosis was evaluated in a continuous series of 962 osteosarcoma subjects treated between 1981 and 2001. Patients diagnosed during growth (group 1) were separated from those diagnosed in adulthood (group 2). Height (H) and final height (FH) were expressed as standard deviation scores (SDS), calculated by national reference data. Group 1 subjects were above the 50th centile and their mean H-SDS values (0.31 ± 1.1) were significantly higher than the mean FH-SDS values (P < 0.0001) of the group 2 subjects, both in males and females. In contrast, the mean FH-SDS (0.01 ± 1.1) of group 2 did not differ from that of the reference population. The highest incidence of osteosarcoma was at 12.5 years in females, 14.5 years in males. These data confirm previous observations of an association between osteosarcoma development and height, at least in growing individuals. The higher incidence during the pubertal spurt, in the anatomic sites of greater growth and in taller individuals, suggests that growth factors play an important role in the pathogenesis of this bone cancer.