The aim of this study was to evaluate sexual dimorphism in male (n = 43) and female (n = 39) springboard and platform divers at the 1991 World Diving Championships in Perth, Western Australia. In addition to recording the age, height and body mass of each diver, measurements were taken of 8 skinfolds, 13 girths, 8 breadths and depths, and 11 lengths. The derived variables were somatotype, sum of 6 skinfolds (SUM6) and height-adjusted proportional z-scores. The mean (+/- s) values for the males were: age, 22.2 +/- 4.6 years; height, 171 +/- 9 cm; body mass, 66.7 +/- 10.2 kg; somatotype, 2.0-5.3-2.4 +/- 0.5-1.0-0.8; SUM6, 45.9 +/- 11.4 mm. For the females, these values were: age, 20.9 +/- 3.8 years; height, 161 +/- 6 cm; body mass, 53.7 +/- 5.5 kg somatotype, 2.8-3.8-2.8 +/- 0.7-1.0-0.9; SUM6, 65.6 +/- 17.0 mm. The males were taller, heavier, more mesomorphic, less endomorphic and had a smaller SUM6 than the females (P <0.01). A series of discriminant function analyses were applied to groups of variables. Wilks' lambda and the overall. F-ratio were highly significant (P <0.001), with correct classification of sex reaching 84-99%. The best combinations of variables were as follows: (1) absolute size, 3 girths + SUM6, or body mass + SUM6 + height (Wilks' lambda = 0.17-0.35); (2) relative size, 2 girths + SUM6 + body mass. Combinations of breadths and lengths, skinfolds, or somatotype components, were less successful, with 84-87% classified correctly (Wilks' lambda = 0.40-0.48). We conclude that, in addition to the expected absolute size differences, sexual dimorphism between male and female divers can be explained by differences in relative size, skinfolds and somatotype, and that these differences may have implications for coaching and selection in diving.