The purpose of this study was to determine the effects of exercise habituation on body composition and anthropometric characteristics in cardiac patients. The subjects, comprising 20 patients with coronary heart disease, aged 43-69, participated in our supervised exercise programme for 38.0 (SD 12.5) weeks while in hospital. The intensity of most exercise was set at the lactate threshold. Analyses of the data indicated that small but significant reductions were observed in body mass (mb) [-1.4 (SD 1.8) kg], abdominal girth [-1.4 (SD 2.6) cm], chest girth [-1.3 (SD 1.8) cm], body mass index [-0.6 (SD 0.7)], and skinfold thicknesses at all the sites measured. As expected, fat-free mass, derived from either bio-electrical impedance (BI) or a skinfold technique with commonly used regression equations, remained essentially unchanged, while there were significant decreases in body fat (BF) and %BF. The absolute amount of change in mb (i.e. delta mb) was significantly associated with delta abdominal girth (r = 0.506), delta BF estimated by the BI technique (r = 0.476), and delta BF estimated by the skinfold technique (r = 0.451). Although the period of the exercise programme [38.0 (SD 12.5) weeks] varied greatly among subjects, it was found not to be associated with delta mb and alterations in body composition. We concluded that aerobic exercise induced significant decreases in BF and many anthropometric variables, independent of the duration of exercise, and that a combination of abdominal girth, BI and/or skinfold measurements would be advantageous in estimating primarily exercise-induced alterations in BF in cardiac patients.