The aim of our study was to compare plicometry and bioelectrical impedance analysis as methods for body composition assessment (fat mass-FM%, fat-free mass-FFM) and to extrapolate population-specific equations in order to predict the plicometrically-derived FM and FFM on the basis of body resistance. A cross-sectional study was carried out at the Diabetes Out-Patient Clinic, Department of Internal Medicine, University of Palermo, Italy. We examined 102 subjects (45 men, 57 women, age range 30-45 years) divided into three groups: Group N (36 normal weight subjects, body mass index-BMI = weight/height2 = 24.5 +/- 0.8), Group O (35 obese subjects, BMI = 35.1 +/- 0.4), and Group GO (morbid obese subjects, BMI = 49.2 +/- 1.2) as well as a control group of 30 subjects (12 men, 18 women, age range 30-45 years, BMI = 20.9-63.1). We measured the following parameters: bioelectrical impedance analysis (body resistance: R), plicometry (triceps, biceps, subscapular, and suprailiac skinfolds), height (Ht), weight (W). The two methods were comparable in Groups N and O. In Group GO, however, especially in men and for FM%, measurements were significantly different. Plicometrically-determined FM% and FFM-Kg were significantly correlated to R.W/Ht2 (r = 0.821, p < 0.0001) and to Ht2/R (r = 0.867, p < 0.0001) respectively. The following prediction formulas for plicometric FM and FFM from body resistance and anthropometric variables were obtained: 1) FM% = 22.3 (R.W/Ht2)-2.2, and 2) FFM-Kg = 1.1 (Ht2/R)-3.1.(ABSTRACT TRUNCATED AT 250 WORDS)