UNLABELLED: The object of the present study was to establish the relationship between the size an body segments in Colombian adult subjects, to get an equation that allows to estimate the size from those segments and confront it with those described in the literature. SUBJECTS AND METHODS: Group 1: Subjects over 18 years were chosen without deformities, amputations neither edemas. Age, sex, weight, size, forearm length (Fl), lower segment length (LSL), navel-heel length (NHL), navel-malleolus length (NML), lower member length (LML), knee height (KH), knee-external malleolus length (KML). Group 2: The same variables were quantified and to these subjects were established the accuracy and precision of the obtained equations and were confronted with those described in the literature. Statistical analysis. The gotten results were analyzed by means of the Student "t" test, correlation analysis, and simple linear regression and multiple step by step. An alpha error of 5% was accepted. RESULTS: 189 subjects were evaluated in group 1 and 180 in group 2. 23 male and 156 female 37.5 +/- 17.5 years old and a size of 161.2 cm +/- 9.6 cm. There was a significant correlation (p < 0.0001) between size and all the measured segments. Predictive size regression equations were established choosing the following segments: FL, NML, KML, because of their easy determination, precise reference points and high "t". The estimated size was not significantly different from the real one just in those that use KML, age an sex: Male: Size = (KML x 1.21) - (0.117 x age) + 119.6. Female: Size = (KML x 1.263) - (0.159 x age) + 107.7. The equations described by Chumlea et al. using the KH just predict the size adequately in male, but overestimate it in female. CONCLUSIONS: 1. The equations using the KML predict the size in the colombian studied sample. 2. The Chumlea equations only adequately predict the size of the male colombian subjects.