Stature reconstruction from long bones in ancient population samples: an approach to the problem of its reliability.

Stature estimates provided by the application of the anatomical method of Fully and Pineau (1960) to well preserved Neolithic skeletons (39 males and 27 females) from seven different European countries are compared with those drawn from lower-extremity components by means of regression equations commonly used to predict stature of earlier European populations. The analysis of data, carried out with reference both to the sample of origin of the skeletons and to stature classes, suggests that the equations of Pearson (1899) and of Trotter and Gleser for Negroes (1952, 1977) yield very good estimates in female samples, leading to errors below 2 cm in most of the cases. These equations, and those of Olivier et al. (1978), prove useful for stature reconstruction in males too, with the exception of very low (below 154 cm) and very tall (over 179 cm) individuals. Formulae of Breitinger (1938) yield values consistent with those resulting from the anatomical method only within a range including medium-high statures, while the corresponding Bach (1965) formulae for females provide poor approximations. The Trotter and Gleser formulae for Whites (1952) yield very unsatisfactory evaluations, except in specimens taller than 180 cm, and usually overestimate stature both in male and in female skeletons. The results obtained from the long bones by means of alternative approaches to the least-squares regression formulae (Model II regression, and femur/stature ratio) and their efficacy in predicting stature of the samples under study is discussed in the final part of the paper.