ACCURACY TESTS OF TOOTH FORMATION AGE ESTIMATIONS FOR HUMAN SKELETAL REMAINS

Estimations of age from tooth formation standards for a large (n = 282 ) sample of subadult skeletal remains from a 19th century historic cem etery sample were analyzed. The standards of Moorrees et al. (1963a,b) for the permanent and deciduous teeth, and Anderson et al. (1976) for the formation of permanent dentition were employed in a variety of co mbinations to calculate mean dental ages. Tests of accuracy and bias w ere made on a small sample (n = 17) of personally identified individua ls, and age of attainment scores were compared to age of prediction sc ores for each individual. The resulting dental age distributions for t he skeletal sample were compared to documented burial records for the cemetery to determine the representativeness of the skeletal sample. T hese comparisons showed little difference between age of attainment ve rsus age of prediction methodologies. The standards of Moorrees et al. (1963a,b) were observed to provide the most accurate estimates of age with a standard deviation of one-half year. The standards of Anderson et al. (1976), while easier to use and more extensive, are problemati c in that the original reference sample begins at three years of age, while the sample used by Moorrees and colleagues begins at birth. The skeletal age distributions compare well to the overall chronological a ge distribution for the cemetery. These results affirm that tooth form ation age estimates for subadult skeletal remains from archaeological or forensic samples provide accurate assessments of age at both the in dividual and population level. (C) 1993 Wiley-Liss, Inc.