Evaluation of the eye lens method for age determination in roe deerCapreolus capreolus

Eye lens weight curves were obtained from a sample of 522 male and 171 female free-living roe deerCapreolus capreolus (Linnaeus, 1758), collected during the regular annual harvest 1998/99–2005/06 (May–January) in 12 different field hunting grounds in Vojvodina (North part of Serbia). We found no significant differences in eye lens weight between males and females within the same age class (except age class 1). Thus, the eye lens weight data for both sexes within the same age class were pooled for further analyses. The rate of increase in eye lens weight declined at a relatively early age, but lenses continued to increase in weight through age class 9. However, there was a considerable overlap in ranges of eye lens weights within all age classes, which restricted the usefulness of this method. The estimate precision of the eye lens method in determining age was evaluated by the comparison of “known age” (cementum layer technique) and predicted age (eye lens weight technique). We examined model validity on 78 roe deer not used previously to construct the regression line. The lack of correspondence between the two techniques was evident among individuals within older age classes. The percentage of correctly determined age rapidly declined already from age class 4 (<30%).