Tests of estimation of age from tooth wear on roe deer of known age: variation within and among populations

The rate of wear of ruminant teeth may vary according to diet, habitat, and individual tooth characteristics. This variation may cause error in estima ting the age of wild ungulates from patterns of tooth wear. We tested the a bility of 10 observers to accurately estimate age from observation of tooth wear in a large sample of jaws of known-age roe deer (Capreolus capreolas) from three populations.. Although the average error was not large (+/-1.02 years), maximum error ranged from -5 to +6 years for jaws of animals betwe en 1 and 7 years old, with observers generally overestimating the age of yo ung animals and underestimating the age of old animals. We found significan t differences among observers in estimation error. When a "jaw-board" of kn own-age reference specimens was provided, this observer effect was largely controlled for, but accuracy was not improved. Error was partly due to vari ation in tooth wear, both within and among populations. Initial cusp height of the first molar was lower, but tooth wear tended to be slower in one po pulation than in the other two populations,possibly reflecting differences in diet and (or) habitat. Individual variation in tooth wear within populat ions was observed, possibly reflecting variation in tooth characteristics ( e.g., enamel mineralisation), which was a source of error in age estimation from observation. Observers tended to underestimate the age of jaws with a relatively low degree of wear and vice versa. These results show that esti mating the age of roe deer from observation of tooth wear produces biased r esults, severely limiting its application in population studies of this spe cies.