We counted droppings of ungulates and hare on transects in order to assess (1) seasonal changes in detectability and disappearance of pellet groups, (2) whether the detectability varies according to the forest type, and (3) the degree of misidentification between pellets of roe deer (Capreolus capreolus) and red deer (Cervus elaphus). The summer decrease in detectability of pellet groups was the most important factor for all species but European bison (Bison bonasus). Detectability did not significantly depend on forest type. In summer, decay reduced significantly the dropping density of red deer, roe deer, wild boar (Sus scrofa) and European hare (Lepus europaeus) but not those of bison and moose (Alces alces). Misidentification of roe and red deer droppings did not influence much density estimates of red deer but resulted in an important overestimation of roe deer in areas were roe deer were much less common than red deer.