Program slice has many applications such as program debugging, testing, maintenance, and complexity measurement. A static slice consists of all statements in programP that may effect the value of variablev at some pointp, and a dynamic slice consists only of statements that influence the value of variable occurrence for specific program inputs. In this paper, we concern the problem of dynamic slicing of object-oriented programs which, to our knowledge, has not been addressed in the literatures. To solve this problem, we present the dynamic object-oriented dependence graph (DODG) which is an arc-classified digraph to explicitly represent various dynamic dependence between statement instances for a particular execution of an object-oriented program. Based on the DODG, we present a two-phase backward algorithm for computing a dynamic slice of an object-oriented program.