We introduce (ε,δ)-bisimulation, a novel type of approximate probabilistic bisimulation for continuous-time Markov chains. In contrast to related notions, (ε,δ)-bisimulation allows the use of different tolerances for the transition probabilities (ε, additive) and total exit rates (δ, multiplicative) of states. Fundamental properties of the notion, as well as bounds on the absolute difference of time- and reward-bounded reachability probabilities for (ε,δ)-bisimilar states, are established.