Under unitary evolution, systems move gradually from state to state. An unstable atom has amplitude in its original state after many lifetimes (τL). But in the laboratory, transitions seem to go instantaneously, as suggested by the term "quantum jump." The problem studied here is whether the "jump" can be assigned a duration, in theory and in experiment. Two characteristic times are defined, jump time (τJ) and passage time (τP). Both use Zeno time, τZ, defined in terms of H and its initial state as τZ≡ℏ/, with $E_ψ\equiv $. τJ is defined in terms of the time needed to slow (à la the quantum Zeno effect) the decay: τJ≡τZ2/τL. It appears in several contexts. It is related to tunneling time in barrier penetration. Its inverse is the bandwidth of the Hamiltonian, in a time-energy uncertainty principle. τJ is also an indicator of the duration of the quadratic decay regime in both experiment and in numerical calculations (cf. Fig.~2 of PRA 57,1509 (1998).) The passage time, τP, arises from unitary evolution sans interpretation. It is based on a bound of Fleming (Nuov. Cim. 16 A, 232 (1973)): for any H and ψ a system cannot evolve to a state orthogonal to ψ for t<τP≡πτZ/2. By including apparatus in H, τP limits the observation of decay according to the quantum measurement ideas proposed in "Time's Arrows and Quantum Measurement," Cambridge U. Press, 1997, thereby allowing an experimental test of these ideas.