The two pulse fluorescence bleaching experiments reported in an earlier paper [J. Chem. Phys. 117, 454 (2002)] are reanalyzed in the context of additional experiments. The fluorescence bleaching observed in that paper was originally ascribed to ground-state depletion. By analyzing the absorption saturation behavior and the magnitude of the pump-probe signal, we find that the absorption cross section of poly(phenylene vinylene) at 400 nm is two orders of magnitude too small to significantly deplete the ground state given the pump fluences used in that experiment. Instead, the observed depletion is due to a combination of exciton-exciton annihilation at early times and dark state luminescence quenching at later times. Different experiments have different sensitivities to the fluence-dependent quenching, with time-resolved experiments like transient emission and fluorescence decay time consistently underestimating the exciton-exciton annihilation rate. Experiments that measure the integrated fluorescence, such as fluorescence saturation and bleaching, result in a consistent value for the exciton-exciton annihilation constant of 6x10(-9) cm(3)/s, while the dark state quenching constant is estimated to be at least 1.2x10(-8) cm(3)/s. Indirect evidence based on the wavelength and sample dependence of the dark state formation suggest that the dark states are charge-separated polarons. The relatively large quenching constants are consistent with what has been observed in other conjugated polymers and suggest that the long-lived dark states in particular are effective quenchers, capable of quenching thousands of surrounding chromophores. (C) 2003 American Institute of Physics.