The fluorescence decay of cod parvalbumin (both its Ca2+-loaded and Ca2+-depleted forms) is found to be a nonexponential process. The decay data can be fitted either by a double-exponential decay law or by a distribution of decay times. To try to distinguish between the double-exponential and distribution fits, we have collected frequency domain and steady-state fluorescence data as a function of temperature and concentration of the quencher acrylamide. We argue that the correct decay law (i.e., double exponential or distribution) must be consistent with all the data collected as a function of temperature and quencher concentrations. We employ a global analysis procedure to simultaneously fit multiple data sets that are linked by an Arrhenius or Stern-Volmer relationship. For the Ca2+-loaded form of parvalbumin, the distribution model provides a consistent and reasonable fit for all of the frequency domain and steady-state data. The double-exponential model requires more fitting parameters, and some of these assume unreasonable values when this model is fitted to all of the data. For the Ca2+-depleted form of the protein, it is not clear whether the double-exponential or distribution model is superior. For our steady-state solute quenching studies we present a novel analysis in terms of a distribution of quenching constants.