Models of stratospheric photochemistry are generally tested by comparing their predictions for the composition of the present atmosphere with measurements of species concentrations. These models are then used to make predictions of the atmospheric sensitivity to perturbations. Here the problem of the sensitivity of such a model to chlorine perturbations ranging from the present influx of chlorine-containing compounds to several times that influx is addressed. The effects of uncertainties in input parameters, including reaction rate coefficients, cross sections, solar fluxes, and boundary conditions, are evaluated using a Monte Carlo method in which the values of the input parameters are randomly selected. Results are probability distributions for present atmospheric concentrations and for calculated perturbations due to chlorine from fluorocarbons. For more than 300 Monte Carlo runs the calculated ozone perturbation for continued emission of fluorocarbons at today's rates had a mean value of -6.2%, with a 1-sigma width of 5.5%. Using the same runs but only allowing the cases in which the calculated present atmosphere values of NO, NO2, and CLO at 25 km altitude fell within the range of measurements yielded a mean ozone depletion of -3%, with a 1-sigma deviation of 2.2%. The model showed a nonlinear behavior as a function of added fluorocarbons. The mean of the Monte Carlo runs was less nonlinear than the model run using mean values of the input parameters.