We discuss the atmosphere variability and instrumental uncertainties which limit the determinaton of the atmospheric lifetime of fluorocarbons. These factors produce substantial uncertainties in the current estimates of fluorocarbon lifetimes. However, because fluorocarbon production has remained relatively constant over the past few years, the atmospheric lifetime may now be determined with considerably improved accuracy. Analysis of the data by trend and total overburden techniques are compared: the trend method is to be preferred after the first year or two because it allows systematic errors to be reduced. The detectability of a sink is illustrated for a network of four stations spaced around the globe at which measurements are made at least daily. We conclude that 5 years of operation of such a network should enable us to establish whether atmospheric lifetimes of 20 years or less exist.