Outgassing rates from the walls of a vacuum chamber are usually measured by one of two methods, (1) the throughput method, where the outgassing rate is derived from the pressure drop across a known conductance during evacuation, or (2) the pressure rise method (aka the gas accumulation method) where the chamber is sealed off at a low pressure and the rise in pressure with time measured. A model of the throughput method is analyzed to show that readsorption of desorbed gas results in the measured outgassing rate Q(m) being considerably less than the intrinsic rate Q(i) (the outgassing rate in the absence of readsorption). The ratio of outgassing rates Q(m)/Q(i) varies with the pumping parameter SIA (S is the pumping speed and A the adsorbing area) and the sticking probability s. Q(m) is found to vary with the pumping parameter and the sticking probability at zero coverage s(0), except when a/As-0 much less than kt(-1), where a is the effective area of the orifice, A the adsorbing area, k is a constant involving the heats of adsorption, and t the pumping time. Measurements by the pressure-rise method on stainless steel systems show that the gas evolved is hydrogen, and that the pressure rise is linear for as long as 10(3) h with the pressure increasing by a factor as large as 10(6), indicating a constant outgassing rate. It is shown that if readsorption occurs then the p(t) curve will be nonlinear. A model is proposed to explain the observed constancy of outgassing rate and the apparent absence of readsorption which assumes that (a) the outgassing rate is desorption limited and (b) the rate of diffusion of H atoms to the surface is much greater than the rate of desorption of H-2 or the rate of arrival of molecules from the gas phase at empty sites.