The magnitude of the proton gradient (delta mu H+) driving solute accumulation in Saccharomyces cerevisiae has long been in doubt, principally because of the lack of an agreed method for assaying its electrical component, the membrane potential (delta psi). In the present work, the size of the cytosine gradient (delta mu cyt) that the yeast generated was used as a measure of the driving gradient (delta mu H+). The selected yeast lacked cytosine deaminase and overexpressed cytosine permease, a 1 H+/cytosine system. delta mu cyt, assayed in washed cell suspensions fermenting glucose and containing 0.5 or 50 mM KCl, was about 260 mV at pH 4 or 5, falling to about 194 mV at pH 7. As a first estimate, -delta mu H+ was thus at least as large at the respective pH value. A 20 mM solution of the lipophilic cation tetraphenylphosphonium lowered delta mu cyt to a value roughly equal to the magnitude of the pH gradient (delta pH). A mathematical model was used to correct the first estimates of delta mu H+ for the effect of cytosine leakage outside the symport. In such a system, delta mu cyt cannot exceed the equivalent ratio Vmax/KmL, where Vmax and Km are kinetic parameters of the symport and L is the rate coefficient for leakage. The feasibility of assaying delta mu H+ depends on it not being much larger than that ratio. The model was tested successfully against observations made with yeast preparations depleted of ATP. After correction, -delta mu H+ during fermentation was estimated to be up to 25 mV larger than delta mu cyt and at least 70 mV larger than previous estimates in the literature involving lipophilic cations. From a knowledge of delta pH, delta psi was in turn deduced and compared with the maximum methylamine gradient (delta mu M) the yeast formed. The results supported the claim in the literature that, at acid pH, delta mu M is a measure of delta psi.