When a solution containing a single monobasic weak acid is titrated potentiometrically with a strong base, non-linear regression analysis of the data yields a random scatter of the residuals around the best fit to the appropriate equation. If the same equation is used when a second weak acid is also present, systematic errors arise and yield a deviation plot having a characteristic shape. If the amplitude of that plot is substantially larger than the random error of measurement, the presence of the second acid can be detected, and its concentration can then be evaluated by non-linear regression onto the equation that takes its presence into account. The amplitude increases as the relative concentration of the second acid increases, and as the difference between the values of pKafor the two acids increases. Examination of these dependences shows, for example, that it should be possible to detect 1 per cent of an impurity for which pKadiffers by 0.57 unit from that of the major component.