In a small mountain stream (Erlenbach, Alptal, central Switzerland), water samples were taken during 15 years by two different methods: weekly-bulked, discharge-proportional sampling and fortnightly grab (instantaneous) sampling. Compared with continuous, in-stream measurements of the electrical conductivity, the discharge-proportional sampling proved to give very accurate and precise estimates of solute fluxes. Relatively to discharge-proportional samples, integration methods based on the grab samples resulted in more or less biased fluxes of: Ca, Mg, K, Na , Cl−, and electrical conductivity. Unweighted means were all positively biased due to concentrations decreasing with the discharge. Means weighted by discharges averaged fortnightly were less positively biased. On the other hand, weighing by the instantaneous discharge gave negative biases. Fluxes calculated from regressions of concentrations against discharge were also negatively biased, but the accuracy could partly be improved by including further factors into the regressions, especially the seasonality. If both grab and discharge-proportional samples are taken in parallel for some time, it is possible to calibrate the fluxes calculated from the first against the second. Calibration periods of 1–7 years were tested and found to improve the accuracy in 8 other years used for validation. Five years were sufficient to achieve unbiased regression estimates and a precision better than 10%. Replacing a less accurate grab sampling scheme by a discharge-proportional sampling should thus be done with a transition period during which both methods are run in parallel, 5 years being advisable for most solutes and for small streams.