Measurements of sedimentation were combined with water samples to calculate settling velocity of suspended particulate matter (SPM) in lakes. The study sites were open water stations and enclosures in Lake Erken (Sweden) and Lake Balaton (Hungary). Settling velocities were found to vary considerably both inside and outside the enclosures. Within sites, the differences between 25th and 75th percentiles of measured settling velocities of SPM were two-or three-fold. Median settling velocities of SPM ranged from around 0.5 m/d in the enclosures of Lake Erken to more than 8 m/d in the open water of Lake Balaton. Special relevance was attributed to flocculation, which is known to be affected by, e.g., SPM concentration and turbulence. Even though not directly measured, the less turbulent environment inside the enclosures was suggested to explain the low settling velocity compared to the open water environment. Settling velocity apparently correlated with water current speed (r 2 = 0.66; n = 12). Stepwise multiple linear regressions were used to relate the variability in settling velocity of SPM to the variability of possible controlling factors in a number of data subsets. In most cases, one variable describing the total amount of settling material (e.g., SPM) and one variable reflecting the composition of settling material (e.g., total phosphorus) were chosen. The use of suspended solids concentration to predict settling velocity in mass balance models was discussed. It was found that the mean slope between SPM and v SPM was close to 0.1 (m 4 d -1 g -1).