Sedimentation is part of the swimming behaviour of cells in the gravity field. Determinations of sedimentation are indispensible for the isolation of graviresponses of cells (gravitaxis, gravikinesis). Using a low-speed microscope centrifuge (NIZEMI) we have determined the sedimentation rates of various protozoan cells (Paramecium tetraurelia, Paramecium caudatum, Tetrahymena pyriformis, Didinium nasutum and Loxodes striatus) as a function of acceleration up to 7 g, iron uptake, deciliation, two different immobilization agents (NiCl2 , MnCl2), and Ca2+-Mg2+ composition of the experimental solution. The experimental results suggest a linear relationship between sedimentation rate and acceleration up to 2 g intercepting the origin (zero sedimentation rate at 0 g). Beyond 2 g the sedimentation curves tend to depart from linearity, depending on cell shape and size, state of feeding, immobilization procedure, and time passed since the onset of sedimentation. The time-dependence of observed mean values of sedimentation rate rises with acceleration due to the continuing loss of the faster sedimenting cells from the observed field. The original distribution of sedimentation velocities of a cell sample can be calculated from the data, as long as cells of the highest velocity classes have not escaped registration. Correcting experimental data for time-dependent modification of the velocity distribution modifies the sedimentation curves toward linearity. A strict linearity of the velocity-g relationship was nevertheless not obtained for g-values beyond 2 g suggesting that the departure from linearity is not only a time-dependent artifact but is affected by additional factors such as the surface properties of cells and the proximity of solid walls enclosing the experimental fluid space.