Lipid analogues and glycosylphosphatidylinositol (GPI)-anchored proteins incorporated in glass-supported phospholipid bilayers (SBL) were coupled to small (30 nm diameter) fluorescent beads whose motion in the liquid phase was tracked by intensified fluorescence video microscopy. Streptavidin (St), covalently attached to the carboxyl modified surface of the polystyrene bead, bound either the biotinylated membrane component, or a biotinylated monoclonal antibody (mAb) directed against a specific membrane constituent. The positions of the beads tethered to randomly diffusing membrane molecules were recorded at 0.2 sec intervals for about 1 min. The mean square displacement (rho) of the beads was found to be a linear function of diffusion time t, and the diffusion coefficient, D, was derived from the relation, rho(t) = 4Dt. The values of D for biotinylated phosphatidylethanolamine (Bi-PE) dispersed in an egg lecithin:cholesterol (80:20%) bilayer obtained by this methodology range from 0.05 to 0.6 micron 2/sec with an average of mean value of D = 0.26 micron 2/sec, similar to the value of mean value of D = 0.24 micron 2/sec for fluorescein-conjugated phosphatidylethanolamine (Fl-PE) linked to St-coupled beads by the anti-fluorescein mAb 4-4-20 or its Fab fragment. These values of D are comparable to those reported for Fl-PE linked to 30 nm gold particles but are several times lower than that of Fl-PE in the same planar bilayer as measured by fluorescence photobleaching recovery, D = 1.3 microns 2/sec. The mobilities of two GPI-anchored proteins in similar SBL were also determined by use of the appropriate biotinylated mAb and were found to be mean value of D = 0.25 and 0.56 micron 2/sec for the decay accelerating factor (DAF, CD55) and the human Fc gamma RIIIB (CD16) receptors, respectively. The methodology described here is suitable for tracking any accessible membrane component.