The STORM imaging procedure. (A) Schematic of a cell in which the structures of interest (gray filaments in this case) are labeled with photoswitchable fluorophores (not shown). All fluorophores are initially in the nonfluorescent state. The red box indicates the area shown in panels B–D. (B) An activation cycle: A sparse set of fluorophores is activated to the fluorescent state such that their images (large red circles) do not overlap. The image of each fluorophore appears as a diffraction-broadened spot, and the position of each activated fluorophore is determined by fitting to find the centroid of the spot (black crosses). (C) A subsequent activation cycle: A different set of fluorophores is activated, and their positions are determined as before. (D) After a sufficient number of fluorophores has been localized, a high-resolution image is constructed by plotting the measured positions of the fluorophores (red dots). The resolution of this image is not limited by diffraction but by the accuracy of each fluorophore localization and by the number of fluorophore positions obtained. (Adapted, with permission, from Bates et al. 2008, ©Elsevier.)