In a regular microscope, the fluorescence of molecules is excited by light of wavelength λ_ex focused by an objective lens characterized by its numerical aperture (NA). Light emitted by these fluorophores (at wavelength λ_em > λ_ex) is collected by the same objective lens and imaged by the tube lens as a spot (the point spread function (PSF) of the microscope) of radius r_Airy = 0.61λ_em/NA (see Figure 2.1). Two point sources separated by a distance d significantly smaller than r_Airy cannot be resolved easily,¹  as their PSFs merge into a single spot (the so-called Rayleigh criterion). This diffraction effect has defined the resolution limit of microscopes for more than a century and has been one of the motivations for the development of electron microscopy. According to the Rayleigh criterion, the classical resolution limit of a good microscope objective (i.e. with NA = 1.4) is 0.44λ_em. Thus, when observing cellular structures labeled with fluorophores such as the green fluorescent protein (GFP), emitting at λ_em = 510 nm, details below 200 nm cannot be resolved. In the past 10 years, methods have been developed that overcome this constraint and allow for almost unlimited resolution (i.e. super-resolution) in optical microscopy.