N is the total number of photons registered by the detector, and it is seen that the spatial resolution of the microscope, taken as its ability to localize a point source, Δmin , is improved from Abbe’s original limit by the factor 1/√N. This means that with 100 detected photons from the single point source Δmin is a factor of 10 smaller than Abbe’s limit and with increasing numbers of detected photons there is no strict limitation to how much the resolution can be improved. For this, the structure of interest must be so sparsely labeled with fluorophores that their separation is larger than the diffraction limit. This condition, however, leads to yet another problem. According to the sampling theorems by H. Nyqvist (1928) and C. E. Shannon (1948) a spatial resolution, Δmin, in the reconstruction of a structure requires that the structure isuniformly sampled with a spatial frequency higher than 2/ Δmin. This condition, however, is at odds with the condition that the nearest neighbor distance between labels is larger than Abbe’s diffraction limit. In fact, what is required is the dichotomy of a very sparse distribution of fluorescence labels to save the assumption that they are all separable point sources in conjunction with a very dense distribution to fulfill the sampling theorems of Nyqvist and Shannon. A conceptual solution to this logical dilemma was first provided by Erik Betzig.