A procedure for the correction of spherical aberration in microscopic applications, wherein various recordings of a specimen to be observed are taken and evaluated for the purpose of changing the setting values of the optical system. The correction values are stored in a correction matrix as a function of the recording position, the recording time, the wavelength, and the temperature, wherein the determination and storage of the correction values are carried out in each recording position in the x, y, and z coordinates, and/or the correction values are determined after a selection of grid points by interpolation, so that the correction values of the interpolated correction matrix are the starting values for the subsequent exact determination by measurement.