Nuclei can fluoresce in much the same way as atoms do. A major difference between nuclear and atomic fluorescence, however, lies in the magnitude of the recoil energy involved in the emission or absorption of a photon. For atoms, where the recoil energy is small, fluorescence is readily achieved. For nuclei the recoil energy is much greater and, for identical nuclei in free atoms, puts the energies of emission and absorption out of step. It was Rudolf Mössbauer's discovery that, for nuclei placed in solids at low temperature, there is, however, a finite probability that emission or absorption of the gamma-ray photon will take place without absorption or emission of a phonon. This means that the solid will be in the same internal state before and after the event, so that, in effect, the recoil is taken up by the crystal as a whole and not by an individual atom. This makes the recoil energy immeasurably small and puts the emission and absorption processes back in step. The result is a spectroscopy having the the resolution of the lifetime uncertainty of the excited nuclear state, or about 10-8 electron-volts. Usually, this resolution is good enough to resolve the so-called hyperfine energy-splittings of the nucleus caused by electronic fields. Often, the hyperfine energy levels are scanned by moving a radioactive source repetitively toward and away from an absorber. Through the Doppler shift, this varies the energy of the gamma-rays arriving at the absorber. Some of these ideas are illustrated in the graphic below.