Time-resolved infrared spectroscopy offers the unique opportunity to
 disentangle the main components of infrared emission from magnetic
 cataclysmic variables: The first harmonics of the cyclotron radiation
 and the photospheric emission of the late-type secondary star. This
 allows to determine the configuration and strength of the magnetic
 fields and to derive the distances of the systems.
 We have selected a sample of different types of magnetic cataclysmic
 variables and propose to study the three most promising targets.
 The fundamental cyclotron frequencies of these systems are expected
 to lie at ~5 um or even longer wavelengths and a spectroscopic coverage
 of the lowest harmonics will only be possible with ISO. As the
 cyclotron flux emitted towards the observer varies with the viewing
 angle with respect to the magnetic field, observations at different
 orbital phases are required to obtain full information about the field.
 The infrared spectra will be compared with cyclotron model spectra to
 determine the field strengths, shock temperatures, and geometries of
 the accretion regions.
 Due to the short orbital periods of AM Her stars, these systems contain
 late M-type main sequence secondaries which in most cases can only be
 detected in the infrared. Using mass-radius and mass-luminosity
 relationships, the infrared flux of the secondaries is an important
 indicator for the distances of cataclysmic variables. Hence, flux
 determinations of the secondary stars will yield the accretion
 luminosities and provide information on the space density of magnetic
 cataclysmic variables.