Clinical heavy ion beams provide the most accurate means of external radiation treatment of tumors available. The study of energetic heavy ions on individual cells have been hampered by shortcomings in available detector technology. The fluorescence nuclear track detector developed by the Landauer Inc. is a small chiplike detector with all the properties needed for these kinds of studies. It is biocompatible, autoclavable, does not require post-irradiation chemical processingand allows the readout of a sufficient number of physical parameters for the determination of a particle’s radiobiological and physical properties. Previous studies in clinical spread-out Bragg peaks have shown a discrepancy in what is detected to what is seen in Monte Carlo radiation transport simulations. While the fluence of heavy primary ions agrees to within one percent, the detection of fragments is underestimated by at least an order of magnitude. In this thesis the performance of the track-recording has been assessed for light ion fragmentsfrom hydrogen, helium and carbon primary particles in physically optimised spread-out Braggpeaks. The underestimation of light ion fragments for carbon beams, mentioned in previousstudies, was reproduced. The underestimation of light ion fragments is prevalent also for helium primary particles. No significant amount of lost tracks are seen in hydrogen beams. The analysis verifies the hypothesis that the detection of large angle trajectories for light ion fragments is not a principal limitation of the detector. The underestimation most probably stems from some limiting features of the readout and image analysis procedure.