The remarkable finding that B follows Fe in almost direct proportion from the earliest times to the present, with little change of slope (if any) between halo and disk metallicities (cf Figure 1, open circles, Duncan et al. 1997), has stimulated the need for a revision of the classical cosmic-ray (CR) spallation scenario. A straightforward interpretation of Duncan et al. results is that the net rate of production of B (and Be too, since it shows a similar trend) does not depend on the CNO abundances in the ISM. It has been suggested that the CR spallation most important for light element production is not primarily protons and alpha particles colliding with CNO nuclei in the ISM (Reeves et al. 1970; Meneguzzi et al. 1971). Rather, it is C and O nuclei colliding with ambient protons and alpha particles, probably in regions of massive star formation (Cassé et al. 1995). This decouples light element production from the metallicity of the ISM and results in the approximately linear relationships observed, but some aspects are difficult to justify (see Prantzos, this volume).
Particle Acceleration by Supernova Shocks and Spallogenic Nucleosynthesis of Light Elements
