We investigate the structure of the scalar mesons f 0 (975) and a 0 (980) within realistic meson-exchange models of the ππ and πη interactions. Starting from a modified version of the Jülich model for ππ scattering we perform an analysis of the pole structure of the resulting scattering amplitude and find, in contrast to existing models, a somewhat large mass for the f 0 (975) (m f 0 = 1015 MeV, Γ f 0 = 30 MeV). It is shown that our model provides a description of J/ψ → φππ/φKK data comparable in quality with those of alternative models. Furthermore, the formalism developed for the ππ system is consistently extended to the πη interaction leading to a description of the a 0 (980) as a dynamically generated threshold effect (which is therefore neither a conventional qq state nor a KK bound state). Exploring the corresponding pole 1 position the a 0 (980) is found to be rather broad (m a 0 = 991 MeV, Γ a 0 = 202 MeV). The experimentally observed smaller width results from the influence of the nearby KK threshold on this pole.

We review the current status of rho-omega mixing and discuss its implication for our understanding of charge-symmetry breaking. In order to place this work in context we also review the photon-hadron coupling within the framework of vector meson dominance. This leads naturally to a discussion of the electromagnetic form-factor of the pion and of nuclear shadowing.Comment: 52 pages with 14 figures. This published version includes updated references and minor changes to tex

No abstract

A dynamical model for S-and P-wave correlated 2π (and KK) exchange between a kaon and a nucleon is presented, starting from corresponding NN → KK amplitudes in the pseudophysical region, which have been constructed from nucleon, ∆-isobar and hyperon (Λ, Σ) exchange Born terms and a realistic meson exchange model of the ππ → KK and KK → KK amplitude. The contribution in the s-channel is then obtained by performing a dispersion relation over the unitarity cut. In the ρ-channel, considerable ambiguities exist, depending on how the dispersion integral is performed. Our model, supplemented by short range interaction terms, is able to describe empirical K + N data below pion production threshold in a satisfactory way.

Within a broad class of models we show that the amplitude for rho^0-omega mixing must vanish at the transition from timelike to spacelike four momentum. Hence in such models the mixing is either zero everywhere or is necessarily momentum-dependent. This lends support to the conclusions of other studies of rho-omega mixing and calls into question standard assumptions about the role of rho-omega mixing in the theoretical understanding of charge-symmetry breaking in nuclear systems.Comment: 8 pages. Publication details added to title pag