Hadrons are composed of quarks and gluons which interact in a complex way to form bound- and resonance states of color singlets. It is of paramount importance to understand the non-perturbative mechanisms for the formation of these states from first principles. Important contributions are made by numerical simulations which begin to provide a degree of accurancy such that quantitative comparisons with experiment become possible.
Of particular interest is the internal structure of the nucleon, as probed by leptons in a wide range of Q2 and Bjorken x. The experimental programs at MAMI and JLAB provide important insights on low-momentum form factors and the role of the pion cloud, the onset of QCD scaling and the spatial- and momentum distributions of quarks in the nucleon. The orgin of the the nucleon spin remains a central problem in experimental and theoretical studies with high-energy leptons at COMPASS and HERMES and closely relates to the gluon distribution functions, for which a detailed understanding is still lacking.
Based on quark models it has long been suggested that there should exist hadrons composed not only of quark-antiquark states or three-quark states but also hadrons consisting of genuine "multi-quark" configurations might exist. An alternative explanation for such states is the exstence of "hadronic molecules" as bound- or resonant color singlet hadrons. The interplay between these descriptions of observed data remains the focus of current debates. QCD predicts the existence of hadrons purely made of gluons, so-called "glueballs". There are many theoretical indications for the existence of such states, but they have not yet been indentified experimentally. A possible explanation is that glueballs strongly mix with quark-antiquark states and hence aquire a large width. Glueball searches and multi-quark spectroscopy will be a central part of the future PANDA project at FAIR in proton-antiproton annihilation experiments.
Experiments of the BABAR, BELLE and BES for the charmonium have revealed a rich spectrum of narrow states near the D-Dbar threshold, that came a a total surprise. The origin of these states and their detailed structure remains a challenge to experiment and theory.
Closely related to the molecular structure of hadrons is the general question of the the spin- and flavor dependence of meson-meson and meson-baryon interactions. In the non-strange and the strange sector these are currently studied in detail at COSY. Meson-exchange models are generally used to describe the data. In the future it is hoped that lattice QCD will provide ab-initio input for such calculations. First numerical results for the nucleon-nucleon and hyperon-nucleon interaction are already available. In the near future new experimental results, especially in the strangeness sector are expected from J-PARC.
There is considerable effort to understand the properties of light nuclei from QCD directly. Except for the Deuteron, it is at present computationally probitive to compute nuclear observables directly on the lattice. A more economic, yet also fundamental route is effective field theory (EFT). It is well-known that chiral perturbation theory is the correct low-energy limit of QCD. Within this framework two- and three-body interactions can be constructed to a given order in the expansion parameter Q2/4F&pi2. These are then used in ab-initio few-body calculations to fix some of the low-energy constants of the EFT. The properties of larger nuclei can then be inferred in a model-independent way. At present nuclei up to mass 10 can be treated in this way. EFT can also be applied to homogeneous nuclear- and neutron matter to compute the equation of state of neutron stars.