At high energy, exclusive meson photo-and electro-production give access to the structure of hadronic matter. At low momentum transfers, the exchange of a few Regge trajectories leads to a comprehensive account of the cross-sections. Among these trajectories, which are related to the mass spectrum of families of mesons, the Pomeron plays an interesting role as it is related to glue-ball excitations. At high momentum transfers, the exchange of these collective excitations is expected to reduce to the exchange of their simplest (quark or gluon) components. However, contribution from unitarity rescattering cuts are relevant even at high energies. In the JLab energy range, the asymptotic regime, where the players in the game are current quarks and massless gluons has not been reached yet. One has to rely on more effective degrees of freedom adapted to the scale of the probe. A consistent picture, the Partonic Non-Perturbative Regime, is emerging. The properties of its various components (dressed propagators, effective coupling constants, quark wave functions, shape of the Regge trajectories, etc..) provide us with various links to hadron properties. I will review the status of the field, will put in perspective the current achievements at JLab, SLAC and Hermes, and will assess future developments that are made possible by continuous electron beams at higher energies.
The p(γ, φ)p reactionThis picture works best when the φ meson is emitted at the most forward angles (that drive the total cross section). At higher −t > 1 GeV 2 , the Pomeron exchange contribution falls down faster than the experimental data as can be seen in Figure 3. The data are better reproduced by modeling [5,6,2] the Pomeron in terms of the exchange of two gluons as depicted in Figure 4. When the two gluons couple to the same constituent quark in the nucleon (bottom part), the interference between their coupling to the same constituent quark or to different quarks in the φ meson induces a node in the cross section. This node disappears when the coupling to different correlated quarks in the nucleon is allowed (top part). The two gluon contribution matches the Pomeron contribution at the lowest −t, but provides more strength at high −t.It turns out that, in the high energy limit, the Pomeron and the two gluon amplitudes have the same functional form and the same structure at low −t. The treatment of the quark loop in the vector meson is the same in the two approaches. The Pomeron Regge pole drives the dynamics of its exchange. Since it is a perturbative-like calculation there is no Regge factor in the two gluon exchange amplitude: