Unitarity driven nonperturbative effects in deep inelastic scattering

We study several semi-inclusive production mechanisms in deep inelastic scattering and high energy hadronic scattering and their impact one the total virtual photoabsorption cross section (the structure function) as dictated by the unitarity relations for the scattering matrix. We first consider the $\overline{d} - \overline{u}$ asymmetry in the proton sea, driven by the pionic content of the proton. The relevant pion fluxes are determined by an analysis of inclusive production of forward neutrons and Delta isobars in high energy proton proton collisions. We consistently take into account the contributions from the exchange of the reggeized rho meson, as well as the background to the neutron production coming from the Delta decays. With the obtained parameters a gooddescription of the recent Fermilab data is achieved. The following parts of this thesis are devoted to the diffractive production mechanism in deep inelastic scattering and its role in the theory of nuclear shadowing. We use the color dipole formulation of diffractive deep inelastic scattering and apply the formalism to the calculation of the shadowing corrections of the nuclear parton distributions at small x. The formalism explicitly supports the well known relation between diffractive production and nuclear shadowing. By explicit inclusion of the q$\overline{q}$g-Fock state of the virtual photon, we evaluate the contribution to nuclear shadowing from the triple-Pomeron region of the diffractive mass spectrum. We relate this contribution to the shadowing correction of the nuclear gluon distribution and give predictions for the latter. In the last part of the thesis, we investigate the influence of the s-channel helicity violating diffractive amplitudes an the spin structure function g$_{LT}$ = g$_{1}$ + g$_{2}$ of the proton. We find that this mechanism leads to a steep small-x rise of g$_{LT}$. In particular it is an explicit mechanism that violates the Burkhardt-Cottingham sum rule and the Wandzura-Wilczek relation. Hence the assumptions behind these appear unjustified