Abstract
The possibility of splitting the inclusive cross section for relativistic nuclear collisions into a direct and a thermal component is examined. It is pointed out that the definition of a direct term is model dependent. Besides this difficulty, it is explained why two particle correlation measurements can give ambiguous information on the size of the direct component. The decomposition of the inclusive cross section in the context of Monte Carlo calculations, according to the number n of collisions undergone by the ejected nucleons, is considered. Numerical results for Ne + Ne and Ca + Ca at EA=800 MeV show that the cross sections cannot be split into two parts that could be regarded as a direct and a thermal part, but rather indicate a continuously rising departure from equilibrium as n decreases. Dominance of some values of n in some regions of the momentum space is exhibited. The Δ-particle spectra show a completely different pattern and look similar for all values of n.