DESCRIBING ACTUAL CRITICAL-BEHAVIOR FROM FIELD-THEORY - A DELICATE MATTER

The renormalization flows evolve in a space involving an infinite number of degrees of freedom. This aspect is not conspicuous in the field theoretical version, which only accounts for a specific kind of degrees of freedom. We shortly present and discuss two examples that illustrate the difficulty of going beyond this ''simplification'': (1) The case of negative correction-to-scaling terms in critical behaviour (usually thought to be not describable by held theoretical techniques). (2) The domain of validity of a theoretical constraint on the critical behaviour of the specific heat (namely R(Bcr)) that we had calculated in the framework of field theory and is not a universal quantity in the usual sense.