It is to be noted that the statement (V-1) is restricted to reversible processes. This excludes the possibility of evaluating the total change of entropy in terms of added heat or radiation alone for systems which involve diffusion or viscosity. Both of the latter processes can bring about changes of entropy within the system, even if dQ is zero. Furthermore, the system must be homogeneous with regard to temperature if (V-1) is to be valid, otherwise the effect of heat conduction within the system can lead to internal generation of entropy. Thus the condition of reversibility of the process places rather severe limitations on the use of (V-1).