Statistical Thermodynamics of Rubber Elasticity

The basic concepts and procedures underlying the theory of rubber elasticity are analyzed. The arguments advanced by Wall and Flory against the network theory of the authors are shown to be without foundation. The alternative theory which they advocate, and which is rather widely used, is criticized on two grounds: (a) it assumes a distribution of chain extensions that does not vary with network extension in the way the actual distribution functions can be shown to vary, and (b) the methods used to compute the numbers of network configurations from the distributions of chain extensions are invalid, except in the most trivial cases. The concept of ``interval dilation entropy'' is discussed and is shown to lack any physical basis. It is concluded that all the methods of calculation that lead to logarithmic terms in the network configuration entropy are incorrect, and that these terms, which affect the theory of swelling, should be discarded.