The present study devises a method to get quantitative information for proteins on the theoretically important heat capacity at constant volume. For this purpose expansion coefficients of both the native and unfolded state of a variety of proteins have been determined and used together with compressibility coefficients to calculate the difference between isobaric and isochoric heat capacity, (c(p)-c(v)), for the unfolding transition. This difference Delta(c(p)-c(V)) = (c-c) - (c-c) turns out to be a positive number that is larger than the experimental isobaric heat capacity change Delta(N)(D)c(p) = c(P)(D)-c(P)(N) for the proteins studied. The generally observed positive heat capacity change on unfolding can therefore alternatively be interpreted as resulting from the difference in work involved in changing the intra- and intermolecular interactions including the weak, highly distance-dependent, van der Waals interactions, for the unfolded and native state, respectively. The difference in expansion work against the atmospheric pressure is negligible. This macroscopic interpretation cannot rule out that part of the denaturational heat capacity increase is also due to the different interaction with water of the native and unfolded conformations of the protein.