Comment on "Protein phase diagrams: The physics behind their elliptic shape" [J. Chem. Phys. 121, 12671 (2004)]

The article by Lesch "Protein phase diagrams: The physics behind the elliptic shape" [J. Chem. Phys. 121, 12671 (2004)] is criticized since it involves assumptions, which violate a basic statistical theorem: the combined variance of a linear combination of statistically independent variables is always given by the sum of the variances. Thus their main conclusion that correlated fluctuations in the transition volume and enthalpy are the origin of the observed nearly elliptical stability diagrams of proteins should be questioned.