Comment on "Anisotropic superconductors with repulsive average interaction"

In a recent paper Whitmore and Carbotte concluded that a sufficient condition for the superconducting state to be stable relative to the normal state over a finite temperature range $0<T<\mathrm{}{T}_c$ is that the attractive part of the effective electron-electron interaction vary over the Fermi surface. We show that this surprising conclusion is not true in general but results from the assumption of a separable attractive interaction.