Comment II on "Resonant and antiresonant frequency dependence of the effective parameters of metamaterials".
In a recent paper, Koshny, Markos, Smith, and Soukoulis [Phys. Rev E 68, 065602 (2003)] report that their computations show that the product of imaginary parts of electric permittivity and magnetic permeability in a passive media is negative. They also criticize a well-known theorem that both imaginary parts are positive as a result of the second law of thermodynamics. I argue that this criticism has no ground and that computational evidence may result from inadequate introduction of the very concept of epsilon (omega) and mu (omega) in a photonic crystal.