The problem discussed is whether the relation between Stokes and anti-Stokes light scattering in liquids follows the famous expression I-S/I-AS = exp(hv/kT). It is demonstrated that the disagreement of this expression with the results found by Watanabe et al. [Chem. Phys. Lett. 333 (2001) 113] is related to the wrong procedure of the determination of the origin of the spectra used in this work. It is shown that in framework of the standard paradigm the relaxational mode in light-scattering spectrum of liquids is placed at a non-zero frequency about 10(-2) cm(-1) in Stokes side of spectrum. (C) 2003 Elsevier Science B.V. All rights reserved.