Abstract
In a recent paper [Phys. Rev. B 47, 3005 (1993)], Carini et al. have reported on low-frequency Raman-scattering measurements made on samarium phosphate glasses. They analyze their data with the model of Martin and Brenig, which has been (and continues to be) traditionally used to account for the universal so-called ‘‘boson peak’’ dominating low-frequency Raman spectra of glasses. The aim of this Comment is to recall important work published in the past years, which has shown that the model of Martin and Brenig is not generally applicable to explain low-frequency Raman scattering in glasses.