The close packed (001) face of the anisotropic molecular crystal biphenyl has been studied using X-ray reflectivity upon approaching the melting point. No evidence for surface melting has been found, in contrast to the observations of Chernov and Yakovlev [Langmuir 3 (1987) 635] for the same crystal face against glass. We speculate that in the latter case the surface melting is not intrinsic but induced by the roughness of the glass.