Kamitsos et al. [Phys. Rev. B 48, 12 499 (1993)] studied the infrared-
(IR) reflectance spectra of bulk SiO2 gels and attributed the high-fr
equency shoulder near 1200 cm(-1) to the TO component of the out-of-ph
ase antisymmetric stretching vibration of the oxygen atoms along a lin
e parallel to the two adjacent Si atoms, contrary to the assignment of
Almeida and Pantano [J. Appl. Phys. 68, 4225 (1990)], who attributed
this shoulder to the LO component of the in-phase antisymmetric stretc
h, indicating that the latter assignment was wrong. This criticism is
contested, based on evidence provided by a previous publication that w
as overlooked [R. M. Almeida, Phys. Rev. B 45, 161 (1992)] and on the
errors introduced in the Kramers-Kronig transformation by the extremel
y low IR reflectivity measured by Kamitsos et al. for their gel sample
s (as low as 2% for the 1086 cm(-1) peak, compared with values in the
range of 20-31 % for gels prepared in our laboratory).