It is shown that the data reported by Nozawa and Horikoshi [Jpn. J. Ap
pl. Phys. 32 (1993) 626 on the tilting of GaAs layers grown on Si by m
igration-enhanced epitaxy can be explained by the layer-by-layer natur
e of the growth process. It is also pointed out that the azimuthal ang
ular shift of the GaAs-Si X-ray rocking curve peak position from the s
ubstrate miscut direction does not hint to the in-plane rotation of th
e epilayer as a whole.