Discussed is the classical theoretical description of the experimental
ly established thermal redshift of spectral lines. Straightforward cal
culation of the observable spectrum from a canonical ensemble of monoc
hromatic radiators yields an overall blueshift rather than a redshift.
It is concluded that the customary explanation of the thermal redshif
t as a second-order Doppler effect does not bear closer examination, a
nd that in fact, the phenomenon ''thermal redshift'' is not yet fully
understood in classical terms. (C) 1998 Published by Elsevier Science
B.V.