I take a simple model of the hydrogen atom in a universe without spatial cu
rvature. The Maxwell equations are formulated on the background cosmic spac
etime. For a class of cosmic metrics, which includes the de Sitter universe
, these equations admit solutions corresponding to an atom whose radius rem
ains strictly constant during the expansion. In the Einstein-de Sitter univ
erse approximate calculations show that the atom expands, but at a rate whi
ch is negligible compared with the general cosmic expansion.