We discuss ellipsometric experiments seeking evidence of broken time-reversal symmetry in the high-temperature copper oxide superconductors. We use both a generalized symmetry analysis, and a magnetoelectric model suggested by Dzyaloshinskii, to argue that the hypothesis that best fits the various experimental results is the scrPscrT-invariance hypothesis—which assumes broken scrT symmetry in each plane, and antiferromagnetic (alternating) order of the broken symmetry in the c direction. We suggest two experimental tests of the scrPscrT-invariant model; one of these is sufficient to rule out any other broken-symmetry state and so has the potential to be extremely useful.

• Received 29 June 1992

DOI:https://doi.org/10.1103/PhysRevB.46.14078