I repeated yesterday's test (but adding spectral line modes), using the same source, same method, at a time 1 hour later. The effect seen yesterday at 12.5 MHz BW in continuum is completely absent today. The sensitivities are all normal for the continuum modes.