"News Features" wrote in message
news:lDumk.35103

I hope you're wrong about the two lights switching places; somehow, I
can't
imagine the "bad" light gradually becoming RFI-quiet over time... by what
mechanism could that happen?

I can think of one, but it's obscure and unlikely. Some electrolytic
capacitors must be "formed" by application of voltage to develop their
rated characteristics. (However, it's applied DC that usually does such
forming, and I think all production flourescent lamps run exclusively on
AC.) If an unformed cap were installed as a filter and if it were later
formed by applied voltage, that is the mechanism that occurs to me.

Contradictions welcome.