Loren Amelang wrote:

I agree that as soon as transformers, or even dynamic mic coils are
involved, things get really complex.

And this encompasses at least half the mics in existence, so you can't
ignore them as part of a general discussion.

And in my line-level example, I'd personally prefer to have direct,
DC-coupled connections between each source and my amp - but of course
that would eliminate almost all commercially produced equipment.

Yup. But you can always build your own capacitorless electronics for a
condenser mic.

My best guess is that I am hearing effects of the DC blocking
capacitors in my source outputs - I don't know what else to suspect.

That's possible. Capacitors can be pretty non-linear. But either you're
awfully sensitive or you have a good imagination. Trouble is that you
can't eliminate them without changing a lot of other things, so you
can't compare with and without.

No transformers. Variable resistance across the amp inputs.

This could certainly affect the frequecy response of a mic with output
capacitors.

I can't "see" anything testing with pure tones, but then what I hear
is mainly changes in the "space" of the soundfield, rather than in the
tonality of the music, so the differences are probably tiny phase
shifts.

You can see and measure phase shift with test equipment. You just
aren't trying hard enough.