Thanks for your suggestions. Here's the latest...

The alt. light is no longer affected by bumps. It stays on at startup, glowing at about half intensity, and almost fades completely out when revved up to about 4K RPM. The voltmeter never shows more than battery voltage, however, so the alt. is never putting out proper charge.

I've replaced the whole subsection of the wiring harness that runs down by the oil p. sender, alternator, etc... so all of that stuff is fresh and good. Just to be certain, I double-checked the integrity of that signal path back to the grey plug, and it checked out. I also checked the fuses, particulary #13, and they're all in good shape.

When I pulled the old voltage regulator, the brushes were practically gone. I thought this was a promising sign (it had to be the brushes!), so I expected popping in the new regulator with shiny new brushes would do the trick. WRONG! But the behavior of the light is exactly the same, so at least I know it is completely unrelated to the regulator. Unless... can brush 'dust' accumulate and cause problems inside the alternator body?

I believed that if I pulled the small red wire off it's spade and measured the voltage at that spade with the car running, I should see at least battery voltage (which is what keeps the warning light from turning on.) If there is a fault within the alt./regulator unit, the warning light essentially grounds out to some degree through this spade. Is this right?

I still need to verify the big connections to the battery and to the block, to make sure I don't have any issues with corroded/high resistance wires, but aside from that, all indications seem to point to a problem inside the alternator.

I appreciate all the responses. Yes, working in -22C weather is worse - your hands keeps flash-freezing to your tools, and the icicles on your eyelashes make it really hard to see what you're working on!
--
Speed Racer, '83 240 R, '74 164 E, '93 940 OL1 (Manchester, CT)