So, dear reader, a few follow-ups.
I measured the resistance of the tongue after water, then after a mild salt solution, seen below:
and found no difference in resistance. I was unwilling to go "more salty" in the name of science. Today.
My little brother commented that you have to measure current in series, and voltage in parallel. So the question I had earlier about why the current dropped when I added another element in parallel reflects my igg-ner-unse of how things work.
Ah well. His electrical kung fu looks like this:
Mine, like this:
The setup below was reprised, and instead of pushing the key I just measured across the leads, putting the multimeter in series as it should be:
complete with the 9V battery. The measured resistance of the tongue today was about 320 kohms.
So the resistance du jour was around 835 kohms, and with the battery consistently putting out about 9.7 volts, that predicts around 11 microAmps.
Measured amps across the tongue was in the range of 14 microAmps. Given the literature I read, I was expecting a threshold around 1 microAmp, perhaps lower. Need More Resistance!
Turns out that to get down to 1 microAmp (an order of magnitude drop from 10 microAmps) I need an order of magnitude more resistance. 9.7 Mohms, to be precise. I don't have enough resistors to do that, even if I wire them all in series I barely cross 1 Mohm. So we downgraded the battery to AA and 1.56 volts.
So Mr. Ohm predicts 1.8 microAmps with this approach, and indeed we were in the ballpark. Turns out my threshold is somewhere just shy of 2 microAmps. Turns out in my previous experiment that when I was pushing the switch that "connected" the circuit, I wasn't actually changing the current, despite a change in the multimeter reading. More on that later.
So Experiment 1, subexperiment 1 is done. Threshold: around 1.8 microAmps.