Action Potential bandwidth

Iain McClatchie iain at 10xinc.com
Sun Aug 20 22:42:56 EST 2000


Bill> So your hypothesis would seem to have at least one very easily
Bill> testable prediction:  you should be able to shine light on one
Bill> end of a cut piece of nerve and see it come out the other end.

Heh heh.  Read about the difficulties people have launching light
into single-mode glass fiber.  You have to hit a target just a few
microns in diameter, with light almost exactly normal to the surface.
Essentially, you need a laser and an optical bench accurate to a
micron or so.  People don't use single-mode fiber for moving lots
of light: they use multimode, which is a lot cheaper and a lot easier
to launch into, and which has a core diameter of 50 or 62.5 microns.

So how do you cut a nerve in such a way that you can see a cross-
section of the bilipid layer, without screwing up that bilipid
layer?  Maybe you could float a tiny conical mirror in the bilipid
layer, so that light shining on the mirror would be reflected into
the layer.

Another unfortunate issue is that the bilipid layer doesn't have to
be particularly lossless, since transmission across few hundred
microns to a neighboring ion channel would be all that's necessary.

Finally, I have only vague guesses at the frequency of light used.
Dropping an ion across 100 mV is enough energy for a 12 micron photon,
which is way deep infrared.  Maybe multiple ions get dropped across
the voltage to successively pump the emitting electron to higher and
higher energies, in which case the wavelength used could be nearly
anything.

Another testable hypothesis would be that an active neuron, sliced
through the axon, ought to have light emitted from the sliced
cross-section (unless it was so disturbed by cutting that it stopped
data transmission but continued action potentials).  Avalanche photo
detectors wouldn't be terribly sensitive to the angle of emission,
nor to the wavelength emitted, so long as the photon energies were
less than about 10 ion drops.

-Iain McClatchie                            650-364-0520 voice
http://www.10xinc.com                       650-364-0530 FAX
iain at 10xinc.com                             650-906-8832 cell






More information about the Neur-sci mailing list