Water fluxes in nerve fiber.

The hydrostatic (Lp) and osmotic (LPD) filtration coefficients and the efflux rates of tritiated water were measured in the giant axon of Loligo vulgaris. The Lp was 8 to 14 X 10(-8) cm/sec/cm H2O and the LPD was two orders of magnitude smaller (3 to 6 X 10(-10) cm/sec/cm H2O). In axons whose diameter was approximately 500 micron, the time (t1/2) required for a reduction in the axonal labeled water activity to one half its initial value was 38 to 48 sec. The rate limiting structure for solute flux was made ineffective by (1) storing the axon in isosmotoc KF at 0-2 degrees C for one month to one year or by (2) fixing the axon in 2-4% glutaraldehyde for 3 to 7 hr. The criteria of ineffectiveness of the rate limiting structure for solute flux were (1) a reduction of LPD to immeasurably low values, (2) the absence of electrical properties characteristic of plasmalemma, and (3) a marked increase in the rate of efflux of Na22. In such impaired axons the Lp and the t 1/2 of tritiated water efflux were unaffected. This independence of solute and solvent flux in conjunction with the finding that the hydraulic conductivity determined by bulk osmotic and hydrostatic pressure gradients is not equivalent (i.e., LPD/LP less than 1) indicate that the rate limiting structures for solute and solvent flux are in series. Solvent fluxes appear to be surface-limited, not bulk-limited. We have been unable to resolve whether the surface structure involved in limiting solvent flux is the sheath (Schwann layer and adhering connective tissue) and/or the cortical layer of the axoplasmic gel.