Electrophysiology of the Glands 203 



distinguishes the hyperpolarization in salivary gland cells from that 

 found in the heart, spinal motor neurons and crustacean stretch 

 receptor cells (see Eccles, 1957). In all these cells the degree and 

 the sign of the potential change produced by nerve stimulation is 

 dependent on the resting potential of the cell. If the resting poten- 

 tial is depressed below normal by the passage of current, hyper- 

 polarization is accentuated ; when the resting potential is increased 

 by some 10 mV to 20 mV, the hyperpolarization disappears and 

 is replaced by depolarization. Analysis has shown that in all these 

 cases the potential is due solely to an increase in conductivity for 

 one or more of the several ions crossing the membrane so as to 

 permit the cell potential to approximate more closely to a diffusion 

 potential (or combination of diffusion potentials). The cell thus 

 tends to approach the equilibrium potentials for the ions con- 

 cerned. If the resting potential is lower than the equilibrium 

 potential, hyperpolarization results ; if it is higher than the equili- 

 brium potential, depolarization results. Clearly, the hyperpolari- 

 zation in Type I cells cannot be explained in this way but must be 

 due rather to the active movement of an ion relatively unhindered 

 by the electrochemical gradient. The existence of active ion move- 

 ments of this kind have been established for sodium in the frog 

 skin (Ussing and Zerahn, 1951) and for chloride in the gastric 

 mucosa (Hogben, 1955). Accompanying the hyperpolarization the 

 conductivity of the outer cell membrane increases from 74 mmho/ 

 cm 2 to 141 mmho/cm 2 . The increased ion movement causing the 

 hyperpolarization could be due to increased cation extrusion from 

 the cell or, alternatively, to active anion entry. Lundberg favours 

 the latter explanation because it involves active movement of an 

 ion in the same direction as that in which secretion is presumed 

 to be going on. However, potassium loss from these cells does 

 occur (see page 208), but whether this involves active ion transport 

 is quite unknown. Because secretion proceeds normally when the 

 gland is perfused with a salt solution, in which the sole anion is 

 chloride, Lundberg favours chloride as the likely ion. He has 

 attempted to test this hypothesis by replacing chloride in the per- 

 fusing fluid by another anion and measuring the effects on cell 

 potential change and secretion produced by injecting 1 fig acetyl- 

 choline (Lundberg, 1956, 1957c). Replacement of chloride by 

 bromide produced no change, but replacement by nitrate reduced 

 both the secretion rate and the hyperpolarization to about 15 per 



