THE
atmospheric electric elements measured were potential gradient and residual ionization in a closed vessel. It was felt that any possible effects of the eclipse would be more likely to show themselves in these elements than in the ionic content or the atmospheric electric conductivity as measured at the earth's surface. At any rate, any effect on conductivity or ionic content other than those arising from changes in potential gradient or extra-terrestrial penetrating radiation would probably be the result of secondary causes, such as change of humidity, atmospheric radio-active content, temperature, etc., resulting from the eclipse, and would not be of such primary interest as changes in the potential gradient and residual ionization themselves.
Even such changes of the conductivity as would result from change in intensity of ion production would only show themselves with considerable lag. Thus, for example, if q is the rate of production of ions, $2 the number of ions per c.c., the numbers being supposed equal for both signs, as an approximation, and if (Y is the coefficient of recombination of ions, we have which leads to dn for a case where II + o when t = O. Or, since fzO, the equilibrium value of n, is given by ?&)=A* a I+; t=2&olog-y.