332 CARNEGIE INSTITUTION OF WASHINGTON. 



to account for the ionic content over the ocean. The radioactive material in 

 the air over the land and in the soil is sufficient by itself to account for more 

 than the ionization as ordinarily observed there with the Ebert ion-counter, 

 so that there is a surplus to account for the existence, over the land, of a large 

 number of the so-called "Langevin ions," which on account of their small 

 mobility are not measured with the Ebert apparatus. 



The part of the paper dealing with the origin and maintenance of the 

 Earth's charge was, in its essentials, the same as the author's paper published 

 in Terrestrial Magnetism, volume 20, pages 105-120, 1915, and abstracted 

 on pages 339-341 of the 1915 annual report, published in Year Book No. 14. 



Atmospheric-electric observations made aboard the Carnegie. W. F. G. Swann. (Pre- 

 sented before Section II, Second Pan- American Congress at Washington, January 3, 

 1916.) 



The paper commenced with a brief review of the atmospheric-electric 

 problems awaiting solution, and a statement of the reasons which cause a 

 peculiar interest to attach to ocean observations. A description was then 

 given of the atmospheric-electric equipment aboard the Carnegie on her 

 fourth cruise. The paper concluded with a discussion of the 1915 results of 

 the Carnegie's fourth cruise. (See pages 326-329). 



On the ionization of the upper atmosphere. W. F. G. Swann. Terr. Mag., vol. 21, pp. 1-8, 

 1916. 



From various points of view there are indications that the upper atmosphere 

 is to be treated as a region of relatively high electrical conductivity. In so 

 far as ultra-violet light is frequently considered as one of the main ionizing 

 agencies in the upper atmosphere, it is of interest to calculate the amount of 

 ionization to be expected from this source when the Sun is treated as a black 

 body. 



Applying the Planck radiation fomiula, and assuming, in accordance with 

 the experiments of Hughes, that, for air, ionization does not set in below 

 wave-length 135 /z/x, it appears that only about 1.6X10"^ of the total solar 

 radiant energy is available for atmospheric ionization. 



In an example quoted by Schuster in connection with his theory of the 

 diurnal variations of terrestrial magnetism it is shown that, if the upper 

 atmosphere is treated as a shell 300 km. thick, at a pressure of 1 dyne per 

 square centimeter, a conductivity of 10~^^ e.m.u. would have to exist in 

 the shell in order to account for the magnetic effects under discussion. If 

 the conductivity be taken as 2ev{q/a)^, where q is the rate of production of 

 pairs of ions per cubic centimeter, a the coefficient of recombination of the 

 ions,e the electronic charge, and v the specific ionic velocity, it becomes possi- 

 ble to calculate X if q, v, and a, are known. Assuming the value hv for the 

 energy necessary to produce an ion, h being Planck's constant and v the 

 frequency corresponding to the wave-length 135 mm> q can be calculated for 

 a case where the ionization is uniform throughout the 300 km. layer, and is 

 all due to the ultra-violet light. Assuming further that the laws governing 

 the variations of a and v down to the pressure and temperature of the 300 

 km. layer are the same as those deduced from laboratory experiments, the 

 conductivity to be expected in this layer as calculated in this manner, 

 is only about 10~^ of the value 10~^' e.m.u. required for the magnetic 

 problem. Even if the whole of the Sun's energy could be utilized in produc- 

 ing ionization, the conductivity accounted for would still be far too small. 

 Any departure from a condition of uniform ionization in the layer would act 

 in the direction of reducing still further the conductivity to be accounted for 

 by the ultra-violet light. 



