48 
DR. S. CHAPMAN ON THE SOLAR AND LUNAR 
the sun, cannot be an important factor. Some form of corpuscular emission may be 
supposed to be responsible. Schuster, in his second memoir, showed that only very 
rapidly moving corpuscles could possibly be so regarded ; these would act as fertilizers 
in the absorbing layer by producing ions through collisions with molecules. Such 
corpuscles might be few in number compared with the total number of ions thus 
liberated ; but if they are supposed to be of both signs, their speed of transmission 
from the sun must be great in order that re-combination may not take place on the 
way, while, if they are of one sign only, the accumulation of charge in the earth’s 
atmosphere may present difficulties. The hypothesis therefore stands in need of 
numerical examination similar to Swann’s discussion in the case of ultra-violet 
radiation, but at present the necessary data for this are wanting. It is difficult to 
imagine further alternatives, however, and the existence of the conducting layer 
itself can hardly now be questioned. 
It may be mentioned that, since the intensity of the ionizing agent varies as the 
square of the resulting conductivity, the former must be from 100 to 150 per cent, 
greater at times of sunspot maximum than at times of minimum, the increase in the 
conductivity being from 35 to 60 per cent. 
The phenomena of electric wave transmission also afford evidence on the present 
subject, and some conclusions of Eccles # may be mentioned. Three strata of the 
atmosphere are proposed, the highest one (first suggested by Heaviside in 1900) 
being strongly and permanently conducting, while the lowest is permanently non¬ 
conducting. The middle layer, the lower surface of which was roughly estimated to 
be 50 miles high, is a conductor by day and a non-conductor at night, the transitional 
region being fairly definite. The magnetic phenomena discussed in this paper indicate 
a layer resembling the middle stratum in Dr. Eccles’ theory, but give no evidence of 
the higher, permanently conducting layer. 
Part YI.— The Theory of the External Solar and Lunar Diurnal 
Magnetic Variation Fields. 
§ 22. Outline of the Mathematical Theory for the General Law oj Atmospheric 
Conductivity. 
In Parts II. and III. of this paper it has been shown that the major portion of the 
solar and lunar diurnal magnetic variations is due to magnetic forces which possess a 
potential, and are therefore attributable to electric currents. These were found to 
be situated mainly above the earth’s surface, and in Part IV. the internal current 
system was shown to be probably caused through induction by the external current 
system. The remaining task involved in the explanation of the whole phenomenon 
consists, therefore, in accounting for the externally circulating system of electric 
* Eccles, ‘Rov. Soc. Proc.,’ A, vol. 87, p. 79, 1912. 
