SECTIONAL TRANSACTIONS,—A.t 353 
is estimated that in summer the resistance of a column reaching right up to the K.H. 
layer and with 1 cm.” cross-section varies between 2-2 x 107! ohms at 2h. and 3-3 x 10% 
ohms at 18h., whilst in winter the range is between 3-4 x 10%! ohms at 5h. and 8-7 x 102! 
ohms at 17h. The resistance increases steadily during the hours in which pollution 
is being produced. The estimates are of the right order of magnitude, but the ratios 
are of more significance than the exact figures. 
Potential gradient depends on the strength of the current and on the resistance 
of the air near the ground. The double oscillation of potential gradient in the course 
of the day is explained by the double oscillation in resistance. The specific resistance 
of the air has minima in the early morning and in the afternoon. The difference 
between the types of variation of total resistance of the atmosphere and of specific 
resistance close to the ground is explained by the fact that in the hours during which 
the ground is being warmed pollution diffuses to considerable heights. 
The high potential of the K.H. layer is attributed to the action of thunderstorms, 
but the mechanism by which thunderstorms maintain the circulation of electricity is 
not discussed in this paper. 
There is a well-known difficulty in evaluating the magnitude of the air-earth 
current. The vertical current in the free atmosphere is regarded as a pure conduction 
current, positive and negative ions moving in opposite directions. At the surface of 
the ground only the positive current is effective. The negative current is thought 
to be counterbalanced by the transport of space-charge by eddy diffusion. Such 
observations as are available for testing this hypothesis are considered in the paper, 
but the need for more observations is stressed. 
Dr. H. Jerrreys, F.R.S.—The New Seismological Tables. 
Friday, September 25. 
Dr. F. Scuiesincer.—The Use of Photographs of Very Wide Angular 
Field in Astronomy. 
Untilrecently, the largest plates used to determine astronomical positions covered 
four square degrees, Experiments made by the writer at Allegheny and at Yale have 
shown that plates up to 25 square degrees could be used to advantage, and that these 
are particularly adapted to compiling zone catalogues and deriving in this way the 
proper motions of many stars, with much less labour and considerably greater 
accuracy than is possible with the meridian circle. Such plates must be reduced by 
means of comparison positions for a few of the stars, especially observed for this 
purpose with the meridian circle, and experience has shown that the observation of 
these comparison stars constitutes about one-half the total labour of the whole work. 
Tf plates considerably larger than 25 square degrees could be used, this half of the 
labour would be saved, and a still further, though small, increase in accuracy would 
be expected. Experiments have, therefore, been made at Yale with plates up to 
_ 140 square degrees. It has been found that these yield results which are slightly 
_ superior to those covering the 25 square degrees. Plates of this very large size can 
_ be reduced without the use of special meridian observations, as the positions in the 
_ Standard Catalogues already published or soon to appear (like the General Catalogue 
_of Boss) are sufficiently numerous and sufficiently accurate to give good determina- 
tions of the plate constants, though the number of these constants is greater than 
with the smaller plates. 
: 
3 
‘Dr. H. Spencer Jonss, F.R.S.—Nova Picioris. 
_ Nova Pictoris was discovered on 1925 May 25. From that date until 1926 
March 13 an extensive series of spectrograms was obtained with the four-prism 
Spectroscope attached to the 24-inch refractor at the Cape Observatory. The 
spectrum from March 1926 to 1931 has been studied with the aid of objective prism 
ectra, obtained mainly at the Union Observatory, Johannesburg. The spectrum at 
iscovery was a strong absorption spectrum of type cF5, many of the lines having 
ission borders to the red; the emission was strongest at the red end of the 
spectrum. Broad emission bands appeared at about the time of the first maximum. 
The sequence of spectral changes was generically similar to those observed in previous 
1931 AA 
