28 Proceedings of Royal Society of Edinburqh. [dec. 6, 
edge of the sides, they can only receive and transmit radiation from 
and to that part of the ground that would he visible to an eye 
placed where the bulbs are ; and that in theory the greater distance 
of the upper thermometer from the ground would be exactly 
counterbalanced by the larger area capable of acting on it, and the 
radiation effects would be the same in both cases. The thermome- 
ters were so mounted that the difference in the size of the screens 
made little difference in the exposure to ground radiation. If the 
screens themselves get heated and cooled by ground radiation, they 
would correspondingly heat and cool the air as it passed through 
the louvres, and thus affect the thermometers. In this latter case 
the heating and cooling ought to be inversely as the square of their 
distances from the ground — that is, the lower screen should be 
affected nearly six times as much as the upper. 
The afternoon difference may also be due to convection currents 
arising from the heated ground; these would affect the lower 
thermometer more than the upper, but it is difficult to see how 
any such cause could give the morning difference, in which the 
upper thermometer reads above the lower. It is possible, however, 
that in fine weather the layer of air next the ground is so much 
cooled by contact with the ground that there is a continuous 
gradient of temperature rising with height above ground, at least 
as high as the level of the upper screen. I hope to repeat the 
experiment under more favourable conditions of weather, and also, 
if possible, when the ground is covered with snow. 
PRIVATE BUSINESS. 
Messrs Asutosli Mukhopadhyay, M.A., <fcc., Joseph James Cole- 
man, and John James Burnet were balloted for, and declared duly 
elected Fellows of the Society. 
