72 
Proceedings of Royal Society of Edinburgh, [jan. 17 , 
be observed, that as long as the readings were taken, the open one 
always read a little higher than the closed one. 
The curve C shows the temperatures given by the new screen, 
fig. 1. It will be observed that its readings are very much 
below those of either of the Stevenson screens. This difference 
varied from time to time, according to the amount of radiation, but 
it always remained during the whole time of the observations a good 
deal lower than the others. At 12.10 it attained its maximum 
difference, being at that hour 2° *3 lower than the ordinary Stevenson, 
and 0 o, 8 lower than the closed Stevenson. 
In addition to the observation shown on PI. III. a number of 
others have also been taken, but the particulars need not be given 
here. The general result has always been the same. The Stevenson 
screen with closed bottom always read lower than the open one, and 
the C screen lower than either, the amount depending on the radia- 
tion, wind, &c., at the time. The observations on PI. II. fig. 2, 
show the readings of the three screens on the 18th October. On this 
occasion the difference in the screens was not great. It will be seen 
that the open Stevenson only went 0°'5 higher than the C screen, 
while the closed one was only 0 o, 2 higher. This small difference 
was due to the lateness of the season, the small amount of radia- 
tion at the time, together with the strength and direction of the 
wind. The wind was fresh, and from the north-east ; it therefore 
entered by the cold side of the Stevenson screens, and tended to 
keep the side of the screen on which the sun was shining colder 
than if it had come from the opposite direction. 
Further, in addition to these observations, made at short in- 
tervals of time with the ordinary thermometers, the temperature 
has been taken almost every day in the three screens by maximum 
thermometers, and the result is always the same. On very few days 
do they all read alike ; only when there is stormy weather and dense 
masses of clouds ; at all other times there are differences, but the 
three screens always keep the same relative position, C being lowest, 
and the closed screen lower than the open. It may be mentioned, 
that in these trials with the maximum thermometers, the same 
instruments were not always kept in the same screens, but were 
generally changed every morning, to equalize any instrumental errors, 
and screens A and B were worked alternately open and closed. 
