1887 .] Mr John Aitken on Thermometer Screens. 
71 
proportion to the excess of its temperature above that of the air, the 
lowest curve — namely C, in the 7th October observations — is taken 
as the base line, and the excess of the temperature of the grass above 
C is marked off. For instance, at 11.30, the temperature of the air 
as given by C was 57°, while the grass was 70°, or 13 degrees above 
C. Read in this way, the G observations show us that at 12.10 the 
grass was 18 ‘5 degrees above the temperature of the air. 
No curve is given of the silvered-bulb observations, though they 
were taken at the same time. The reason for this is that they would 
only confuse the figure, as they were practically the same as those of 
the screen C, only they were sometimes a little higher and at other 
times a little lower than C, owing to the smaller inertia of the 
silvered bulb. 
The day on which this trial was made was fine, with some wind 
from the south-west till near mid-day, when it fell, the air was clear, 
and there were a few passing clouds. It w T as not thought advisable 
to take readings till near 11 o’clock, as the Stevenson screens were 
quite wet in the morning, and tended to read low, owing to the 
evaporation. At 10.45, when the readings were begun, the bottom 
of screen A was open, and B closed. Readings were taken with the 
screens in this condition till 11.45. After which screen A was closed, 
and B opened. At 12.30 the condition of the screens was again 
reversed, B being closed and A open. 
It will be noticed that from 10.45 to 11.45 there was not much 
difference between the open screen A and the closed one B, even 
though the radiation effect was strong, as will be seen from the 
amount of sunshine indicated by the black area below the curves 
and by the curve G. The maximum difference between A and B 
amounted to only 0 o, 5 ; this was probably due to the strongish 
wind blowing at the time. After 11.45, when the bottom of A was 
closed, and B opened, the lines connecting the temperatures cross 
each other, and B now, instead of being the lower, becomes much 
the higher, the open screen showing a maximum difference of 1°‘5 
at 12.10. This great error in the readings of the open screen was 
doubtless due to the wind dying away at this hour. After the 
bottoms were again reversed, the lines again cross each other ; but as 
the sun did not again come out, and the radiation curve G fell very 
low, there was not much difference in the two screens ; but it will 
