152 
MR. GLAISHER ON THE RADIATION OF HEAT, 
At noon on this day a thermometer was placed on grass upon which the sun had 
been shining till Sept. 21 d , 23 h 45 m , and the following observations were taken : — 
1843. 
Month, day, and 
hour. 
Reading of thermometer 
Excess of reading of 
thermometer in air 
above that on grass. 
Remarks. 
In air. 
On the grass. 
h 
m 
Sept. 22. 0 
0 
69-0 
67-0 
2-0 
A few clouds to the north. 
1 
0 
68-9 
64-5 
4*4 
Thin clouds to the north. 
1 
40 
68-7 
64-0 
4-7 
Cloudless. 
3 
30 
67-0 
61-7 
5-3 
Cloudless. 
4 
0 
66-6 
61-5 
5“1 
Cloudless. 
4 
20 
66*3 
61-0 
5*3 
Cloudless. 
4 
35 
65-8 
59-7 
6-1 
Cloudless. 
5 
0 
65-2 
59-0 
6*2 
Cloudless. 
During the whole of this day the temperature of the grass in the shade was several 
degrees below that of the air at four feet above it, and dew remained all day upon 
that portion of grass upon which the sun had not shone. Also the temperature of 
that upon which the sun had been shining soon began to decline, but it did not de- 
scend so low as that of the dew-point, which was about 58°, and therefore no dew 
was deposited upon it. 
This day followed a long period of hot weather without rain, and the day was 
nearly cloudless, with the sun shining brightly. 
This closes the experiments connected with grass with mercurial thermometers, and 
it will be seen that I have paid very much attention to them, as grass is a substance 
upon which many former experiments have been made, the discrepancies between 
which are now fully accounted for. 
I may here point out one important result of the preceding experiments; viz. the 
amount of the corrections dependent upon the portion of clear sky to which the 
bulb of a thermometer is exposed, necessary to be applied to its readings to obtain 
from them the true temperature of the air. As a thermometer at the height of four 
• © 
feet, protected from six-tenths of the sky, was found to read 0°*58 higher than 
another at the same height fully exposed to the sky, it follows that the elfect of pro- 
tecting the bulb of a thermometer from the sky, whose height from the ground was 
four feet, was to cause the readings to increase 0°T for every tenth part of the sky 
protected : hence, if a thermometer be wholly exposed to the sky, it is necessary to 
increase its readings by 
0°1 X number of tenths of cloudless sky ; 
and if the bulb be partially protected from the sky, the correction is 
flO-{ quantity ofsky ' n tentks f ,orn | \ xo 0, l X number of tenths of clear sky. 
V l which the bulb is protected J' 
From this it appears that the thermometer, which throughout all the experiments 
has been considered to have been protected from the effects of radiation, was yet 
subjected to it to the amount of 0 o, 4 on a cloudless sky, or to 0 o, 4 X number of tenths 
of clear sky at times when the sky was partially cloudy. 
