1887 .] 
Mr J. Aitken on Thermometer Screens. 
431 
plan generally adopted of taking the readings of a thermometer 
placed on the grass is open to many objections. The temperature 
indicated by a thermometer so placed is affected by the greater or 
less amount of heat communicated upwards from the ground, by the 
amount of air circulating just at the place where the bulb happens 
to be, and other local influences, to many of which the ball at four 
feet from the ground is not exposed. By changing very slightly the 
position of the bulb of the thermometer placed on the grass, we 
can greatly alter its readings, whereas this is not the case with the 
ball at four feet from the ground. In the large ball used here there 
are fixed two tubes in a horizontal position. One of these tubes 
holds a maximum, the other a minimum thermometer. 
Returning to the table showing the difference in the readings of 
the Stevenson and the C screens. In the fourth column will be 
found the effect of solar radiation in heating the 40 cm, ball. The 
figures given are not the temperatures given by the ball, but the ex- 
cess of this temperature above that of the air, or, in other words, it is 
the heating effect of the radiation. For instance, on the 23rd of July 
the temperature of the ball was 84°, the air was 64°, and the heating 
effect of the sun was thus 20° as entered. In the fifth column are 
the solar radiation temperatures taken by the black bulb in vacuo. 
These temperatures are treated in the same way as those given 
by the black ball ; the figures show how much the thermometer was 
heated above the temperature of the air. 
It will be noticed that though the error of the Stevenson screen is 
due to radiation, yet it follows the indications of neither of these radia- 
tion thermometers. This is quite to be expected, because the error of 
the screen is the effect of radiation as modified by wind. Though the 
readings of the ball give the effect of radiation as modified by wind, 
yet these readings alone do not tell us how much they are affected 
by wind. For instance, the black ball might be heated to only a 
small amount, either by a strong sun checked by a strong wind, or 
by a feeble radiation unchecked by wind. If, however, we compare 
the temperatures of the black ball with those of the black bulb in 
vacuo , we at once see the effect of the wind. FTo two of these readings 
for any day bear the same relation to each other. When there was 
much wind, the black ball was heated only to about 04 times the 
temperature of the black bulb ; whereas when there was little wind 
VOL. XIV. 28 / 1/88 2 E 
