1919-20.] 
Thermometer Screens. 
175 
absorption of radiant heat. Compared with the silver-sheathed fine-bulb 
thermometer, the gilded-bulb instrument read about 0 o, 5 higher. The 
next protection tried was gold leaf. The bulb and part of the stem were 
dipped into a thin solution of spirit varnish, taken out, shaken to get rid 
of the last drop, and then turned bulb up so as to have as thin a layer of 
varnish as possible. A piece of gold leaf previously prepared was placed 
on the bulb, and was at once drawn round it by the capillarity of the 
varnish. A blast of air immediately applied caused the creases to fold 
closer to the glass. After drying for two or three days, the gilded bulb 
was rubbed with a fine brush and the superfluous leaf removed. This 
gold-gilt thermometer when tested against the silver-sheathed fine-bulb 
instrument gave the same reading when the air temperature was fairly 
steady. It was not, of course, so sensitive to rapid changes. 
In the following tests the gold-gilt thermometer was used as a standard, 
since its thermal inertia was about the same as that of the other 
thermometers in use. The silver-sheathed thermometer was, however, 
always placed alongside as a check on the gold-gilt one. 
The thermal inertia of the Stevenson screen is very marked. It lags 
behind both in a rising and in a falling temperature, the enclosed 
thermometer reading lower in, the former case and higher in the latter than 
a freely exposed shaded thermometer. 
There is also increased thermal inertia in the maximum and minimum 
thermometers, which are usually constructed with heavy metal frames. 
These are nearly in contact with the bulbs, thereby checking the free 
circulation of the air. This adds needlessly to the thermal inertia, as direct 
testing clearly proves. Maximum and minimum thermometers should 
have the frames cut away all round the bulb to give free access of air. 
All screens must have more or less thermal inertia, but we should 
endeavour to keep this factor as low as possible and effect a reasonable 
compromise. The thing to be avoided is too great thermal inertia, especially 
when combined, as in the case of the Stevenson screen, with the “ bottling 
up ” of heat consequent on the long-continued action of radiation. 
As clearly established by prolonged observations, the simple double- 
board sunshade with the thermometer bulb freely exposed under it, as 
shown in fig. 1, is more efficient than the Stevenson screen, besides being 
very much simpler in construction. It may, however, be greatly improved. 
Thus radiation from below can be intercepted by a small horizontal screen 
a placed under the bulb, as shown in fig. 2 ; and secondary radiation 
from the lower surface of the shade proper can be intercepted by a small 
screen, similar to a, and placed above the bulb. Finally, the radiation 
