636 
Proceedings of the Royal Society 
during the trials. The first trials were made with the thermometer 
case as it came from the work, but without polishing the inside 
surface that radiated heat to the thermometer. A number of trials 
were made with it in this condition, of which it is unnecessary to 
give the detailed results. Ib may, however, be mentioned, that with 
the usual draught and with no protection between the thermometer 
and the walls of the case — that is, with the bulb exposed all round 
to radiation from bright tin at 100° — the thermometer rose to 18°, 
the temperature of the air being 12 ° ; that is, it was heated 6° above 
the temperature of the air. When one thickness of wire-cloth and 
two thicknesses of muslin were placed between the bulb and the 
case, the radiation was almost entirely checked, and the thermo- 
meter rose only about 0°‘2 above the temperature of the air. 
As it could not be expected that the inside of the case would keep 
bright after being long in use, it was thought advisable to make the 
tests under the most disadvantageous conditions possible. The 
inside of the case was therefore painted black. When this was 
done, a great increase in the heating effect on the thermometer was 
produced, as was to be expected. 
In the following tables are given the results of some experiments 
made with the inside of the case blackened, and with fine cloth as 
the protecting material. The protection effected by different 
numbers of thicknesses was tried, and the results are given. The 
cloth used in these trials was book-muslin, about 56 meshes to the 
inch. 
Muslin. 
1. 
Air. 
2. 
Case. 
3. 
Difference. 
4. 
Protection. 
13°-4 
31°'6 
18“ *2 
None 
13° 4 
20° -9 
7° - 5 
1 Thickness 
7* 1 
O 
CO 
17°-4 
4°-0 
2 
13°-25 
13°-6 
0°-35 
4 
13°-3 
13° -3 
o 
o 
O 
12 
In the above table, in column 1 is given the temperature of the 
air of the room; in column 2 the reading of the thermometer sur- 
