G88 
Proceedings of the Ployed Society 
through the indraught tube, and radiated to the thermometer ; and 
i:>art is due to the air which enters the tube having first touched the 
outside of the apparatus, and in this way got heated before arriving 
at the thermometer ; while the silvered thermometer is less heated 
by diffused radiation when exposed freely to the air than the 
thermometer in the fan apparatus. 
All these experiments have been made with the silver coating 
deposited on the bulb by chemical means. A comparison of man}" 
different experiments with different thermometers shows a variation 
in the protecting power of the different coverings due to their 
greater or less thickness, and also to their greater or less perfection, 
and freeness from scratches which remove the silver from the 
glass. In some of the experiments the thermometers had two and 
three coatings of silver put on them, by placing them in successive 
baths ; the surfaces of the bulbs being simply washed when taken 
out of one bath and placed immediately in the next. When taken 
from the final bath, some of the bulbs looked dull and dusty, but 
acted quite well, while others were improved by polishing with 
rouge. 
These chemically deposited silver coverings being rather delicate 
for practical purposes, the next thing to be done was to get a silver 
sheath prepared for the thermometer bulb. This sheath was made 
of thin sheet silver and fitted easily over the bulb, and it also 
covered part of the stem. The silver coating being dissolved off 
the thermometer bulb, the silver sheath was polished, and fitted on, 
and the thermometer tested as before with the fan apparatus. The 
result in this case was not so good as with the chemically deposited 
silver. The readings of the thermometer now almost exactly 
corresponded with those of the fan thermometer; that is, the 
silver sheath reduced the readings 1° lower than the clean glass, 
but gave about 0°‘25 higher than the deposited silver. These 
higher readings suggested that this sheath was not made of pure 
silver. On inquiry, I found this to be the case ; so another sheath 
was prepared, and special precautions taken to have the silver as 
pure as possible. 
This second sheath was tried on the morning of the 18th, when 
there was a bright sun and almost no wind. As in the previous 
experiments, the thermometer was simply placed under a horizontal 
