of Edinburgh, Session 1885 - 86 . 
635 
In experimenting on this method of protecting the thermometer, 
different materials were used for making the porous tube ; the 
results will be here given of the trials with muslin, wire-cloth, and 
cotton wool. The material under trial was rolled up in the shape 
of a tube and placed inside the case of the draught tube, as shown 
at g , and the protecting power of one, and of a number of thick- 
nesses of the different materials was measured. 
As the question to be determined was, how much the thermo- 
meter surrounded by the hot case was heated above the tempera- 
ture of the air when protected in this manner by the different 
materials, it was necessary that the temperature of the air should 
be very accurately determined. In conducting the investigation, it 
soon became evident that a thermometer hung up in the room 
could not be relied upon for this purpose, as it was too much 
affected by radiation, from the fire, from gas flames, and from the 
body of the observer ; for the amount of difference required to be 
measured was not degrees, but fractions of a degree. Another 
draught screen was therefore prepared, and a thermometer placed 
in it. The thermometer in this screen was well protected by 
concentric tubes ; a gas flame placed close to the case had no 
effect on it, and all radiation from below was easily checked. 
The draught was kept up in this screen, as in the other, by means 
of a jet of gas in the upper part of the tube. The temperature of 
the air as given by this screen could be relied upon to a small 
fraction of a degree. In the trials, the two draught screens were 
placed as near each other as convenient, generally about one-third 
of a metre apart. 
The thermometers used in these trials were graduated to Centi- 
grade degrees. The scale of both of them was very wide, and each 
degree was divided into tenths — the scale of the one showing 
the temperature of the air was 4*2 mm. to the degree, and the 
other 5 ’5 mm. to the degree. The thermometers were carefully 
compared with each other in water, and corrections made for 
difference. 
In experimenting with this arrangement of apparatus, the gas 
was lighted in both draught tubes to produce a circulation of the 
air. In the screen under trial a Ho. 1 jet was used turned down 
to one quarter of full flame, and kept as constant as possible 
