634 
Proceedings of the Royal Society 
best method of causing the walls of the tube to acquire the tem- 
perature of the air, and at the same time prevent heat passing from 
the outer case to the thermometer, would be to surround the bulb 
with a tube of some open or porous material, such as cloth, through 
which the air could be made to pass, and to cause the air to enter 
one end of this tube, and to pass radially through its walls in the 
manner shown in Plate XXIII. fig. 1, where a is the outside case 
heated by radiation, t the thermometer, and c a tube of some porous 
material surrounding the thermometer and placed inside the draught 
tube as shown. The arrows in the figure show the direction of the air 
currents in the draught tube under these conditions, from which 
it will be seen that the air currents are made to flow in a direction 
the opposite of that in which the heat tends to move ; and it was 
expected that the outward-moving air currents would check the 
inward movement of the heat, and prevent it arriving at the thermo- 
meter; also that the passing air would cause the inside surface 
of the tube rapidly to acquire the temperature of the air, and to 
follow its changes quickly. 
The trials with this arrangement were quite successful. The 
advance of heat was so perfectly held in check by it that the 
thermometer kept at the temperature of the air, even when the 
surrounding tube was highly heated. As these experiments may 
have some interest, I shall give somewhat in detail the results of 
some experiments made with some different materials used for this 
purpose. As it was winter, and no experiments of this kind were 
possible in the open air, a draught screen was prepared for making 
the trials in the laboratory. This apparatus is shown in fig. 2 : 
a is the case surrounding the thermometer t ; c is the draught tube, 
heated by a gas jet d, for promoting a circulation of air; e is a 
screen placed below the gas flame to prevent heat being radiated 
downwards to the thermometer case. The air enters the lower end 
of the case, and is drawn upwards by means of the column of heated 
air in the draught tube c. As it was necessary to put this method 
of protecting the thermometer to as severe a trial as possible, the 
outer case a, surrounding the thermometer, was jacketed by the 
case f which could be kept hot with steam. The thermometer 
could be thus exposed to radiation from a surface all round it having 
a temperature of 100° C. 
