1920] on Low Temperature Studies 249 



(4) Action of a large charcoal thermoscope. Instead of the small 

 1 gramme charcoal bulb (Fig. 1) and projected manometer shown 

 on June 8, 1906, a cylindrical bulb A (Fig. 3), containing 30 grammes 

 of charcoal, was connected to a manometer B, of which the limbs 

 were 2 feet high. The bulb A was immersed in liquid air to saturate 

 the charcoal to atmospheric pressure with clean dry air. A 3-way 

 stopcock C at the top of the charcoal bulb A gave connection either 

 to the manometer or to a liquid air trap (not shown) to clean the air 

 before being condeused. An annular tube E, open at the bottom, 

 covered the bulb, and was connected airtight above by rubber tubing. 

 An opening with a stopcock F led to a press-ball, by means of which 

 the liquid air could be forced out from the annular space round the 

 charcoal bulb. The vacuum vessel (r, in which the arrangement was 

 immersed, was silvered half-way round, the remainder being clear ; 

 hence, when rotated, it screened off the radiation directed upon the 

 charcoal from the front. 



A dispersed beam 3 feet broad, from an arc lamp a yard away, 

 was used to illuminate and actuate the thermoscope. AVith the 

 silvered half of the vacuum vessel towards the lantern, and C turned 

 to connect A to the manometer, the movement of the liquid was 

 scarcely perceptible ; but when the charcoal was exposed through the 

 unsilvered glass, a displacement of 30 cm. in 15 seconds resulted, 

 the annular space being full of liquid air. When this was emptied 

 by the press-ball connected to F, the displacement in the same time 

 was over 60 cm. 



Measurements of Transmissive Powers. 



In the cell employed for laboratory measurements the sensibility 

 depends finally on the degree of relative isolation between the 

 saturated charcoal and the liquid air or oxygen of the bath. If the 

 charcoal is in close contact with the cell walls, the absorbed radiation 

 is unable to raise the temperature of the charcoal sufficiently to affect 

 the manometer to any extent ; in the other extreme, not only are 

 the initial displacements more erratic, but the time taken to re- 

 equilibrate is too great. An arrangement that is very sensitive and 

 quite workable is to lower the level of the liquid in the bath below 

 the cell, but the space above the liquid must then have a very small 

 temperature gradient. This condition can be secured by efficient 

 cooling of the enclosing walls, as in the gas or air-cooled cell (see 

 below). 



The principal application of the arrangement was for measuring 

 the relative absorptions of infra-red radiation up to 100° C, exhibited 

 by a great number of substances when immersed in the form of 

 plates of various thicknesses in the liquid air above the cell. The 

 proportionate absorption was determined by two successive readings 

 with the shutter raised for \ to 1 minute to expose the cell to a 



