1920] on Low Temperature Studies 251 



cell are shown at A. The metal capsule is covered by a transparent 

 rubber membrane through which can be seen the tray of charcoal. 

 This tray (shown separately at (B) below) was a tambourine of chiffon 

 gauze, stretched by a pair of light brass rings of square section. A 

 thin layer of clean dry absorptive charcoal— less than 1 gramme, in 

 pieces of about J mm. — was spread on the tambourine, which fitted 

 loosely into the capsule ; the charcoal was thus kept free in the cell 

 space. The shutter above the cell was a simple hinged brass lid 

 actuated by a light chain of silver or softened platinum wire, 

 running through eyes and over a small pulley above (not shown). 

 The lid, when closed, rested on a flat ring platform C, carried by a 

 sleeve fitting on a thin (1 mm. bore, \ mm. walls) german silver 

 tube leading out from the bottom of the cell. The plates of material 

 whose transmissivity was to be measured were placed on the 

 platform, the shutter being hollowed out sufficiently to allow this. 



The form of the cell employed for plates of silver chloride or 

 rock-salt is shown at S. The wall of the cell was a spherical segment 

 of pure assay lead about h mm. thick soldered to a brass disc as 

 base, from which the thin german silver tube led to the manometer. 

 The silver chloride (or rock-salt) plate was luted to the upper rim of 

 the lead, which was flattened to receive it. The interior of this cell 

 was occupied by an ebonite block, hollowed out above to contain the 

 charcoal, as in the metal cell, and with a small opening below, to 

 connect to the exit tube. The upper surface of the ebonite fitted 

 under the lead rim, without being luted to it, as the conti actions of 

 lead and ebonite differ so considerably at low temperatures that 

 otherwise there would be a tendency to crack the plate. Thus the 

 lead rim was left free to follow the contraction of the rock-salt 

 plate. 



In a later form of cell the lead shell was soldered to a brass ring 

 below, slightly coned to fit a similar ring on the periphery of the 

 brass base plate. The ebonite was then used only for the first 

 shaping of the lead rim, and the charcoal was supported in the lower 

 half of the cell, as in the metal capsule already described. This 

 arrangement allowed either charcoal or membrane to be altered as 

 required without mutual disturbance. 



The cell was immersed in liquid air (or oxygen or nitrogen, etc.) 

 in the inner of two vacuum vessels (shown at 1) and I) arranged in 

 the usual way for good isolation. The thin german silver tube led 

 from the cell to a wider tube in connection with the manometers E 

 and CI. Two three-way stopcocks («) and (b) were inserted on each 

 side of manometer E, which itself had a plain stopcock (c) closed 

 only when it was registering. The third connection of (a) was 

 sealed to a liquid air trap H, of the usual annular form, leading to 

 the supply of gas used to saturate the cell charcoal. With liquid 

 oxygen or old liquid air in I, room air was good enough for this 

 purpose ; but nitrogen, hydrogen and helium were all employed in 



