58 Mr. W. Makower on the Molecular 



graduation was measured by completely filling the diffusion- 

 vessel with mercury, and then running out the mercury to 

 the first graduation and weighing ; for the first vessel used 

 this volume was found to be 5*0 c.c, and for the second 

 vessel 12*4 c.c. The upper end of the diffusion-vessel was 

 drawn off and closed by a dense porous plug of plaster-of- 

 Paris (P), 1 cm. in diameter and *5 cm. thick, fitting tightly 

 into the glass*. Just below the porous plug was a three- 

 way capillary stopcock, T, provided with a syphon-pipette 

 for collecting gases over mercury when desired. The lower 

 extremity of the diffusion-vessel was connected by stout 

 rubber pressure-tubing with a cylindrical mercury reservoir 

 R of 3 cm. diameter which could be raised or lowered at 

 will. By means of a short piece of rubber pressure-tubing 

 the diffusion-vessel was connected to a cylindrical brass 

 vessel Y, along the axis of which ran an insulated brass wire 

 a connected to earth through the key k, by raising which 

 the wire a and its connexions could be insulated ; a difference 

 of potential of about 400 volts was maintained between a 

 and the outside of the brass vessel by a battery of 200 small 

 storage-cells S, which gave a field sufficient to saturate the 

 air in Y. All the connexions were carefully shielded from 

 external electrostatic effects by inclosing them in earthed 

 conductors. 



The emanation was obtained from a solution of radium 

 bromide in water. By lowering the reservoir R, a stream of 

 air could be carried through the solution and thence through 

 cotton- wool to remove dust, and through a small sulphuric- 

 acid bubbler along the capillary c into the vessel A. 



To carry out an experiment, a certain volume of air mixed 

 with emanation was sucked into A and well mixed by raising 

 and lowering the reservoir R several times in succession. 

 The volume of air in A was then read to *1 c.c. and a known 

 volume carried into Y. It was important not to force any 

 air through the porous plug during this operation, and there- 

 fore to prevent this the pressure of the air in Y was reduced 

 by means of a water-pump attached at e, after which the 

 stopcock / was closed. The stopcock T was then turned so 

 as to allow a certain volume of gas to pass into Y, the 

 mercury reservoir R being meanwhile raised and the level 

 of the mercury in A and R finally adjusted to equality. 

 Care was always taken that the pressure in Y had been so 

 far reduced, that at the end of this operation the pressure was 

 still below atmospheric. By turning the tap t communication 



* The plaster-of-Paris was inserted while still moist, and after it had 

 been compressed was allowed to set in position. 



