464 Perkins — Molecular Weight of Radium Emanation. 



were shrunk at D and D'. For outside jacket, A, a piece of 

 tubing of nearly 5 c,n outside diameter was squared off to exactly 

 the same length jas the distance between the shoulders D and 

 D'. Iron plates, K and K', with accurately ground surfaces 

 were made to bear at D and D' on the outside tube and the 

 ends of the outside jacket, thus forming a diffusion chamber. 

 Collars B and B 7 , threaded and shouldered inside, pulled the 

 plates against the ends of the outside jacket ; while collars E 

 and E', threaded on the inner tube, served a similar purpose at 

 D and D\ In order to bring down the size of the ends of the 

 inner tube, iron plugs, F and F', were threaded into them. 

 These were reamed out and ground to tit thin-walled iron tubes 

 l cm in diameter. One of these tubes communicated with a 

 hydrogen generator, while two others were alternately inserted 

 in F for collecting and condensing the diffused vapor. The 

 hydrogen generator was used for maintaining a slow current 

 through C to continually remove the diffused vapor from the 

 plug. 



Some early experiments showed that when mercury was 

 heated in air, red oxide of mercury began to appear at 275° C. 

 and probably at a slightly lower temperature ; in addition, small 

 quantities of iron scale were formed. By performing the experi- 

 ments in an atmosphere of hydrogen, all oxidation effects were 

 eliminated. Before entering the diffusion apparatus, the hydro- 

 gen was bubbled through bottles of copper sulphate to remove 

 hydrogen sulphide and then passed through a calcium chloride 

 drying tube. 



For the purpose of introducing mercury or emanation into the 

 diffusion chamber, small holes were bored through the collars 

 D and D' and still smaller holes through the plates about mid- 

 way between the outer and inner tubes. Into these holes in the 

 plates were threaded small iron caps G and G', closed by iron 

 plugs L and 1/ with conical tips. These iron plugs had slotted 

 heads and could be screwed into place, then set in with a 

 wrench. 



Some previous experiments with asbestos fibeivshowed it to 

 be especially adapted for use as porous plugs. When a mixture 

 of fine asbestos fiber and water was run through fine holes under 

 pressure, very uniform and porous plugs were formed of any 

 desired thickness, which seemed to suffer very little change in 

 porosity over wide ranges of temperature. 



To serve as a base for porous plugs of this kind, 40 small cup- 

 shaped holes were bored through the middle of the inner tube 

 C as follows : Small borings were made with a drill of nearly 

 3 mra diameter reaching almost through to the inside of the tube. 

 Through the remaining 0*2 mm , holes were bored l mm in 

 diameter. Before the plugs were formed, all the different parts 



