824 



Professor Dewar 



[April 1, 



fraction of the space filled by the solid did not exceed one-half. 

 The effect of any considerable inequality in the thickness of the non- 

 conducting powders was ascertained by comparing two vacuum tubes, 

 one having double the thickness of vacuum space of the other, and 

 each then filled with precipitated silica. Taking the unfilled vacuum 

 tube as the unit for comparison as described above, then the single 

 thickness of silica increased the insulation to 6 and the double 

 thickness to 8. The following table contains the results of a number 

 of experiments with triple sets of double-walled tubes filled with 

 different substances, when exhausted and unexhausted. The results are 

 expressed in the relative times required to volatilise the same small 

 volume of liquid air from each tube. This is most readily done, 

 after filling each tube with the same volume of liquid air, by noting 

 the time required to fill a given vessel standing over the pneumatic 

 trough with the gaseous air distilling off. 



In each triple set the unit taken for comparison is the time value 

 of the free vacuum spaced tube. 



Empty Tube 1 I'O 



Charcoal 5 0'7 



Magnesia 2 0*6 



Vacuum. Air. 



Empty Tube 1 1-0 



Lampblack 5 07 



Silica 4 0-7 



Vacuum. 



Empty Tube 1 



Graphite 1"3 



Alumiua 3' 3 



Vacuum. 



Empty Tube 1 



Lampblack 4 



Lycopodium 2*5 



Empty Tube 

 Calcium carbonate 

 „ fluoride 



Vacuum. 

 . 1 

 . 2-5 

 . 1-25 



Empty Tube . . 

 Barium carbonate 

 Calcium phosphate 



Vacuum. 



Empty Tube 1 



Phosphorus (amorphous) . . 1 

 Mercuric iodide 1*5 



Empty Tube 

 Lead oxide 

 Bismuth oxide 



Vacuum. 

 . 1 

 . 1-3 



. 2-7 



Vacuum. 

 . 1 

 . 2 

 . 6 



From these experiments it will be seen that silica, charcoal, 

 lampblack and oxide of bismuth all increase the insulation to 4, 5, 

 and 6 times that of the empty vacuum space. In tubes generally 

 which did not reach such a high vacuum the relative insulating 

 effect of these powders could be raised as much as 1 to 8 or 1 to 10. 

 In this case the influx of heat per unit of time in the vacuum tube 

 which did not contain any finely divided powder was necessarily 

 much greater. As the chief communication of heat is by molecular 

 bombardment the fine powders must shorten the free path of the 

 gaseous molecules, and the slow conduction of heat through the 

 porous mass must make the conveyance of heat energy more difficult 

 than when the gas molecules could directly impinge upon the outer 



