186 



PROFESSOR H. B. DIXON AND MR. E. C. EDGAR 



The weights employed were a brass hectogramme and its subdivisions to a gramme, 

 and, for the submultiples of a gramme, small platinum weights. The hectogramme 

 was taken as the unit and the separate weights were carefully compared with it. 

 Since all our measurements of mass were relative and not absolute, it was not necessary 

 to determine the absolute mass of our unit. In comparing the gramme of platinum 

 with the brass gramme marked Z. a correction was applied for the different weights 

 of air displaced by them. The values of all the weights are given below : 



VALUES of the Brass Weights. 



VALUES of the Platinum Weights. 



The palladium bulb, when charged with hydrogen and sealed off, varied in weight 

 from about 419 grammes to 425 grammes. It was counterpoised by a vessel made of 

 the same glass and of approximately the same volume, weighing 400 '00097 grammes. 

 The brass and platinum standardised weights were used to complete the equilibrium. 

 The only vacuum corrections necessary to apply to the weighings were (i.) that for 

 the difference in volume between the small weights used before and after the com- 

 bustion, i.e., the volume occupied by (approximately) 1 gramme of brass, and (ii.) for 

 possible changes in the buoyancy of the bulb. 



The glass counterpoise was made the same volume as the bulb when first used in 

 Experiment I. It was not considered necessary to alter it so as to make it exactly 



