$6 Prof. Chattock, Miss Walker, and Mr. Dixon on 



Our conclusions may be summed up as follows : — 



1. The values of V for point-discharge in gases are in approxi- 



mate agreement with those obtained for the same gases 

 when ionized by x-rajs. 



2. In non-conducting liquids V appears to be of the same 



order of magnitude as in ordinary electrolytes. 



3. In hydrogen for negative discharge only, and probably in 



turpentine for both kinds of discharge, a considerable 

 " back-discharge" of ions seems to occur from a brass 

 or platinum ring towards the point. 



-4. The amount by which this back-discharge lowers the 

 wind-pressures in the case of hydrogen is closely pro- 

 portional to the strength of the current from the point. 

 This is proved by the constancy of z in each group of 

 observations. 



$. Negative ions in constant -fields travel at very varying rates 

 in hydrogen ; the same being true of oxygen and air, 

 but to a much less extent. These variations are real, and 

 not attributable to errors of observation or other accidents. 

 Positive ions are not similarly affected to anything like 

 the same extent, if at all. 



6.. To account for this it is suggested that a certain number 

 of the negative ions may owe their origin to occluded 

 gases in the point : that this number may be variable 

 owing to changes in the aceessibiiitv of the occluded gases 

 and other causes ; and that the specific velocity of the 

 positive ions may be constant owing to their being formed 

 exclusively of the gas outside the point.. 

 In conclusion we wish to express our thanks to Mr. H. 



Strachan for much assistance in the work of taking the 



■readings, 



Note on a Sensitive Pressure- Gauge. 



As will be seen from the magnitudes of the wind-pressures 

 we 'have measured and the smoothness of the curves obtained, 

 our gauge was sensitive enough to indicate pressures of a few 

 ten-thousandths of a millimetre of water. As it presents 

 features which we believe to be new, it may be of interest to 

 describe it. 



It was necessary to nse a well-known combination of oil 

 and water to obtain the requisite sensitiveness. In fig. y 

 EBBE is a glass U-tube, about 4 centim. in diameter at the 

 wide parts and 1 centim. diameter at A. From centre to 

 centre of the limbs is about 13 centim., and from D to 8 

 about 25 centim. The low<er part of the tube is filled with 



