110 Dr. Gr. A. Liebig on the Electrostatic Force 



balance ; and g the force of gravity at the place in which the 

 observations are made. 



In order to determine whether the method of finding when 

 the plates were in contact, by observing when the sparks just 

 disappeared, was an accurate one, the plates were connected 

 with a battery and galvanometer, and alternate readings taken. 

 As a result, it was found that the former method gave, in 

 general, values about - 5 -Jq millim. too small, though they were 

 occasionally too large ; but in any case the difference was 

 small enough to be neglected. 



As the plates between which the sparks passed were sphe- 

 rical, a correction to the measured difference of potential was 

 rendered necessary in order to reduce the values to plane 

 surfaces. 



In the preliminary experiments a considerable difficulty 

 was encountered in measuring the difference of potential for 

 sparks of small length when the plates were close together. 

 Unless the machine was run very slowly, the sparks were given 

 off in such rapid succession that the suspended disk had not time 

 to swing back to zero before a new impulse acting on it drew 

 it downward again. Even the damping effect of that part 

 of the metal case which is above it cannot prevent this ; and a 

 reading taken under such circumstances would unquestionably 

 give values very much too large. However, by turning the 

 Holtz machine very slowly indeed, and by increased expe- 

 rience, errors of this kind were finally avoided. 



Results. 



The results of the experiments are given in Tables I., II., III., 

 and IV., and the curves corresponding to them on Plate II. In 

 each diagram one curve gives the distance and total electro- 

 motive force, and one the relation between distance or length 

 of spark and electromotive force per centimetre. The values 

 are in all cases in C.G.S. electrostatic units ; and about 150 

 observations were used to plot each of the curves. In order 

 to show the form of the curve for the total electromotive force 

 more plainly, it was found desirable to construct this on a 

 scale three times as large as that used for the force per unit 

 length. The ordinates are therefore to be measured according 

 to one of two scales, both of which are constructed on the 

 Plate. 



