Capacity of Electrolytes, 349 



Conducting Cylinders. 



III. If the cylinder is a conductor and is surrounded by a 

 nonconducting medium, the lines of force are drawn toward 

 it, as lines of magnetic force are gathered up by a piece of 

 iron put into a magnetic field. The resultant force is always 

 one of attraction toward the electrodes. This follows from the 

 tendency of the lines of force to shorten, and being more 

 numerous on the side toward the origin, the conductor is drawn 

 in that direction. If the surrounding medium is a conductor, 

 the case is not modified providing the solid is a very much 

 better conductor. Thus, suppose the cylinder is carbon and 

 the medium water. The latter is a fair conductor, yet carbon 

 conducts millions of times better than does distilled water, 

 and the lines of force will be arranged sensibly the same as 

 though water were a nonconductor and carbon a perfect con- 

 ductor. The resultant force on the cylinder in a given field 

 will evidently be proportional to the specific inductive capacity 

 of the medium of the field. Tins is a much simpler case than 

 that of a dielectric solid in a liquid, where the conductivity of 

 the liquid and the specific inductive capacity of the solid 

 affect the result. As before, an accurate adjustment of the 

 position of the conducting cylinders is not necessary, and 

 using the same cylinders with the same difference (or known 

 differences) of potential, this would seem to be an excellent 

 method of measuring the specific inductive capacity of liquids. 

 Accordingly, cylinders of carbon (cut out of an arc-light 

 carbon) were mounted in the torsion-balance and the force of 

 attraction under various conditions measured. 



1. The force, as we should expect, is accurately propor- 

 tional to the square of the difference of potential for air and 

 all liquids, conducting and nonconducting. 



2. The force is approximately proportional to the volume of 

 the cylinder. This was found to be roughly true for three 

 pairs of cylinders of nearly equal length, but of very different 

 volumes, i. e., 0*73, 5*7, and 10*0 cubic centimetres respec- 

 tively. The force increased somewhat faster than the volumes, 

 which would be expected, as the larger cylinders extend 

 farther into the stronger parts of the field near the electrodes. 

 For small changes in the volume the force may be taken as 

 proportional to the volume. 



3. The following tables give series of measurements made 

 w T ith the second pair of cylinders (volume 5*7 c.c.) in air and 

 turpentine. Other measurements agreed with these almost 

 exactly. 



