Prof. G-. Quincke on Electrical Expansion. 33 



16. With the same difference of electric potential, 



the increase in volume — is, 

 for the same kind of glass and the same thickness, 

 three times the increase of length -p 



17. The same result was obtained by simultaneous measure- 

 ment of the increase in volume and the increase in length of a 

 thermometer condenser with a long cylindrical bulb. The 

 bulbs of the thermometer condensers were very uniform tubes 

 of flint-glass, 800 to 1900 millims. in length and 0'362 to 

 0*621 millim. in the thickness of the glass. The bubble of the 

 spirit-level of the contact-lever was observed with a micro- 

 scope and micrometer-eyepiece, which permitted the measure- 

 ment of an increase in length of 0*000008 millim. 



18. The relationship — = 3-r- is not consistent with the 



' r v I 



hypothesis that the thickness of the glass is diminished by the 

 attraction of the unlike electricities on the coatings of the con- 

 denser, and that the volume of the bulb of the thermometer 

 is thus increased indirectly by "electric compression." 



19. Electric expansion takes place equally in all directions, 

 in the same way as thermic expansion. The hypothesis that 

 electric expansion results from heat caused by feeble electric 

 currents traversing the glass between the two coatings is nega- 

 tived by the facts stated under 6, 8, and 15. 



20. The electric expansion of glass is most simply shown 

 by the use of tubes with excentric bore, in which the glass is 

 thicker on the one side than on the other. Such a tube is 

 curved, after cooling, with the thinner wall on the convex side, 

 since the thicker wall cools more slowly and contracts more 

 forcibly than the thinner wall. Such a glass thread, closed 

 at the lower end, filled with water, and immersed in a deep 

 vessel of water, forms a " glass-thread electrometer," of which 

 the water within and without the tube serves for the coatings. 



When these coatings are placed in contact with the coatings 

 of a charged Ley den jar, the thread curves still more, since 

 the thin wall is more expanded by the electric force than the 

 thick wall. The deflection of the glass-thread electrometer 

 may amount to several millimetres, and may be conveniently 

 observed by means of a horizontal microscope with micrometer 

 eyepiece. 



Upon discharge, the end of the bent glass thread returns to 

 its original position immediately with flint-glass, more slowly 



Phil. Mag. S. 5. Vol. 10. No. 59, July 1880. D 



