FiQ. 17. 



1890.] on Problems in the Physics of an Electric Lamp. 49 



any vacuum, is not necessary for the production of the " Edison 

 effect." Here is a carbon horseshoe-shaped conductor, not enclosed 

 in any receiver (see Fig. 17). Close to the negative leg or branch, 

 yet not touching it, we have 

 adjusted a little metal plate. 

 The sensitive galvanometer 

 is connected between this 

 metal plate and the base of 

 the other or positive leg of 

 this carbon arch. On sending 

 a current through the carbon 

 sufficient to bring it to bright 

 incandescence, the galvano- 

 meter gives indications of a 

 current flowing through it, 

 and as long as the carbon 

 endures, which is not, how- 

 ever, for many seconds, there 

 is a current of electricity 

 through it equivalent to a 

 flow of negative electricity from the plate through the galvanometer 

 to the positive electrode of the carbon. The interposition of a thin 

 sheet of mica between the metal plate and the negative leg of the 

 carbon loop entirely destroys the galvanometer current.* 



These experiments and brief expositions cover a very small 

 portion of the ground which is properly included within the limits 

 of my subject. Such fragments of it as we have been able to explore 

 to-night will have made it clear that it is a region abounding in 

 interesting facts and problems in molecular physics. The glow- 

 lamp and the electric arc have revolutionised our methods of 

 artificial lighting, but they present themselves also as subjects of 

 scientific study, by no means yet exhausted of all that they have to 

 teach. 



[J. A. F.] 



Edison effect " experiment shown with 

 carbon in open air. 



♦ This last experiment is due to my assistant, Mr. A. H. Bate. 



Vol. XIII. (No. 84.) 



E 



