In Sir Humphry Davy's experiments with his arc light he was led to believe that the light between the two points of carbon would be produced even in an absolute vacuum, if it were possible to create one. Several scientists at the time disputed this contention, and M. Masson, Professor of Physics in the _École Centrale des Arts et Manufactures_ in Paris was particularly active in combatting the idea, maintaining that the arc had the same cause as the electric spark--the transport by electricity of the incandescent particles of the electrodes through the atmosphere. It was certain, at any rate, that no light was produced when the opposing carbons were brought into contact with each other, or were, on the other hand, separated too widely; and since there was a constant wearing away and shortening of the points, and thus a constantly increasing space between them, the great difficulty in making a practical lamp lay in regulating this distance automatically. It was finally accomplished, however, by the invention of a Russian officer, M. Jablochkoff, in 1876. The "Jablochkoff candle," as his lamp was called, marked an epoch in the history of electric lighting. One great merit of this invention was its simplicity, and while it has long since gone out of use, having been superseded by still simpler and better devices, it must always be recalled as an important stepping-stone in the progress of artificial illumination. The name "candle" for Jablochkoff's lamp was suggested by the fact that the two carbons were placed side by side, instead of point to point, the light at the top thus suggesting a candle. Between these two carbons, and extending their whole length except at the very tips, was an insulating material that the arc could not pierce, but which burned away at a rate commensurate with the shortening of the carbons. In this manner the points were kept constantly at the proper distance without regulating-machinery of any kind. This ingenious apparatus had the additional advantage that it could be placed on any kind of a bracket or chandelier that was properly wired, thus dispensing with the cumbersome frames and machines of the point-to-point carbon arc lights then being introduced. One difficulty at first encountered in using the Jablochkoff candle was the starting of the voltaic arc. In doing this it was necessary that contact be made between two carbon points, whether they lie parallel or point to point, and the necessary slight separation for producing the light effected later. To accomplish this Jablochkoff joined the tips of the carbons of his candle with a thin strip of carbon, which quickly burned away when the current was turned on, leaving the necessary space between the points for the arc.