NATURAL PHILOSOPHY. 105 



it is peculiarly advantageous, as a burner accidentally put out by a puff of 

 wind, is instantly lighted again. An improvement on this invention has 

 been patented, March, 1858, by the original inventor. It consists in placing 

 the platina coil by the side of the burner, instead of above it, and in the 

 flame. The use of platina, though very costly, is necessary, as it is the 

 cheapest metal which does neither melt nor burn under the circumstances 

 described. The apparatus, thus improved, has been lately applied to the 

 1,500 burners of the Senate Chamber in Washington, and is said to give 

 complete satisfaction. 



GATCHELL'S LIGHTNING RODS AND POINTS. 



A committee of the Franklin Institute recommend the following improve- 

 ments in the construction of lightning rods, introduced by Mr. J. L. Gatchell, 

 of Maryland. These improvements consist, first, in the use of a rope of 

 twisted copper wire (containing eighteen strands of wire, about J^. inch in 

 diameter), by which are gained greater conducting power, freedom from 

 breaks or joints in the conductor, and a flexibility which allows it to be 

 adapted to any irregularities of form over which it may be carried. 



The second modification is the substitution of a copper for a platina 

 point, and the increasing the angle of the point, so as to approach that 

 recommended by the Committee of the French Academy of Sciences. The 

 advantages here gained are, greater conducting power by the substitution of 

 copper for platina, and secondly, a counteraction of the liability to fusion by 

 rendering the point much less acute. The preservation from, oxidation is 

 entrusted to a zinc ball attached below the point. 



ON THE ELECTRICAL LIGHT. BY H. W. DOVE. 



The experiments, in connection with the results of the prismatic investi- 

 gation of the spark, appear to me to lead to the following conclusion : 



A wire, becoming red-hot by heat, is first red, then orange, and lastly 

 white; so that it behaves like the combination of light which is obtained 

 when a screen is drawn away from the spectrum concealed by it, in such a 

 way that the red end first becomes visible, and to this the violet is finally 

 added. The increase of brilliancy, from the slightly luminous brush to the 

 bright spark, behaves quite otherwise. In this case, it is as if the screen re- 

 moved first set free the violet end, and then the other colors. This distinc- 

 tion of itself renders it improbable that the phenomena of electrical light, in 

 the state of less brilliancy, can be ascribed to a gradually increasing ignition 

 of solid particles. They rather resemble the weakly luminous flame of hy- 

 drogen, which becomes white by solid ignited carbon in the so-called gas- 

 flames, or by other solid matters, as in the Drummond light. The true 

 electrical light is produced at great distances in the surrounding, isolating, 

 aeriform medium, when the latter is attenuated. With this colored light be- 

 longing to the strongly refrangible part of the spectrum, phenomena of 

 ignition may be combined, by particles torn away from the positive and neg- 

 ative bodies. If these particles be only at a red heat, the impression of a 

 violet light is produced by their mixture with the electric light. To this class 

 belong the column of light in the electrical egg, and the basal point of the 

 brush, and, lastly, the indented reddish sparks of an electrical machine, at 

 distances to which a white spark does not pass. If particles at a white heat 

 come together, the whole is white, as in the sparks of Leyden jars; in oppo- 



