FteBRUAET 4, 1910] 



SCIENCE 



167 



cateSj oxides, etc., show characteristic emis- 

 sion bands. 



One of the most attractive fields of arti- 

 ficial light production has long been that of 

 luminous gases or vapors. It has seemed 

 as though this ought to be a most satisfac- 

 tory method. The so-called Geissler tubes 

 in which light is produced by the electrical 

 discharge through gases at low pressure are 

 familiar to all. The distribution of the 

 energy emitted from gases is still further 

 removed than that of solids from the laws 

 of a black body, and a large proportion of 

 the total electrical energy supplied to a 

 rarefied gas may be emitted as lines and 

 bands which are within the range of the 

 visible spectrum. These lines, under defi- 

 nite conditions of pressure, etc., are charac- 

 teristic of the different elements and com- 

 pounds. The best known attempts to util- 

 ize this principle are the Moore system of 

 lighting, in which long tubes of luminous 

 gas are employed, and the mercury lamps, 

 which, while more fiexible on account of 

 size, are still objectionable because of the 

 color of the light. A simple form of mer- 

 cury arc is shown. 



It is rather interesting that the efficien- 

 cies of all of these various sources of elec- 

 tric light are not nearly so widely different 

 as one would expect from a consideration 

 of the widely divergent methods of light 

 production employed. 



From the light of a vapor or gas to that 

 of an open arc is not a wide step, but the 

 conditions in the arc are apparently quite 

 complex and there is a great deal of room 

 for interesting speculation in the phenom- 

 ena of an arc. Briefly, there are two kinds 

 of arcs to be considered in lighting. One 

 has been in use for a century, the other for 

 a few years only. The first is the successor 

 to Sir Humphry Davy's historical arc 

 between charcoal points. In this kind of 

 arc the current path itself is hardly lumin- 



ous and the light of the lamp is that given 

 by the heated electrodes. In ease of direct 

 current it is the anode, or positive elec- 

 trode, which gets the hotter and gives far 

 the greater part of the light. In the car- 

 bon arc shown, it will readily be seen that 

 the light is emitted by the heated solid car- 

 bon of one electrode. This gives a steady 

 source of light, but is not so efficient as an 

 are in which material in the arc stream 

 itself is the source of light. The arc may be 

 made to play upon rare earth oxides, and 

 these, being heated to incandescence, in- 

 crease the luminosity, but this has not 

 proved useful. The more common way is 

 to introduce into the carbon electrode cer- 

 tain salts which volatilize into the are and 

 give a luminous effect. Here cerium flu- 

 oride, calcium fluoride, etc., are used, and 

 the color of the arc, just as in the ease of 

 gas mantles, may be varied by varying the 

 composition of the electrodes. This is seen 

 in the arc from the carbon electrodes con- 

 taining such salts. 



I have arranged several different kinds 

 of ares, and before each is a magnifying 

 lens, to throw the image of the arc upon a 

 screen. This permits our seeing the phe- 

 nomena of the arcs and observing the char- 

 acteristics of each. The very essential dif- 

 ferences between the plain carbon arc and 

 the luminous or flaming arc is readily 

 noticed. In the latter case the greater 

 part of the light is due to the incandescent 

 metallic vapors in the space between the 

 electrodes. Substitution of one chemical 

 for another in such flaming arc electrodes 

 has covered quite a wide range of chemical 

 investigation. Salts are chosen which give 

 the greatest luminosity without causing the 

 formation of too much ash or slag. Some 

 compounds of calcium, for example, are 

 practicable, while others are not, though all 

 of these would, under suitable conditions, 

 yield the calcium spectrum. 



