THE PROGRESS OF THE ELECTRIC LIGHT 5 



four Nernst lamps. In spite of these drawbacks, the Nernst 

 lamp was a moderate success during several years and might 

 have come into far more general use but for the invention of 

 the metallic filament lamps ; perhaps the best that can now be 

 said for the Nernst lamp is that its production undoubtedly 

 stimulated the investigations which led up to the invention of 

 the metallic filament lamps. 



The earliest of the metallic filament lamps was the osmium 

 lamp invented and developed by Dr. Auer von Welsbach 

 during the years 1898 to 1903 and first introduced into this 

 country in 1905. Little need be said in reference to this lamp, 

 as it is now no longer in commercial use and was never a 

 great success. It is interesting chiefly as the forerunner of 

 the tungsten lamps, which are manufactured by methods similar 

 to those first tried with the osmium lamp and in some cases 

 gradually developed from those methods. Osmium, however, 

 proved unsuitable as a filament, mainly on account of its 

 extreme brittleness when cold ; it is markedly inferior in many 

 respects to tungsten, which has superseded it. 



At about the time when the osmium lamp was introduced, 

 Messrs. Siemens & Halske brought out the tantalum lamp, 

 which they had spent several years in developing. This was 

 the first metallic filament lamp with which any genuine success 

 was achieved. The filament of this lamp is made of tantalum 

 prepared by a special process, fused electrically in vacuo and 

 then drawn into a very fine wire : tantalum being of low 

 specific resistance compared with carbon, a very much greater 

 length of filament is needed for a given voltage and to 

 accommodate this length in the lamp bulb it is wound to and 

 fro over a wire support. As the tantalum filament may be 

 heated by a much higher temperature than a carbon filament 

 the lamp is consequently far more efficient. The average 

 consumption in watts per candle is slightly under 2 for direct 

 current lamps — in fact, 1*85 represents a fair average of the 

 trustworthy tests which have been published. The alternating 

 lamp is not quite so efficient ; its consumption may be taken 

 as 2 watts per candle or a little over. It will be seen therefore 

 that the invention of this lamp halved the cost of energy for 

 electric lighting : as the lamp was in all other respects a 

 thoroughly commercial article, it is not surprising that its 

 success was immediate. 



