Lamps with Direct and Alternating Currents. 481 



the direct current was immediately followed by one with 

 alternating current, so that the means of all the corresponding 

 direct-current measurements may be safely compared with the 

 means of all the alternating-current measurements. Similarly 

 for this investigation it is unimportant whether the standard 

 candle used on one clay was slightly more or less bright than 

 the standard candle used on the following day. Further, 

 since for any two successive observations with the same 

 coloured light with direct and with alternating current the 

 screen was at practically the same distance from the lamp, the 

 fact that the rays of light coming from the lamp made a 

 somewhat different angle with the surface of the paraffin- wax 

 from the angle made by the rays coming from the candle, 

 introduced no error in this investigation of comparative 

 efficiencies, although it might very likely do so in an absolute 

 determination of the efficiency of an incandescent lamp. In 

 fact the simple device of successively taking observations with 

 direct and with alternating currents throughout the whole 

 investigation removed most of the objections that usually 

 may be made against photometric determinations carried out 

 with a candle as the standard of light. 



Disregarding the experiments made with the lamp with the 

 looped filament, for the reason given above, Table II. gives the 

 summary of the results obtained; and the conclusion to be 







Table II.- 



—Mean Values. 









Lamp. 



.2 



I'Sa 



So," 



vi 5 : 



vT 2 



Watts per Candle. 

 White Light. 



* Z m 

 !M OTJ 

 0-J3 fi 



/r. * O 



£ a o 



.So" 1 



61 3 



50 volt M-shapedl 

 filament. J 



50 volt horseshoe- 1 

 shaped filament J 



50 volt horseshoe- 1 

 shaped filament J 



50 volt horseshoe- 1 

 shaped filament J 



20 



19 

 20 

 16 



1-25 



130 

 1-34 

 1-35 



50-5 



50-0 

 50-0 

 50-0 



Gramme. 

 3-053 



Ferranti. 

 3-033 



453 



1 

 453 



167 



Green Light. 

 Gramme. [Ferranti. 



Eedl 

 Gramme. 



jight. 

 Ferranti. 



2-597 

 2-935 

 2-900 



2-534 

 2-966 

 3-073 



3-100 

 3-254 



3-504 



3-100 

 3-164 

 3-477 



Mean of the last three resu 



ts..... 



2-811 



2-857 



3-286 



3-247 





Gramme. Ferranti. 

 3-0490 ' 

















In all the experiments the plane of the filament was perpendicular to the screen. 



