E. Merritt — Light from Incandescent Lamps. Ill 



Some determinations of the radiation from incandescent 

 lamps made by Capt. Abney and Col. Festing* are also of in- 

 terest in this connection. The radiations from the lamp were 

 dispersed by a prism, and the increase in the energy of each 

 ray, as the candle power was raised, was measured by a ther- 

 mopile. The curves found for different wave-lengths were 

 approximately parabolas with vertical axes. The summation 

 of a number of such curves should evidently give a curve sim- 

 ilar to those found in my experiments. 



There is another way of looking at the results of these ex- 

 periments, which may be of some practical interest. An in- 

 candescent lamp, in fact any artificial source of illumination, 

 may be considered as a machine for producing light. Energy 

 in some form being supplied, the machine transforms a cer- 

 tain portion of this energy into luminous vibrations. Now 

 the efficiency of any machine is the ratio of the useful work 

 done by it to the energy expended. The efficiency of an in- 

 candescent lamp, therefore, is the ratio of the energy given 

 off as light, to the total energy consumed. This efficiency, for 

 the lamps tested, can be determined from the results already 

 given by dividing the values of L by the corresponding val- 

 ues of W. The values of this quotient are given in the ta- 

 bles of results. The curves of 

 efficiency and total energy are 

 shown in Fig. 5. The largest 

 value found for the efficiency 

 was slightly over 7 per cent. 



It will be noticed that this 

 " luminous efficiency," as it 

 might be called, does not corre- 

 spond at all with the commer- 

 cial efficiency of the lamp. 

 Take for example the two Edison lamps A and B. A was 

 new and fairly efficient, commercially, for lamps made at that 

 time, giving 16 C. P. at about 80 Watts ; B was old, and com- 

 mercially very inefficient, giving 16 C. P. at something over 

 100 "Watts. Yet the efficiency of A as a light machine was 

 only 3-5 per cent at 16 C. P., while that of B was 7 - 4 per cent. ! 

 The difference must be in the quality of the light. 



Dr. Thomson, whose article on this subject has already been 

 referred to, made several determinations of the efficiency of 

 oil and gas flames. He found that about two per cent of the 

 radiant energy was light. This low value of the efficiency is 

 probably not due to the low temperature of the incandescent 

 portion of the flame, but rather to the fact that the heated 



* " Relation between Electric Energy and Radiation, in Incandescent Lamps." 

 •Capt. Abney and Col. Festing. Proc. Roy. Soc, vol. sxxvii, p. 157. 



