and Surfaces of Incandescence Lamps. 377 
attempted has been in the case of the experiments made by the 
Committee appointed to report on the incandescence lamps in 
the Paris Electrical Exhibition, an abstract of which appears 
in the same volume of the ‘ Journal of the Society of Tele- 
graph Engineers.’ The Committee condense these results on 
four varieties of lamps into the following numbers for the 
efficiencies, or candles per horse-power and electromotive 
force working them, taken at two very different values. 
Edison. Swan. | 
a oN See a 
PEST Morreale cio ssociie sn ses 0.% oo lt 98°39 47°3 54:21 
Candles per horse-power...| 196°4 307°2 1779 262°5 
| Lane-Fox. Maxim. | 
eT Cee aa CS ae ee | 
TIP ese 43°63 48:22 56°49 62:27 | 
| Candles per horse-power...| 178°6 276°9 151°3 239°4: | 
| 
If we take the logarithm of each number, and compare the 
difference of the logarithm of the efficiencies with the differ- 
ence of the logarithm of the corresponding electromotive 
forces for each lamp, we have the following ratios :-— 
Edison. Swan. 
19428 16895 a 
4302 — 49 oop ay 
Lane-Fox. Maxim. 
20277 19928 
4344 me 4231 ewe 
If, therefore, k, and ky be the efficiencies corresponding to 
two observed values v, and v, of electromotive force, 
log k, log hy _ 4.5 
log v1 —log ve 
or b= Cour” 
represents an approximate formula for calculating the effi- 
ciency. If, now, approximately, in the case of an Hdison 
lamp, the candles per horse-power vary as the 4:5 power of the 
electromotive force, and the life varies inversely as the 25th 
power of the electromotive force, it follows that 
1 
or, roughly, that the life varies inversely as some power 
between 5 and 6 of the candles per horse-power. 
The writer was given in America an extensive series of 
observations on the relation between the candles per horse- 
power and the average life, which were as follows :— 
