AND STATICS UNDER THE INFLUENCE OF LIGHT. 
357 
manganin resistances (manganin 100,000 ohms, 6 ohms in the shunt were used, as 
shown in the diagram ; the coil of the galvanometer, also of manganin, is 3 ohms). 
These resistances give, under the conditions of experiment, a sufficient deflection 
(about fl- 15 centims. to — 15 centims.), so that the value of the deflection caused 
by the acetylene light is always measured in standard units, independent of the 
6 oh ms. 
ii imm n n if i n in 11 mm 
Scabs. 
sensitiveness of the galvanometer, &c. (the diagram also shows that the difference 
between the temperature in the quartz vessel immersed in the bath, owing to the 
reaction which takes place in it, and that in the bath is also measured with the same 
galvanometer by means of iron-nickel thermocouples ; this, however, has nothing to 
do with the photometer itself). 
The principle of the measurement of the light intensity consists in its deter¬ 
mination objectively by means of the deflection of the galvanometer and standard 
units (Clark, manganin resistances). We believe that if two sources of light 
(say acetylene gas) at a fixed distance, say 1 metre from the plane containing a 
given thermopile (with a given number of iron-constantan junctions), in the same 
relative position to the thermopile (the lines of the flames being parallel to the 
line of the exposed junctions and seen in the same position from the narrow tube in 
the double copper cylinder of the thermopile), give, or are made to give, the same 
deflection in standard units, then the intensity of the lights must be the same, 
provided they are quite pure and free from smoke and that the burner is cool. We 
assume that the heat effects of any source of light upon the exposed junctions of the 
thermopile, i.e., the rise of their temperature, will be directly proportional to the 
amount of light falling upon them, i.e., to the intensity of the light. This point is 
of primary importance. A careful theoretical investigation of it was absolutely 
necessary, and the investigation confirmed the conceptions. The electromotive force 
of the thermopile being thus directly proportional to the difference of temperature 
of the thermo-junctions exposed to the light, and of those which are left in the dark, 
