316 
PEOFESSOE W. G-. ADAMS AND ME. E. E. DAY ON THE 
which has been as yet tried. They also show that the change produced in selenium by 
exposing it to rays of light is not due to the increase of temperature which would be 
caused in the selenium by those rays. 
From these and a variety of other experiments which have been already described, 
and which point all in the same direction, we conclude that for a slight increase of 
temperature the resistance of the selenium is greatly increased, and for a slight lowering 
of temperature the resistance is greatly diminished. 
The Use of a Selenium Bar as a Thermometer. 
At the beginning of January 1875 there was severe frost and snow, which was 
followed by a sudden thaw on January 5th, when the temperature in the shade rose 
to 44° F. Previous to the thaw, in the room where the experiments were made, the 
resistance of the selenium was about 2 ’25 megohms ; on the day of the thaw the resistance 
had increased to 2*7 megohms, the increase in the resistance of the selenium amounting 
to 450,000 ohms, or nearly one fifth of its total resistance. These results suggest that 
a piece of selenium kept in the dark and forming part of an electric circuit would form 
a very delicate thermometer, and that a very delicate differential thermometer may be 
formed with two similar pieces of selenium balanced against one another. 
Exposure of Selenium to Radiations from the Bunsen Flame. 
The effect of the ordinary non-luminous flame of a Bunsen burner was compared with 
that of the same flame rendered luminous by closing up the air-holes. 
On exposing the selenium for several seconds to the non-luminous flame at a distance 
of 20 centims., only a very slight effect was observed, the deflection of the galvano- 
meter-needle being represented by about 10 divisions of the scale; but on making the 
flame luminous, by closing the air-holes of the burner, the spot of light was deflected off 
the scale with great rapidity. 
On testing for the amount of radiation from the Bunsen flame in its two states, by 
means of a thermopile and a galvanometer, it was found that the actual amount of heat 
radiated from the non-luminous flame was very nearly, but not quite, as great as that 
from the same flame when rendered luminous. 
These experiments with the Bunsen burner were repeated in various ways, and it was 
always found that, while the hot non-luminous flame produced scarcely any perceptible 
effect, exposure to the same flame Avhen rendered luminous produced a diminution of 
the resistance amounting in some cases to nearly one fourth of the whole resistance of 
the selenium. 
In order to examine this point more fully, experiments were made with the electric 
light so as to determine what effect was produced by the obscure heat-rays. The 
selenium was placed at a distance of about 2 feet from the carbon points, and some 
medium which would absorb the luminous rays was placed in the path of the beam. 
A plate of smoked rock-salt and a plate of alum were interposed together in the path 
