194 
tometer, and all the other divisions are arranged in such a 
manner, that the numbers which they bear, are every where 
proportional to the squares of the distances of the places 
where these divisions are marked, from the middle to the 
bottom of the photometer, where the two shades which are 
compared coincide in the experiments. 
To complete the important part of a standard of light, it 
is necessary to have a very steady light, and) at the same 
time, one known to every one. I found these two qualities 
in a good candle. That which I chose, was of those which 
are live to a pound; it was 9 lines in diameter, 10 inches 
and a quarter long, and weighed 90 grammes, (about 3 
ounces) ; it burnt 7 grammes of wax per hour, with a beau¬ 
tiful flame, and with an uniformity so perfect, that it rarely 
varied a fortieth in its ordinary consumption. 
I have given to the light of this candle the value of 100 
degrees, and I placed the candle always at the division of 
the scale of the photometer, which is marked 100 degrees, 
when it is burning. In order to express in a uniform and 
convenient manner the quantities of the different inflamma¬ 
ble substances consumed in the experiments, I divided into 
100 parts the quantity of wax which the candle I used as 
a standard consumed in an hour (=7 grammes), and I es¬ 
tablished one of these parts (= de grammes or 1.34545 
grains) as an unit of weight (unite de poids), and I employ 
it as such in all my experiments. 
We shall see in the sequel, how this arrangement has 
rendered the comparison of the result of my experiments 
easy and intelligible. 
The objects I had particularly in view in the course of ex¬ 
periments, of which I now have the honor of giving an ac¬ 
count to the class, was to determine if the quantity of light 
manifested during the combustion of an inflammable sub¬ 
stance, is always in an invariable proportion to the quantity 
of the substance which is consumed; and as the flame of 
an Argand’s lamp has a double current of air, well regulat- 
