618 REGNAULT’S HYGROMETRICAL RESEARCHES. 
a tube Z/’ is hermetically fixed, which acts as a Mariotte’s tube 
to render the current constant; in the second tubulature 8, a 
thermometer T is fitted, the bulb of which occupies the centre 
of the vessel. The lower end has a single tubulature with a gra- 
duated stopcock R, which has a tube 1 decimetre long, which 
remains filled with water when the current ceases, and prevents 
the external air from entering the lower tubulature when the 
vessel is emptied. 
The tube of Mariotte has a stopcock 7, which allows of stop- 
ping the aspiration of the air, and a U-shaped tube’ filled with 
pumice-stone moistened with sulphuric acid, which remains con- 
stantly fixed to the apparatus. The object of this tube is to 
prevent the vapour passing from the aspirator to the counter- 
poised drying tubes B and C. 
To absorb the humidity of the air, I only employed two U- 
shaped tubes, 0™18 in height, and filled with coarse fragments 
of pumice-stone moistened with sulphuric acid: fine fragments 
would oppose too much resistance to the passage of the gas, and 
the airin the aspirator would no longer present the same elastic 
force as the external air. 
The two tubes intended to absorb completely the humidity of 
the air are very short; I have endeavoured to make them as 
small as possible, because I attach great importance to rendering 
this method eminently practical and easy of execution in all hy- 
grometric experiments. Experiment demonstrates moreover 
that these two tubes completely retain the humidity of the air. 
The first tube generally absorbs the whole of the water, and it 
is rare that the second tube gains 1 or 2 milligrammes in weight. 
I did not however regard this proof as sufficient. I wished to 
ascertain whether, on putting at the end of these two first tubes 
several others filled with pumice-stone moistened with sulphuric 
acid and immersed in refrigerating mixtures, these latter tubes 
would not increase in weight. I attached to the end of the 
second counterpoised tube a third tube which had been immersed 
in ice, then a fourth, which was placed ina refrigerating mixture 
of ice and chloride of calcium, at a temperature of —30 degrees. 
The experiment was made in the usual way; the tubes 3 and 
4 were afterwards placed under a bell-glass with some quick 
lime, where they were left for several hours, that they might 
exactly take the temperature of the ambient air. It was found 
that they presented exactly the same weight as before the expe- 
