1819.]  andon the Laws of the Communication of Heat. 179 
Apparatus destined for Experiments on Cooling. 
The bodies, whose cooling we observed, were, conformably to 
the principles just explained, thermometers of such a size that 
the diminution of their temperatures could be observed with 
recision. We constructed two of them, the bulb of one of 
which had a diameter of about six centimetres ; that of the other 
of two. ‘The first, containing about 3 lb. of mercury, served for 
observations at a high temperature. The smaller one was 
employed for low temperatures, in order to shorten the duration 
of the experiments. it was easy to deduce from the results given 
by the last, those which would have been given by the large one 
if the series of its cooling had been prolonged ; for that, it was 
sufficient to commence the observations with the small thermo- 
meter at a higher temperature than that at which the large one 
had terminated. By determining then the ratio of the velocity 
of the cooling of this last to that of the small thermometer for a 
common excess of temperature, we obtained the number by 
which it was necessary to multiply all the results given by the 
small thermometer to obtain the corresponding velocities of the 
other. 
These two instruments, constructed with all the care possible, 
did not-differ from common thermometers, except in this parti- 
cular, that the tube on which the degrees were marked was 
separated from the bulb by an intermediate tube, the calibre of 
which was very small. We shall see immediately the motive of 
this construction. 
The experiments on cooling in a vacuum, with which we had 
to commence, required that the thermometer could be transported 
into a pretty large space, in which a vacuum could be made very 
rapidly. It was necessary also that the surface which surrounded 
the thermometer on every side should be maintained at a known 
temperature : and as it was requisite that the same apparatus 
should serve for observing the cooling in air and in gases, it was 
requisite that the gases should be introduced into it in a conve- 
nient and prompt manner. All these conditions were satisfied 
by the following construction. 
The enclosure in which the cooling takes place is formed of a 
large, thin, copper balloon M M’ M” M’”’ (fig. 5) whose diameter 
is about three decimetres. The neck of this balloon is ground 
at its upper part so as to be terminated by a perfectly flat surface, 
which is rendered horizontal by means of a level. This balloon 
is plunged almost completely into a large wooden cylindrical 
trough full of water, in which position it is kept by the strong 
cross beams, RR’, RR’. It is evident that the walls of this 
balloon, being very thin and very good conductors, must assume 
constantly the temperature of the surrounding water; and being 
covered within with lamp-black, they cannot reflect any sensible 
quantity of the heat sent to them from the thermometer. Besides, 
M 2 
