318 
On the Measure of Temperature, d^c. 
[Oct, 
common excess of temperature, gave the number by which the results of the smaller 
thermometer should be multiplied, to obtain the corresponding rates of the larger. 
These two instruments, which were constructed with the utmost possible care, 
differed from ordinary thermometers only in this, 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. The reason of this construction will be presently seen. 
The experiments on cooling in a vacuum being those with which we were to 
begin, required that the thermometer should be quickly disposable in a space 
sufficiently roomy, in which a vacuum might be produced without delay. It 
was further desirable, that the case surrounding the thermometer on all sides 
should be kept at a fixed temperature ; and as the same apparatus was to be used in 
observing the progress of cooling in air, or any of the gases, it was indispensable 
that there should be means of readily introducing any of these. These conditions 
will all be found fulfilled in the following arrangement. 
The surrounding case or envelope within which the process of cooling goes on, is 
a large balloon, formed of very thin sheet copper. (PI. V. fig. 1. M M' M" M" ) 
The diameter being about 3 decimetres, (11,82 inches , ) the projecting neck of this 
balloon is ground on its upper surface, so as to be an exact plane, which is adjusted 
to a horizontal position, by means of a level. The balloon is immersed within a 
short distance of the edge of this neck, in a cylindrical tub of water, and is retained 
in an invariable position, by means of strong supports RR'R RA It will be seen, 
that the sides of the balloon being extremely thin, and at the same time, excellent 
conductors, must always be of the same temperature with the water surrounding 
them ; and being covered inside with lamp black, they reflect exceedingly little of 
the heat they receive from the thermometer. And as this error, such as it is, is 
nearly as the loss of heat of the body, it affects proportionally every result. To 
raise the temperature of the water, it was only necessary to introduce steam into the 
tub, by means of the curved tube SUV, which descends to the bottom. 
The mouth of the balloon is closed, by means of a thick plate of glass, AB, 
ground with tlie greatest care, to fit the surface of the neck ; the surfaces in contact 
have sufficient extent, on account of the thickness of the neck, to allow of the appli- 
cation of a little grease, which renders the connection perfect, and cuts off all commu- 
nication with the atmosphere. 
This flat plate has a circular hole in the centre, in which is fitted tightly a cork, 
through which passes the stem of the thermometer ; the graduation commences im- 
mediately above the cork ; and the connecting tube is of sufficient length to allow 
the bulb to occupy the centre of the balloon. By making this tube of a very small 
diameter, the quantity of mercury external to the bulb is diminished ; no current 
can be established, and the enlargement at the commencement of the scale, allows 
the insertion of the stem, into the cork, to be made more firmly. The arrange- 
ment of the flat plate and thermometer is represented, fig. 2, in which the bulb of 
the instrument is placed above the furnace, by which it is heated. The screens 
AA' are plates of tinned iron, separated from each other, which serve to defend 
the glass plate AB from the action of the fire. 
To return to figure 1. The stein of the thermometer, which is outside the balloon, 
it is seen, is covered by a hollow tube, the ground edges of which fit exactly to the 
upper surface of the glass plate. This outer case is terminated above, by a stop 
cock. D, which connects it with a very flexible leaden pipe DEF ; the other 
extremity of which F is itself connected to the plate HK of an air pump, lhe 
tube connecting this plate with the barometer, has another stop-cock T, terminated 
by a neck, to which is attached a tube full of muriate of lime. The gas contained 
in the reservoir V escapes through this tube, after passing through the curved one, 
m n p r s. The reservoir being moveable, it is easy to establish an equilibrium 
between tlie elasticity of the gas , and the pressure of the atmosphere. 
The following was the order of proceeding in each experiment. Ihe water in t ie 
tub being brought to the proper temperature, and the thermometer attached to e 
plate of glass being heated nearly to the boiling point of mercury, it was quick y 
fixed in the balloon ; the case CT, which had been previously screwed to the lea en 
tube, was then brought down on the plate ; and while one person was emp oye 
bringing the surfaces into contact, an assistant was rapidly working the air pd ,n P» 
for the purpose of making a vacuum. The communication between the ba oo 
and the outer case was effected by means of two holes made in the glass disk, nea 
the central opening. , , - r 
If the rate of cooling was to be observed in a vacuum, we stopped when 1 
pump had reached its limit of exhaustion, and the residuum of air in the >a 
