466 Natural-Philosophical Collections. 
proportion. A given mass of air was therefore submitted to pressures, succes- 
sively increasing from 1 to 24 atmospheres. ‘The volume which corresponded 
to each of these pressures was carefully noted; and then there could be substi. 
tuted for the troublesome column of mercury this mass of air, the different vo- 
lumes of which would represent determinate weights with the greatest accuracy. 
This preliminary operation, which, in consequence of local circumstances, lat- 
terly became of absolute necessity, has, moreover, enabled the commission to 
verify, in an accurate manner, one of the most useful laws of physics, that known 
by the name of Marriotte’s law. This law, although it had received a certain 
degree of verification, was far from being established in a satisfactory manner in 
reference to high pressures. The last observations of MM. Dulong and Arago 
leave no doubt now on the subject. A table drawn up by the commissioners 
presents the results of thirty-nine experiments made on the same mass of air, 
submitted to pressures comprised between one and twenty-seven atmospheres, and 
within which Mariotte’s law never contradicts itself in any appreciable manner. 
The commission of the Academy had the very natural desire of profiting by the 
apparatus which it had raised, by submitting to observation two or three other 
species of elastic fluids, and proving whether they also obey Marriotte’s law ; 
but they considered it necessary, before all, to complete the inquiries required by 
the government; and when these were concluded, they found it impossible te 
obtain from the administration of civil buildings the use of the place in which 
their pressure-apparatus was putup. ‘* This circumstance,” said M. Dulong, “ is 
so much the more vexing, that we could have accomplished the determination of 
this important point of the mechanics of gases without any increase of expense, 
and in a very short time ; while it would now require a great expense, and seve- 
ral months of severe labour, to be able to resume the subject at the point where 
we left it.” 
Determination of the Elastic Powers of the Vapour of Water. 
The experimenters, as we have seen, had procured a manometer, by means of 
which they could know the pressure exercised by the vapour with an accuracy 
equal to that which experiments made directly upon mercury would have yielded. 
It was sufficient to put a caldron in communication with the reservoir of this 
manometer, to accomplish the solution of the problem. By this method there 
was obtained the great advantage of avoiding the inconveniences already pointed 
out of great oscillations of the metallic column; and the apparatus was so dis- 
posed, that a vapour caldron could be substituted for the pressure pump, without 
producing any derangement. But, fearing the effects of an explosion in a tower 
whose arches threatened ruin, and especially cn account of the vicinity of the 
college of Henri IV., the commissioners determined upon transporting their ap- 
paratus, with the necessary precautions, to the courts of the observatory. The. 
precise measurement of the high temperature of steam requires precautions, the 
neglect of which has led some observers into great errors. 
The first precaution consists in keeping count of the cooling produced by the 
atmospheric air upon the part which remains placed externally to the caldron. 
It can only be done, with accuracy, by keeping the whole of this part at a con- 
stant temperature; and this was carefully done. 
The second consists in not immediately exposing to the pressure of the steam 
the thermometer, which serves to indicate this temperature, especially when the 
pressure is very great. For, even should a thermometer be found which could 
support it without breaking, it would certainly undergo a pressure which would 
tend to raise the column of mercury by an effect independent of heat, and would 
thus produce a cause of error for which it would be very troublesome to make al- 
lowance. ‘To obviate this inconvenience, which had not previously been perceived 
by any observer, the thermometers were placed in the interior of metallic tubes, 
closed at one end and filled with mercury. They were selected extremely thin, 
in order to oppose the least possible delay to the transmission of heat, and were 
