590 REGNAULT ON THE 
The thermometers employed in the experiments which were 
made at inferior pressures to those of the atmosphere ranged 
only from 0° to 100°; they had from six to eight divisions for 
each degree; it was therefore easy to observe with accuracy the 
goth of a degree. The four thermometers employed in the ex- 
periments with high pressures had a range from 0° to about 240°. 
The centigrade degree was equivalent to from 2°5 to 3 divisions 
of their scale. All these instruments were graduated and veri- 
fied with the greatest care. 
For correcting those portions of the column not immersed in 
the experiments under high pressure, the following experiments 
were made :— 
The water in the retort was made to boil under a certain 
pressure greater than that of the atmosphere; three of the four 
thermometers were immersed in the iron tubes ; in the fourth iron 
tube, the thermometer having a bulb at the upper end was placed, 
with the mercury so separated that the column should remain 
stationary at a few millimetres above the cover, and required 
consequently no correction; the indications of the four ther- 
mometers were then read off. To be enabled to calculate the 
temperature indicated by the double-bulbed thermometer, it was 
only necessary to know the position of the 100° point on the 
thermometer; this positicn was determined in a direct manner 
by immersing the thermometer in the vapour of boiling water 
at the pressure of the atmosphere. 
The temperature of the thermometer with the two bulbs re- 
quiring no correction being known, as likewise that of the other 
thermometers which required correction for those portions of 
their columns above the retort, it was easy to see whether the 
mode of correction employed brought uniformity into all the in- 
dications. It may be well to mention, that the portions of the 
stems not immersed were of various lengths in the different 
thermometers, and required consequently very different absolute 
corrections. 
These experiments were made at about the temperatures of 
110°, 120°, 130°, and 140°: it was observed, that between these 
limits of temperature, corrections being made for the stems, on 
the supposition that their mean temperature was indicated by a 
mercurial thermometer, the reservoir of which occupied a medium 
position, the thermometers never differed from the thermometer 
with two bulbs more than 02. These differences may be 
