OBSERVATIONS FOR FINDING THE VOLUMES OF S AND (£. 
877 
M. Gambey. In form it resembles 91, except that it is not quite so large, in conse- 
quence of the greater density of the metal of which it is composed. The cylindrical 
surface and the ends are very accurately worked. The metal of which it is made is 
greatly superior to that of which the standard lb. and its copies are constructed, as 
not the slightest indication of any defective place can be observed on its surface. 
This kilogramme will be designated by the letter (§. The comparisons of 91 with 
PC No. l-fPC No. 2+B+V were now discontinued, being of secondary importance 
after the acquisition of (B, and serving mainly to control the comparison of (0 with 91. 
A considerable number of observations were made with the stereometer on the 
25th of September, for the purpose of finding the volume of 91 by comparing it with 
that of a hollow brass cylinder M of nearly the same dimensions. Towards the end 
of the series, M being in the cup, the mercury in the graduated tube was observed 
to descend perceptibly though very slowly. This was caused apparently by the 
passage of a small quantity of air into the upper part of the graduated tube, where 
the pressure was about half that of the external air. At first it was supposed that 
the air found an entrance either between the rim of the cup and the glass plate 
which closed it, or through the screw joint by which the cup was connected with the 
collar into which the graduated tube was cemented, or, lastly, through the cemented 
joint itself. The cement was then varnished with shell-lac dissolved in alcohol, and 
the screw and glass plate carefully smeared with lard, so as to render the passage of 
air through the joints all but impossible. The mercury in the graduated tube still 
continued to descend when M was in the cup, but not otherwise. It appeared there- 
fore probable that the soldering of M had given way under the changes of pressure 
to which it had been exposed, so as to allow the air enclosed within it to escape 
slowly, when the pressure of the air in the cup was diminished. This conjecture was 
verified, on attempting to weigh M in water on the 22nd and 23rd of April, 1845, 
when M was found to have increased in weight after being left all night in the water 
in which it had been weighed ; and on placing it in the receiver of an air-pump and 
partially exhausting the air, drops of water made their appearance at the junction of 
the plane and cylindrical surfaces at one end of the cylinder. It was evidently useless 
to continue the observations with M, on account of its presumed leakage ; a second 
series of stereometer observations was therefore made on the 5th of October, in which 
the volume of 91 was compared with that of (If. 
Stereometer observations for finding tlie volume o/’91. 
At 16° the mercury contained in the graduated tube between 163’2 mm. and 
108*9 mm. weighed 1812*35 grains, and the mercury contained between 108*9 mm. 
and 0*0 mm. weighed 3614*99 grains. Hence the mercury contained between 0 mm. 
and 136 mm. weighed 4532*87 grains, and each mm. of the tube near 136 mm. con- 
tained 33*38 grains of mercury. The volume of PM is expressed in terms of the 
5 Y 2 
