OBSERVATIONS FOR FINDING THE VOLUMES OF ^ AND €. 
875 
By the good offices of M. Arago, permission was obtained from the French Govern- 
ment to compare the pound with the standard kilogramme of platinum deposited in 
the Archives on the 22nd of June, 1799, known as the ‘kilogramme des Archives.’ 
The weights selected for comparison with the standard kilogramme, which hence- 
forward will be designated by the letter 91, were PC No. 1 and PC No. 2, together 
with the auxiliary weight B and a platinum weight V of about 192'4 grains, making 
altogether about ]5432'35 grains. In order to obtain a second comparison perfectly 
independent of the former, a kilogramme of bronze was constructed by Mr. Barrow, 
with which, after comparison with 9[, it was my intention to compare PS together 
with each of the four platinum copies of the pound in turn, and other platinum 
weights sufficient to make up a kilogramme. 
By some most unaccountable oversight 91 had never been weighed in water pre- 
vious to its final adjustment. Afterwards, on account of its legal importance, it was 
considered hazardous to immerse it in water, especially as, from the method of pre- 
paring platinum at that time in use (fusion with arsenious acid and subsequent igni- 
tion of the arsenide of platinum under a muffle till the arsenic was burnt away), there 
is reason to suppose that it is not entirely free from an admixture of arsenic which, 
on being wetted, might oxidize and then dissolve, and thus produce a very sensible 
alteration of weight. Its form is that of a cylinder of about 39’4 millimetres in 
diameter and 39‘7 millimetres high, having its edges rounded by a surface 0'75 
millimetre broad, and having a radius of about 3 millimetres. An approximate value 
of its density was obtained by Professor Schumacher and Olufsen in the following 
manner. In August and November 1831, the density of a kilogramme^ in Professor 
Schumacher’s possession was found to be 21*212, by weighing it in water and in air. 
In March 1832, by measuring pairs of diameters at right angles to each other in 
planes cutting the axis in eight different points, and the distances between nine 
corresponding pairs of points in the ends of the cylinder, the volume of ^ at 0° 
appeared to be 471 14*4 cubic millimetres. In the autumn of 1834, Professor Olufsen 
measured two diameters at right angles to each other at the middle and at each end 
of 91, and the distances between eight pairs of corresponding points in the circular 
ends. The volume of 91 at 0°, deduced from these linear dimensions appeared to be 
48615*4 cubic millimetres, and consequently its density 20*644*. These measure- 
ments, though made with the utmost care, appeared to be too few, and confined to 
too small a number of points, to determine the density of 91 in this manner with 
sufficient accuracy. I therefore resolved to compare its volume by means of the 
stereometer, with that of a brass cylinder of nearly the same dimensions, the volume 
of which might afterwards be found by weighing in air and in water. 
The representations of M. Arago procured for me the privilege of forwarding the 
balance and stereometer unexamined from Havre to the Douane in Paris, where I 
received them without being obliged to unpack the cases in which they were con- 
* Schumachee’s Jahrbuch fur 1836, p. 237. 
5 Y 
MDCCCLVI. 
