Intelligence and Miscellaneous Articles. 151 
I have made this verification for air and nitrogen up to 85 
atmospheres; and at the same time I have made the deter- 
minations necessary for referrmg the series to the normal pres- 
sure which is indispensable in the construction of manometers. 
These experiments were made in one of the towers of the church 
of Fourviéres, where, thanks to the kindness of M. Sainte-Marie 
Perrin, the architect of the monument, I had at my disposal a 
vertical height of 63 metres during the whole time necessary for 
frequently repeating the series. I retained the method I had used 
at the shaft at Verpilleux, but with some improvements. The 
glass manometers were so arranged that the gas occupied at least 
the value of 500 divisions of the stem; the diameter of the iron 
tubes was double; the conditions were favourable for exactly 
reducing the volumes of mercury to zero. 
The various series agreed perfectly. For nitrogen the results at 
which I arrived are almost identical with those of my former 
experiments. The differences do not in general exceed a thousandth 
of the total pressure, except in one point of the curve which I had 
already noted as irregular. The differences with M. Cailletet’s 
numbers exist still; between 40 and 60 metres they correspond to 
an error of more than 2 metres of mercury. 
In the following table the values of the product pv for nitrogen 
and air are for the temperature 16°; they are compared with the 
standard pressure, for which they are supposed equal to unity. 
| 
Pressure in| Nitrogen, Air, ‘Pressure in} Nitrogen, Air, 
metres. pu. pu. metres. pu. pu. 
0-76 1:0000 1-0000 45°00 0:9895 0:9815 
20:00 0:9930 0:9901 50°00 0-9897 0-9808 
25:00 0:9919 0-9876 55°00 0:9902 0-9804 
30°60 0:9908 0-9855 60°00 0-9908 0-9803 
35°00 0-9899 0-9832 65°00 0-9913 0:9807 
40°00 0-9896 0:9824 
The deviations for 20 metres are considerably lower than those 
which had been given by Regnault, which is explained by the fact 
that Regnault had worked at 4°. For nitrogen the minimum of 
the product pv corresponds to a pressure of 42 metres; in my first 
researches it was 45 metres. For air it is 59 metres. It must 
moreover be remarked that the variation of pv is here so small near 
the minimum, that an unimportant difference in the pressure may 
displace the minimum ordinate through several metres. 
Within the limits of the above table we may either use air or 
nitrogen for filling the manometers; for higher pressures it will 
be preferable to choose nitrogen, which has been directly studied. 
Tt will be sufficient to reduce proportionally all the numbers which 
