But these experiments lead to a further result. They show that nitrogen, procured from its compounds, when treated in the same way as atmospheric nitrogen, yields no trace of argon. And it must therefore be taken as proved without doubt that argon is actually present in the atmosphere as such, and is not produced by any process to which the nitrogen has been submitted in order to extract it. This point having been settled, the actual percentage of argon in atmospheric air next invited inquiry. It is by no means very easy to absorb quantitatively the whole of the nitrogen from an accurately measured sample of air, for small gains and losses are apt to occur. It is necessary to keep the air out of contact with water as much as possible, because argon, being more soluble than nitrogen, dissolves in larger proportional amount in the water, and is thereby partially removed. The air was therefore entirely manipulated over mercury. The processes were like those previously employed: most of the nitrogen was removed with magnesium, and the residue was freed from all nitrogen by sparking with oxygen. Experiments directed to this end were carried out by Mr. Kellas in Professor Ramsay’s laboratory, and independently by M. H. Schloesing in Paris. The results were identical. “Atmospheric nitrogen” consists of pure nitrogen mixed with 1·186 per cent of its volume of argon. It is now possible, knowing the percentage of argon in atmospheric nitrogen and its density (19·94), to calculate whether Lord Rayleigh’s determinations of the density of atmospheric nitrogen were correct. The weight of one litre of pure nitrogen is 1·2511 gram, and of argon, 1·7818 gram; hence a litre of a mixture of 98·814 volumes of nitrogen with 1·186 volume of argon must possess the weight 1·2574 gram. The actual number found by Lord Rayleigh was 1·2572 gram, which is almost exactly identical with the number calculated.