Page:Popular Science Monthly Volume 47.djvu/538

524 then occurred, that the lightness of the "chemical" nitrogen was due to a partial dissociation or breaking up of the molecules of the gas into single atoms under the conditions of its preparation. This, too, was negatived by experiment. One or the other of the gases must be a mixture, containing an ingredient much heavier or much lighter than ordinary nitrogen. To suppose a lighter ingredient mixed with the chemical nitrogen required the existence of two kinds of nitric acid, which was out of the question. "The simplest explanation in many respects was to admit the existence of a second ingredient in air from which oxygen, moisture, and carbon dioxide had already been removed."

This explanation was put to the test by an attempt to isolate the suspected gas, with the result that by two entirely distinct methods a new substance was obtained from air.

One of these methods was that of Cavendish, already described. Air confined over dilute alkali is subjected to the action of electric sparks, while oxygen is added from time to time until, with an excess of oxygen present, no further absorption occurs. The oxygen is then removed by alkaline pyrogallate, and argon is left.

The second method for the separation of argon is based on the fact that red-hot magnesium unites with nitrogen, forming a nonvolatile compound. Air from which moisture and carbon dioxide have been removed is freed from oxygen by passing it over heated copper, and then from nitrogen by means of magnesium turnings at a red heat. The removal of the last portions of nitrogen is a tedious operation, requiring some two days. The residual gas is pure argon.

The gas obtained by both of these methods is the same, and its behavior proves conclusively that it is a new substance. Prof. Crookes finds that it gives two spectra, according to the strength of the induction current, one characterized by red and the other by blue lines; and testifies that he has "found no other spectrum-giving gas or vapor which yields spectra at all like those of argon"; and that "as far, therefore, as spectrum work can decide, the verdict must, I think, be that Lord Rayleigh and Prof. Ramsay have added one, if not two, members to the family of elementary bodies."

The behavior of argon at low temperatures and under high pressures has been examined by Prof. Olszewski, of Cracow, who is well known for his researches on the liquefaction of air and other gases, its critical temperature—that is, the temperature at which its liquefaction under pressure first becomes possible—is —121º C., and at that point it is condensed to a liquid by a pressure of 50·6 atmospheres. Liquid argon becomes an icelike solid at a still lower temperature, melts at —189·6º, and boils under ordinary