Page:Proceedings of the Royal Society of London Vol 1.djvu/339

Rh maintained by Messrs. Berthollet on the non-existence of oxygen in ammonia, and on account of the inference deduced by Messrs. Gay- Lussac and Thenard, from the action of potassium on ammonia, who conclude that potassium is a compound of potash and hydrogen. Mr. Davy consequently details a variety of processes in which he examined all the circumstances and results of their mutual action.

The potassium employed was procured by passing dry potash through red-hot iron turnings in a gumbarrel, which appears to differ from that obtained by electricity solely in containing a very small portion of iron.

A green glass retort, after the introduction of apiece of potassium, is ﬁrst exhausted, then ﬁlled with ammoniacal gas, and after two subsequent exhaustions, ﬁnally ﬁlled again with the gas in an ex- tremely pure state. The potassium thus exposed to ammonia at common temperatures, loses its lustre and becomes white by a thin crust of potash on its surface; while the gas suffers a slight diminu- tion, and then contains about ﬁfth of its bulk of hydrogen.

When the potassium is heated by a spirit-lamp, the colour changes from White to a bright azure, thence to bright blue, green, and olive, which is the last state to which the whole of the potassium may be brought by continuation of the heat. When sufﬁcient ammonia is present to insure the complete saturation of eight grains of potassium, twelve cubic inches of ammoniacal gas disappear, and nearly eight cubic inches of hydrogen are evolved. The French chemist having stated this quantity to be exactly equal to that given out by the action of an equal quantity of potassium on water, Mr. Davy has made the comparison with great care, and ﬁnds the quantity of hydrogen given out by its action upon water to be just 8% cubic inches.

The olive-coloured compound formed is combustible. heavier than water, and a conductor of electricity. It fuses at a low temperature, and then begins to emit ammonia, till its quantity amounts to 4% inches out of 12 cubic inches that had disappeared. The residuum is then no longer ﬂuid, and begins to give of hydrogen and nitrogen, till the former amounts to 4 inches and the latter to 1,9,, in proportion exactly suited to the formation of ammonia. When this residuum no longer yields any gas. even at a red heat, a quantity of ammonia may still be formed from it by the addition of water, and amounting to about four cubical inches, and along with these about 1,lu th of an inch of hydrogen. The formation of ammonia in this case having proved that a quantity of nitrogen was here combined with the potassium, other experiments were instituted for the purpose of obtaining it separate. By combustion in oxygen gas, a part of the nitrogen, but not the whole, was obtained; but by distillation with red oxide of mercury, the product of nitrogen was greater. For the formation of ammonia from this nitrogen, as much hydrogen is wanted as was originally given out by the ammonia, in the ﬁrst part of the action of potassium upon it; but unless oxygen, as well ashydrogen, be sup-