Page:Proceedings of the Royal Society of London Vol 60.djvu/427

This page needs to be proofread.
On the Occurrence o f Gallium Clay-ironston
395

being collected. The mother liquor from the last crop of crystals was evaporated almost to dryness to expel acid, and, after addition of water, rods of zinc wei-e immersed in the solution, which was then left undisturbed for forty-eight hours. The zinc was found to have been almost all dissolved. The precipitated metals and the residue of zinc were washed and dried. Residue H. The filtrate, after three precipitations with ammonium acetate, was mixed with the mother liquor of the ferrous chloride crystals from the first portion, so that the liquid then represented the whole of the ingot. It was diluted, mixed with an excess of ammonia and ammonium sulphide, to precipitate all the iron and metals of that group still in the solution, and filtered. The filtrate was evaporated to dryness and gently ignited to expel ammonium salts. A residue was left, which contained the alkaline earths and alkaline metals.

The Spectrographic Analysis of the Residues and of the Precipitates.

From the foregoing description it will be observed that by partial solution the metals precipitable by iron maybe looked for along with carbon, and, probably, some phosphides of iron and other metals. Such phosphides yield the flame spectra of the metals only, and not of the phosphorus combined with them. Precipitation with zinc in an acid solution was expected to give a deposit ( and R ) which would yield the spectra of copper, silver, bismuth, lead, thallium, and tin in the oxyhydrogen flame, if these metals were not already precipitated by the iron, and present in the residues A and B ; while D, E, and G are compounds which fall under the category of sesquioxide metals, including beryllium, aluminium, indium, gallium, and chromium. Of these, aluminium and beryllium were expected to show no spectra in the oxyhydrogen flame, and for these it was intended to use spark spectra.

The residue A, when dried in the water oven and gently heated, gave off fumes which indicated that an oil was present, and extraction with ether and subsequent evaporation did, indeed, yield a quantity of a brown oil.

The oxyhydrogen flame spectra of the substances separated were photographed, and the following are particulars regarding their spectra.

The insoluble residue A contained iron, manganese, copper, gallium, sodium, chromium, silver, and nickel.

The lines which served to identify the metals had the following wave-lengths:—

Iron . . . . 4808-0 4046-0 3929-8 3922 0 3904-8 3898-5 3886-5 3860-0 38570 3841-0 3834-0 3826-0 3824-5 3758-4