if 
MATTER OF BLACK TOURMALINES. Al 
was then filled in nearly to the top of the crucible, and the latter, covered with its lid, 
was kept for a couple of hours at a strong heat in a powerful gas furnace. The iron 
filings became agglutinated, and the tourmaline, transformed into a highly vesicular 
mass, was almost completely decolourised. The coal at the top of the mixture remained 
unchanged until the lid was removed, when combustion of course ensued. A small piece 
of the tourmaline thus rendered practically white, when fused into a borax-bead coloured 
blue by copper oxide, immediately gave rise to red spots and streaks of Cu’ O, thus show- 
ing that the iron had not been converted into Fe’ O’, but that deoxidation had really been 
effected, as proved by the change of colour. ' 
(3) Experiment No. 2, with slight modifications, was repeated on other black crystals. 
All were decolourised, some assuming a pale-greyish or yellowish-white colour, and 
others a delicate pink tint. 
(4) A small, dark-green, translucent crystal of Brazilian tourmaline was treated as 
in experiment No. 2, but the heat was kept up somewhat longer. The crystal remained 
unfused, although slightly rounded on the edges; but it was rendered opaque and dull 
greyish-white in colour. In these green and other translucent examples of tourmaline, it 
is probable that the iron is only partially in the condition of Fe’ O*; and that in lightly- 
tinted examples it may be wholly present as Fe O. But in schorl or black tourmaline, I 
hold that it is essentially if not wholly in the state of ferro-ferric oxide, and that conse- 
quently the earlier view of the constitution of tourmaline is in this respect correct. 
Nore.—Want of leisure has prevented me from extending these investigations to 
other black or deeply-coloured silicates, with the exception of some melanites or black 
garnets from Frascati.. Ignited in an open crucible, these became fused or rounded, but 
retained their dark colour. Treated, as in experiment No. 2, in a reducing mixture in a 
covered crucible, they became, practically, white or greyish-white, but presented at the 
same time a singular phenomenon. Perfectly solid crystals (of sp. gr. averaging 3.80) 
separated into two portions, an outer and an inner portion. The outer part expanded to 
about three times the original bulk of the crystal, forming a thin sheath, within which lay 
the whitened but only partially-fused central portion, like a nut within its shell. This 
bears out an observation made some years ago by Rosenbusch, that many garnets differ 
externally and internally, as regards colour and composition—his remark applying 
especially to these black garnets. Damour, in his analysis of a black garnet from Frascati, 
obtained a small amount of titanic acid, and he attributed the black colour to the presence 
of titanic oxide, seeing that many light-coloured garnets contain more iron. But in these 
lightly coloured subspecies, the iron may be regarded as present in the condition of 
ferrous oxide, whilst in melanite (or, at least, in the black external portion of that mineral) 
it is probably entirely in the form of the ferro-ferric compound, Fe’ O‘. 

None of these crystals gave by fusion with sodium carbonate and nitre more than the very feeblest reaction 
of manganese. : 
Sec. IIL, 1886. 6. 
