TRANSACTIONS. 
79 
the inflammable matter of the ore itself, and a smouldering heat, 
never, I was assured, sufficient to occasion fusion, is kept up 
for several months. On examining the lumps of roasted ore, 
small nodules are observed of an iron grey colour, with some 
lustre, resembling in appearance the vitreous copper ore (ciiivre 
sulfure of Hatty). These nodules have been found, when as¬ 
sayed, to be very much richer in copper than the original ore. 
Their specific gravity I found to be 4’6, very nearly that of 
vitreous copper ore. By a few general experiments, I ascer¬ 
tained that they are not entirely soluble in heated nitric acid, 
and that the solution contains a small proportion of peroxide of 
iron, with a much larger one of oxide of copper. The yellow 
copper pyrites, in its natural state, was determined by Mr. Ri¬ 
chard Phillips, to consist of 
2 atoms of protosulphuret of iron, 
1 atom of bisulphuret of copper. 
“It should appear, therefore, that during torrefaction the 
bisulphuret of copper, by parting with one atom of sulphur, is 
converted into protosulphuret, which, by its aggregative attrac¬ 
tion, is collected into small nodules. This fact furnishes an 
additional example, to the few which were before known, of 
changes of molecular arrangement in bodies heated below their 
point of fusion; with this further circumstance, that the attrac¬ 
tion, which causes the aggregation of the particles, is sufficient 
to overcome the obstacle of interposed matter of a different 
kind. The only other instances of similar facts that at present 
occur to me, are presented by crystallites ; and by the products 
of Mr. Gregory Watt’s experiments on Basalt, some of the ap¬ 
pearances of which he supposes to have taken place after the 
fused mass had returned to a solid state. 
“In the second volume of Breislac’s Institutions Gcologiques , 
I was pleased to observe, two or three years after the foregoing 
fact had occurred to me, that a precisely similar observation had 
been made by Brocchi on the roasted copper ore of Agardo, 
which is also the yellow pyrites, and of quality, as to its propor¬ 
tion of copper, not exceeding that of the Parys Mountain. On 
breaking the lumps of roasted ore, similar nodules were ob¬ 
served ; and these, when assayed, were found to contain two 
thirds of their weight of copper, while the surrounding ore had 
lost greatly of its original proportion of that metal. In the cen¬ 
tral parts of some of those nodules, small fibres and plates of 
metallic copper were visible, an appearance which I have not 
observed in the roasted ore of Anglesey. The nodules, thus 
enriched in their proportion of metal, are picked out, and sub¬ 
jected to reducing processes. 
“ M. Breislac adds, that the torrefaction of the ore, at Agardo, 
