OF MONTE SOMMA AND VESUVIUS. Ti 
Microscopical Examination.—The sanidine occurs in large well- 
formed crystals similar in all respects to that of the lower sub- 
division. 
Amphibole differs only by its greater size and perhaps greater 
abundance, due probably to the smaller relative bulk of the matrix. 
Pyroxene is much commoner in this upper subdivision. It shows 
in section the usual light pea-green colour, and graduates in size 
from large crystals down to the microliths. 
Magnetite is pretty abundant in crystals of various dimensions, 
which may be enveloped in the amphibole or pyroxene, and also are 
sometimes slightly decomposed. 
Biotite is moderately common in small well-formed crystals, and 
similar to that from the other pumice. 
Melanite is perhaps more common in this subdivision, though part 
is no doubt erratic. 
Tt is in the ground-mass that the two subdivisions of the pumice 
differ so much from each other. In the first case we found the 
amorphous base predominating over the formed material, but in the 
middle subdivision the conditions are reversed. In the most ad- 
vanced stage of crystallization we find almost the whole of the base 
omposed of microliths of pyroxene and biotite. 
The first of these minerals transmits a feeble green light, more 
intense proportionally to the size of the prism. The average of 
twelve measurements for the angle of extinction gave 414°, and 
the extreme variations were confined within narrow limits. The 
pyroxenes range from the merest trichite, or belonite, through mi- 
eroliths most variable in size, up to large crystals of that mineral. 
It was impossible to find a single microlith of amphibole. 
The biotite, like the pyroxene, is most variable in size, occurring 
nearly always as well-formed hexagonal plates, either single grouped 
together, or in pairs, but more often superposed one upon another. 
They are of the usual sepia-coloured, very polychroic kind. 
In addition to these two minerals, there are a few small mine- 
rals or microliths of magnetite, which are sometimes altered and at 
others not so. 
As already mentioned, the glassy base may be almost absent; but 
it seems that a small quantity always remains. 
When we examine that part which forms the border-line between 
the two subdivisions, we find that the crystalline structure is inter- 
mediate in development between the very vitreous white and the 
highly microlithic green pumice. In fact, it is possible to collect 
all gradations from one to the other. The green colour of the 
middle subdivision is due to the abundance of formed pyroxenic 
material. 
The pisolitic tufa proves, on microscopical examination, to consist of 
the same minerals as compose the pumice broken up into fine frag- 
ments. The pisolitic concretions sometimes attain the size of a 
walnut, but usually average that of a good-sized pea. When split 
open or sectionized, they are seen to be composed of coarse grains en- 
veloped in concentric shells of a very fine brown powdery substance. 
