274 
THE MEDITERRANEAN NATURALIST 
2. Microcrystalline pumice, in wliicli surface ' 
cooling has produced pyroxene and ieueite. The ' 
amount of vitreous base diminishes as we reach the j 
top of this division, and is replaced by formed 
material. 
3. The pumiceous ash-bed in which the cement- 
ing vitreous base is nearly all destroyed, so that 
cohesion has become so feeble that the formed mat- 
ter separates, producing an ash composed of crys- 
tals and microliths. The difference is very similar 
to the results of crystallization of a salt in the form 
of large crystals by a slow process, or, in the pre- 
paration of the granular state, by a quick one, as 
table salt, pure Ferrous sulphate, and oxalic acid, 
as they are met with in commerce. 
The increase of the percentage of silica has the 
effect of rendering acid-rocks less easily crystal - 
lizable, just as the amorphous form of sugar retards 
crystallization of other bodies with which it is 
mixed. For the same cause, the viscosity of the 
rock is increased, so that the escape of the enclosed 
gaseous bubbles takes place with greater difficulty; 
and, as a result, the pumiceous character is far 
more common amongst such rocks. 
Mode of Formation and Structure of Scoria . — 
This product, which is often erroneously grouped 
together with pumice, is that spongy variety of lava 
which covers or underlies a stream. When the 
magma does not contain sufficient volatile cons- 
tituents to tear it asunder before it issues from 
the volcanic vent, it will pour down the slopes of 
the cone, giving up what remnants of aqueous mat- 
ter are still dissolved within if Should this be 
considerable in amount, and the temperature of 
the lava rather low, in basic examples we shall get 
an irregular broken-up cinder-like mass, that will 
continue to float on the surface, and cover it in 
some cases to a great thickness. Prof, ljudd, who 
fully appreciates this fact, gives a striking example 
of this side by side with an equally interesting- 
illustration of the opposite condition. On the 
contrary, should a basic magma be remarkably de- 
void of dissolved water, its surface wili not be bro- 
ken up, and it will assume forms like any other 
viscid body in movement. In the case of a water- 
bearing acid lava, the scoria surface will be much 
thicker, on account of the difficulty with which the 
gaseous materials escape in consequence of the 
viscidity of the paste: whilst in nearly nonaqui- 
ferous acid lavas the surface figure.- that result will 
be more marked, and more characteristic of an 
intensely viscous magma, as illustrated in the 
marnmelon volcanoes, such as the islands of Reu- 
nion, Hawaii, the obsidian stream of Yulcano, or 
some of the central French groups. On the other 
hand, the Vesuvian lavas of 1858 and 1S72. as 
pointed out by Judd, are respectively typical of 
aquiferous and non-aquiferous magmas, which 
may be further illustrated by the ti mbyte of Monte 
Olibano, and the Lava deFArso of Ischia. 
From the mode of formation of scox-ia we must 
expect it to exhibit two very marked differences in 
structure and mineral composition from pumice. 
In the latter the vesicular cavities are of all sizes, 
ranging down to the minutest dimensions, which 
are the most abundant and marked characteristic 
of pumice. This is due to the intermolecular se- 
paration of steam and its union sopra loco into 
vesicles of varying dimensions. In the case ot the 
scoria, the gases are derived from the whole thick- 
ness of the subjacent lava, whu h, in rising ii the 
mass, further unite together, so that the cavities 
are rarely of microscopic size, and may reach very 
great dimensions; and unless the bubbles be of a 
certain size, the large area of their surface in pro- 
portion to their volume increases, so that the fric- 
tion is so much that they could not rise in the 
viscid mass. In the case of pumice we have the 
vesicular structure developing in a complete or 
nearly vitreous magma, which is the principal cause 
of rapid solidification; but in so Ha the bubbles ot 
hot gas that rise from the bottom, which., trom 
being more protected, is the hottest part, through 
a magma already far advanced in crystallization, 
would help to prevent or ward off the cooling of 
the surface. Besides, the scoria will cool slowly, 
resting as it does on the surface ..of a highly-heated 
mass. We therefore may sum up Ip stating that 
pumice is filled by vesicles of all siz.s, but mostly 
small, and approaches the vitroou> state, whilst 
scoria only contains vesicles of large size, and ap- 
proaches a crystalline si ueture. The cjectanvnta 
during strombolian action is a true scoria, being 
dependent upon borrowed steam that rises in the 
magma column, and forms the vesicles. 
I n la vas the presence of vesicular cavities is no 
proof of the actual amount of original vapour, for 
the latter will be allowed to more easily escape in 
