485 
of Edinhurgh, Session 1883-84. 
the distribution of the substances among the different species of 
constituent minerals. This vitreous matter can indeed contain an 
indeterminate quantity of the different bases. On the other hand, 
the difficulties of the calculation are all the greater, as the consti- 
tuent minerals of the ashes may contain, as isomorphs, the bases 
which the analysis suggests. It is none the less true, however, 
that the percentage composition expressed by the analysis supports 
the preceding mineralogical determinations, without permitting the 
species to be precisely determined. It agrees with the interpreta- 
tion that the magma from which the ashes were formed belongs to 
the augite-andesites. 
The vitreous and mineral fragments we have just described from 
the Krakatoa eruption being identical with those which we encounter 
in deep sea sediments, we may conclude that both have a similar 
origin. In certain cases, however, we have in place of augite a 
predominance of hornblende, and sometimes black mica is abundant. 
Again, we find more or less fragmentary crystals of peridote, of magne- 
tite, of sanidine, and, more rarely, of leu cite and of hauyne. We can 
easily understand this variation in composition, following the nature 
of the magma from which the ashes collected in different regions of 
the sea were derived. But in all cases it is the predominance of 
vitreous particles, with their special structure, which indicates most 
clearly the volcanic nature of the inorganic constituents of a sediment. 
If now we consider the conditions which govern the distribution 
of ashes in the atmosphere or at the bottom of the sea, we shall 
be able to show how it is that there is generally a predominance of 
vitreous particles in these ashes. In the first place, these are vitreous 
matters rather than minerals, properly so called, from the moment 
of ejection from the crater. Moreover, we should, in a general way, 
not expect to find that incoherent eruptive matters, which are spread 
out at a distance from the volcano, present a perfectly identical 
composition with those other loose products such as lapilli, volcanic 
bombs, and scoriae, which are projected only a short distance from 
the focus of eruption. Even where there exists a perfect chemical 
and mineralogical identity, in the crater itself, between the lavas 
and the pulverulent materials of the same eruption (the supposition 
being that the ashes arise simply from the trituration of the lavas), 
we can easily understand that these latter, being carried far and wide 
