370 ASH FROM THE ERUPTION AT ST. VINCENT.  [ Aug. 1902, 
Discussion. 
Mr. TEatt called attention to the small quantity of potash revealed 
by the analysis, and thought that this might possibly be due to the 
glassy part having been mechanically separated from the crystalline 
minerals in the air, as the sample analysed consisted mainly of such 
minerals. Hence, if the area could be found where the glass fell, 
the discovery might prove important from the agricultural point of 
view. Nature would have been more bountiful if she had let all 
the glass fall on Barbados, and the minerals in the sea. 
Prof. Jonnsron-Lavis pointed out the uselessness of such dust- 
analyses, except as applied to agricultural deductions at the actual 
spot where the material was collected. He had shown many years ago 
that, in studying a given volcanic deposit, its composition varied with 
the distance from the focus of eruption, owing to the winnowing 
effect of air-currents. In his paper on volcanic ejectamenta, he had 
shown that a volcanic dust may be quite different in composition from 
the initial explosive material that was the motive power of its ejection, 
so that it might be in part made up of the powder of quite different 
volcanic rocks, or even of sedimentary rocks forming the walls of the 
volcanic vent. Thus a volcanic dust might contain a large percentage 
of limestone or quartz of sedimentary origin. Only a careful 
selection of primary eruptive material near the vent, without any 
admixture of supplementary or accidental ejectamenta, could afford 
suitable material for an analysis that would throw any light on the 
nature of the magma of this eruption. 
Prof. Sottas doubted whether the small proportion of potash 
contained in these ashes stood in any necessary connection with the 
winnowing-away of the glass, especially if this amounted to half 
the bulk of the rock, as might well have been the case. The general 
result of analyses which had hitherto been made, was to show that 
the chemical composition of the total phenocrysts of a glassy rock 
did not differ widely from that of the glass itself. No doubt there 
was a general increase in the ratio of potash to soda on passing 
from phenocrysts to glass, but it was on the whole trifling, and not 
sufficient to lead us to expect that the glass of this eruption should 
contain more than, say, an additional 1 per cent. of potash above 
that in the phenocrysts. 
Mr. Prior said that he had examined the dust, and thought that 
it indicated the character of the lavas of eruption: they were 
probably hypersthene-augite-andesites, such as those of the Andean 
and Mexican chains. He had also examined the dust which fell 
on Barbados in 1812; this was finer in grain, but qualitatively of 
the same composition as the dust of 1902. ‘The amount of light 
material (pumice), however, was much larger, and that of magnetite 
less. he older fall is said to have increased the fertility of Bar- 
badian soils. 
