70 
Proceedings of Royal Society of Edinburgh. 
fsESS. 
were in parts plentiful. Vegetation, too, will before long help to 
protect the surface, but at the time of my visit the only traces of 
this that could be found were a small fern and a small plant of 
lichen on the warm ground close to a nearly extinct fumarole in 
the crater itself. 
In and near the crater weathering showed itself in other forms. 
Immediately after the eruption the walls of the crater were nearly 
vertical, rising in some parts to a height of over 1600 feet above 
the crater floor. In the more rocky parts this has not been greatly 
changed, but elsewhere, through the action of rain and frost, 
constant landslips have been taking place, and the precipices are 
now no longer vertical, but are for the most part covered with a 
talus lying at the angle of repose. Again, the rugged “conical 
mounds,” of which hundreds were spread over the crater and the 
main earth-stream, forming a most conspicuous feature in the 
early photographs, are now smoothed and rounded to an extent 
almost inconceivable in so short a time. Another striking feature 
is the rapidity with which many of the rocks lying about the crater 
are crumbling down. These have been subjected to the action of 
steam and acid vapours in the heart of the mountain, by which 
almost all the constituents except silica have been removed.* 
When exposed to the air these blocks quickly fall to pieces, 
spherical flakes which soon fall into sand peeling off them with 
great ease. 
It is worth considering to what extent the gradual decomposition 
of the rock brought about the final catastrophe. The mountain 
was, as it were, bound together by a number of sheets of lava, 
which made it strong enough to resist the steam pressure beneath. 
During the past ten centuries these lava beds have been gradually 
decomposed along certain lines, and at length a time came when 
they were no longer able to resist the steam pressure, and the 
mountain was blown to pieces. May not this, rather than the 
sudden development of a large quantity of steam, as usually 
supposed, be the true history of the eruption ? 
The eruption of Bandaisan is certainly not the first of the kind 
that has occurred in Japan. Professor Sekiya has pointed out some 
* Analyses made by Mr Shimizu give the proportion of Si0 2 in this rock as 
91 '66 per cent., while in the natural rock it was only about 59 ‘6 per cent. 
