518 
Proceedings of the Royal Society 
pressure, and approaches it more nearly the nearer the moment of 
a great eruption. 
He argues, therefore, that immediately before such an eruption 
only a small shock will be sufficient to vaporise a large mass of 
the whole column, and so to displace the whole column above. 
Now, it is a fact that the column is constantly subject to 
such shocks, which occur at intervals of a few hours, and are 
more frequent as a great eruption is approaching. Bunsen accounts 
for these shocks, which are in fact abortive attempts at an 
eruption, in this manner. He observes that it is a feature of 
most of the Icelandic warm springs that, periodically, at certain 
points, great bubbles of steam get formed, and rising soon condense 
in the colder strata above. This is well seen in these rocky 
cavities, 10 or 12 feet deep, which exist in that remarkable region 
of springs and mud-cauldrons in the immediate vicinity of the 
Great Geysir. I observed also something of a like phenomenon at 
the “quhar” by the side of the lake at Laugardalr, where we rested 
on the way between Thingvalla and the Geysir. At that spring, 
however, as the depth of water is quite inconsiderable, the effect is 
more of a continuous bursting of great bubbles of steam on the 
surface, as no condensation takes place, the water being at a tem- 
perature close on the boiling-point. Now Bunsen argues that, if 
at some point in the in-carrying ducts of the Geysir-column (and 
the existence of these ducts is proved by the constant overflow of 
water from the basin), the temperature of the layer of water gets 
raised above the boiling-point due to the pressure, owing to the 
great heat of the surrounding rocks, then a sudden generation of 
steam is the result, and a rise of that steam in the column itself. 
This great bubble is soon condensed, while at the same time its 
sudden formation cooled the water at the point in the duct where 
it was formed. The phenomenon, therefore, possesses a periodic 
character, and the explanation, it must be admitted, seems to account 
well for the conical water-hill, as Bunsen aptly terms it, the sudden 
upheaval of which in the centre of the basin is an invariable 
accompaniment of these subterranean explosions, often of very 
great violence, which are heard and felt recurring at intervals 
under the Geysir cone. The grand display of a great eruption, 
however, does not occur until the temperatures in the whole column 
