519 
of Edinburgh, Session 1874-75. 
are, by the influx of heated water, brought so near their respective 
boiling-points that a slight upheaval at a certain point of the tube 
is sufficent to carry the superincumbent layer to a point where, 
from its temperature and the slight diminution of pressure, a 
further generation of steam (instead of a condensation of that 
already formed) will be the result. 
Bunsen has shown that the mechanical force which this action 
developes is fully sufficient to account for the marvellous pheno- 
mena of a great eruption. Bunsen’s observations and my own 
agree in showing that it is somewhere about the middle of the 
column that the observed temperatures approach most nearly to 
those of the boiling-point due to the pressure. An upheaval of the 
layer at that point, through only a few feet, will be sufficient to 
generate instantaneously an additional volume of steam, the pressure 
of which will again further relieve that of the strata beneath, 
and so cause an additional volume to be generated there. The 
enormous force, which the phenomenon of the sudden upheaval of 
a small column of water is thus capable of calling forth cannot be 
spent in a single eruptive-shot, and hence the explanation of the 
fact that a great eruption lasts sometimes for four minutes. 
No theory of internal steam-cauldrons, filled in succession with 
steam and water, seems at all consistent with observed phenomena. 
It fails to explain how, in the abortive eruptions, no water seems 
to flow from the tube more than the small rivulet which steadily 
finds its way at the indentations over the rim of the basin. What 
flows over the margin, and it was great enough, considering its 
temperature, to cause considerable difficulty in retaining hold of 
the cord and rope, was due solely to the great commotion in the 
basin, and was apparently equal to the fall of a few inches in the 
level of water in the pool when the sudden upheaval had subsided. 
Further, Bunsen actually let his thermometer remain without injury 
at the bottom of the tube during a great eruption, having noted 
on it, a few moments before, a temperature of 9° C. below that of 
the boiling-point due to the pressure. The column erupted on that 
occasion he estimated at 43'3 metres or about 142 feet. 
It will be observed, that there is a remarkably sudden rise of 
temperature at a particular point of the column. Thus while the 
rise between depths of 10J feet and 18 feet is only 7° F., and that 
3 x 
VOL. VIII. 
