OF MONTE SOMMA AND VESUVIUS. 39 
temperatures between that of the fluid, igneous matter, and the cold 
ground-water. 
This is certainly somewhat at variance with our knowledge of 
the critical point of water; but all must admit that laboratory ex- 
periments may fall tar short of the conditions that exist in nature. 
The presence of water enclosed in the cavities of quartz in granite, 
seems to demonstrate the possibility of its existing in a liquid 
state at very high temperatures, provided the pressure is sufficient. 
Let us, however, suppose the critical point of the liquid state 
of water to be passed,—the interstices of the porous materials in 
immediate contact with the lava would be filled by the steam, and 
thus prevent the further access of the cooler liquid. In fact, as we 
shall see later on, a gradual absorption of such water may be going 
on by the incandescent rock, which water escaping, may form the 
principal part of the vapour rising from the volcano. If such liquid 
be derived from the sea, the liberation of those salts may take place 
that gave rise to the theory of marine infiltration Seing the actual 
cause of volcanos. Without continuing for the present a more exten- 
sive consideration of the subject than is necessary to our argument, 
itmay be mentioned that what has been said is not supposed to support 
or contradict any doctrine of vulcanicity. [It would in fact be safer 
to consider these as purely superficial phenomena, quite independent 
of those far more extensive primary causes of the eruption of igne- 
ous matter, whatever they may be. 
When, however, the conditions are changed from a tube filled with 
molten rock reaching far above sea-level to a deep cavity 850 metres 
below that line, from which is escaping at high pressure a column 
of vapour and gases, carrying with it fragments of pumice and 
other rocks, we should look for different results. We should ex- 
pect that water would enter from all sides of the hollow to be im- 
mediately carried upwards in a pulverized state, mixed with the 
finer pumiceous dust, well suited to set as a fairly hard mass such 
as forms the latter part of the materials ejected by the Plinian 
eruption. 
The rush of vapour through the chimney would probably accele- 
rate the influx of water in an analogous manner to that by which 
the liquid rises in the supply-tube of a vaporizer by the blast of 
air crossing its extremity. 
As the eruptive activity diminished, the volcano would assume 
a submarine character, so far as the eruptive vent was concerned, 
somewhat like the island of Santorin. It does not seem very im- 
probable that from time to time there were more powerful spurts 
consisting of a mixture of fine matter with water, so as to form mud. 
Such humid material deposited in large quantities, especially on 
the outer edge of the crater, and rendered more liquid by the 
torrents of rain derived from the condensed vapour, would form an 
immense accumulation of the so-called ‘“ lava d’ acqua,” which would 
sweep down the flanks of the mountain. On its arriving at any 
obstacle, such as a town like Herculaneum, this would be over- 
whelmed in a natural cement, that must almost immediately set 
