TRANSACTIONS OF SECTION C. 613 



rially increased the transmitting power. There is reason therefore to suppose that 

 water, under the considerable hydrostatic pressuies that exist beneath volcanic 

 mountains, and assisted by capillarity, may flow into the volcanic ducts with 

 facility, especially when aided by the intermittent relief of pressure afforded by 

 the rise and fall, or pulsations, of the column of lava. 



As the underground springs also are exhausted by the expulsion of the laro-e 

 volumes of water (as aqueous vapour), another agent comes into operation. When 

 by the continuance of this action, the level of the underground waters in the' 

 sedimentary strata under the volcano is lowered to below that of the sea-level so 

 that their hydrostatic pressure is no longer equal to that exercised by the sea-water 

 the current becomes reversed, and iustead of an outflow from the land, an inflow' 

 of salt water from the sea necessarily takes place through the same exeurrent 

 channels, and thus, taking the place of the displaced fresh water, finds its way to 

 the volcanic ducts. Then, from the exhaustion of the fresh-water supplies and 

 an impeded access of sea-water, the lava flows quietly and unaccompanied by the 

 violent explosions which mark the commencement of an eruption. If, on the other 

 hand, the sea-water gains access more freely through tlie more porous volcanic 

 materials, it may help to maintain, as in Stromboli and Kilauea, a constant volcanic 

 activity. In ordinary cases, however, when the inland waters, after the force of 

 the eruption is expended, regain the ascendant, they again exclude the sea- 

 water, and return to a state of equilibrium, which lasts until the strata are a-^ain 

 disturbed and fractured by a renewed eruption of lava. 



In conclusion, the author conceives that the first cause of volcanic action is the 

 welling up of the lava, in consequence of pressure due to slight contraction of 

 the earth's crust. Secondly, the fluid lava coming into contact with water stored 

 in the crevices of the masses of lava and ashes forming the volcano, the water is 

 at once flashed into steam, giving rise to powerful detonations and explosions. 

 Thirdly, follows an influx of water from the underlying sedimentary strata into 

 the ducts of the volcano , and, lastly, as these subterranean bodies of water are thus 

 converted into steam and expelled, the exhausted strata then serve as a channel to 

 an influx of sea-water into the volcano. A point is finally reached when, owino- 

 to the cessation of the powerful shocks and vibrations, and the excessive drainage 

 of the strata, the flow of the lava is effected quietly, and so continues until the 

 lava ceases to rise. 



6. The Connection hehveen the Intrusion of Volcanic Eoch and Volcanic 

 Eruptions. By Professor Sollas, M.A., F.E.S.E. 



In a volcanic eruption there are concerned, first, the elevation of the lava column 

 in the axial pipe of the volcano, and next the explosion by which the lava is 

 ejected into the air. The author attempts to find a vera cmisa for the latter. 

 Sorby's researches on included cavities prove that steam at a high tension must 

 have been everywhere present throughout plutouic rocks when these were in a 

 state of fusion, and the presence of steam in ejected lava is well known. Judd's 

 researches show that plutonic rocks are in many cases the solidified remains of 

 deep-seated lava from which volcanoes were supplied. It may therefore be shown 

 that the axial pipe of a volcano is occupied by fused rocks permeated bv steam, 

 which is probably in a liquid state, and the tension of which at any point in the 

 tube will depend on the hydrostatic pressure due to the lava column above it. 

 Any sudden diminution of this pressure will lead toa sudden expansion of the steam, 

 and 80 tend to produce a volcanic explosion. 



The mere elevation of the lava in the volcanic pipe cannot directly produce a 

 diminution of pressure, though an overflow at the surface of the ground would; 

 but this infers that the overflow of lava should precede an eruption, which is not 

 the case ; hence the author concludes that an overflow of lava from the sides of 

 the pipe takes place underground, and the pressure on the lava column being re- 

 duced beneath the point of outflow, an eruption follows. The abundant presence 

 of intruded sheets and dykes of igneous rock, known to occur beneath volcanic 

 cpnes, thus stands in close connection with the production of volcanic explosions. 



