OF THE HAWAIIAN ISLANDS. 451 



great pressure it can be heated red hot without assuming the condition of vapor, and that 

 sudden motion or changes of pressure, such as we know are constantly taking place in the 

 earth's crust, might bring about a contact, in such a way as to cause an expansion of the 

 water into steam, and cause an elevation of the combined water and melted mass to 

 the surface, we must point to the quiet welling out of the Hawaiian lavas, and ask for an 

 explanation of that. 



What we see with these lavas is quite contrary to the supposition that water causes their 

 activity. Often they pass into the ocean and cool rapidly beneath the water ; at other times 

 when the motion is violent, the water obtains access to the interior of the mass and cools it 

 still more rapidly, breaking it into sand. Where the fresh water gains access to the incan- 

 descent lava the effect is to cool it more or less rapidly in proportion to the quantity of 

 water. Everywhere on the surface the tendency of water is to cool the melted mass by its 

 evaporation, and whatever its action may be at great depths, under great pressure, on 

 being set free on the surface it should cool the mass from which it escapes. There is not 

 enough water set free from any volcano known, to account for the elevation of such vast 

 weights of lava to such great heights. To do the work steam must pass from a high tension 

 to a lower one, and must finally escape into the air. In its high tension it may be con- 

 densed to water by extreme pressure, but on reaching the atmosphere it must lose all this 

 tension and become ordinary steam at the pressure of sixteen pounds to the square inch, or 

 unite chemically with the lava. Analyses show us that the last is not the case, and no more 

 steam is evolved from the Hawaiian volcano than must result from the rainfall on the 

 heated surface of the exposed lava. Prof Dana intimates that the rainfall may be the fuel 

 of the volcano; if so, why should not constant eruption attend a season of rain? I have 

 been on the banks of the lake of liquid fire during a severe rain, and the water that fell on 

 the surface, instead of stimulating to increased action, darkened the crust, and nearly closed 

 the vent. 



We know so little of the laws of combined heat and pressure, that at present theory runs 

 wild on the condition of the interior of our globe. One supposes a solid core surrounded 

 by a melted coat, which in turn is covered by the solid crust of the earth ; another supposes 

 pockets or reservoirs of heated matter left in the solid crust of the cooling globe. All, 

 however, are obliged to admit some source for the great streams of lava which are poured 

 out upon the earth's surface. To explain all the phenomena of volcanoes it matters little 

 what the interior of the earth may be, so long as we have a crust and a bed of melted rock 

 below it. In the earliest ages of the globe, when the crust was thin, it was frequently broken 

 as the earth cooled and contracted, owing to its being a poor conductor of heat, and con- 

 tracting on its surface faster than it transmitted heat from beneath, and the escaping caloric 

 found vent in numerous volcanoes over these fissures, precisely as on a smaller scale the 

 surface of the lava in Kilauea hardens and cracks, and allows the lava to boil up through 

 the cracks. As the crust became thicker, the rate of cooling diminished, and the escaping 

 heat needed fewer orifices, so that volcanoes became extinct. But the cooling process does 

 not go on perfectly regularly ; the crust is a better conductor in some places than in others, 

 and this conductive power may vary from time to time. A high mountain with its rough 

 surface abstracts more caloric from the earth by radiation than a smooth lake ; continents 

 more than the ocean. The balance is destroyed, and caloric must be transmitted through 

 the earth to restore it. In what way the temperature of the earth's crust, at moderate 



