BY THE CONDITIONS OF THEIR EXTRUSION. 7 



repose again becomes active, their immediate products, gases, are the first to be 

 thro-nm out; only after a considerable time, do we have any outbreak of lava. This 

 is essentially what we should expect on general mechanical grounds. The evidence of a 

 relatively low temperature in the lavas shows us that the heat beneath volcanos is not 

 nearly so great as it is in an ordinary blast furnace ; even if elevated to the temperature that 

 man frequently produces in the arts, there would be nothing to keep many of the enclosed 

 unmelted fragments from being completely fused by the action of their molten envelopes. 

 The interaction of the gases which escape from volcanos in the full tide of eruption, 

 is too complicated for understanding. It will be worth while, however, to consider some of 

 their effects. In the first place it is clear that the great expansion that takes place in 

 them must have an effect in lowering their temperature, and that of the substances 

 through which they pass. It is not possible, however, to separate this action from the 

 other effects, in the way of increased temperature which their evolution produces. The 

 movement of the gases would tend to produce this increase of heat in either of two 

 ways : by the direct mechanical effort of their friction ; or by the recombination near the 

 mouth of the volcano of the gases which had been separated in the more remote regions 

 whence the volcano draws its supplies of matter. I am inclined to give little value to 

 this action. There is, however, a way in which the gases act to liquify the lava which is 

 worthy of attention. When the pressure of the contained gases becomes very great, they 

 must necessarily tend to squeeze the rock containing them out in the direction of least 

 resistance. The softening that would give this liberty of movement, is most likely to 

 occur in sandstones and other highly siliceous rocks, and it is in them that we find the 

 largest proportion of water to be volatilized. At the time of their deposition our sand- 

 stones often contain as much as one fifteenth of their bulk of water, whUe the other more 

 compact rocks do not usually contain more than from one fiftieth to one hundredth of 

 their bulk of that substance. The escape of gases from the other more refractory materials 

 would, as far as it escaped, have to pass through the rigid framework of the rock. 

 Experience in coal and other mines shows us pretty clearly that the densest of our rocks 

 are somewhat permeable to gases, and that these gases, under the pressure of their gen- 

 eration from chemical changes, will make their way out through the interstices of the par- 

 ticles which compose the rock. In this way the gases of heated rocks, which are the 

 more permeable to such substances by virtue of their heating, probably find their way 

 towards the volcanic outlets. So we may reasonably assume that the gases which are the 

 main spring of volcanic activity, pass towards the vent in either of two ways : by 

 driving the rock before them when it is fluid enough to admit of it, and by interstitial 

 movement when this is not possible. 



The Lavas of Massive Eruptions. 



I propose now to consider the outlines of the probable history of that class of eruptions 

 commonly termed massive. It is only of late years that it has been made tolerably plain 

 to us that extensive outflows of lava occasionally, though rarely, take place without passing 

 through the craters of true volcanos. As yet we know but little of these extrusions; they 

 seem to have been formed in only a few of the known volcanic districts of the globe. 

 They have been observed in the Cordilleras of North and South America, in the Deccan, 



