On the Domite Mountains of Central France. 147 



Now, it will generally be conceded, I imagine, that every vol- 

 cano which aspires to rise to any height above the surrounding 

 level will, as a preliminary measure, throw up a cone of ashes 

 and scoriae, and this we find to be actually the case both in 

 modern eruptions, and in these extinct but well-preserved craters 

 which form the bulk of the Auvergne range. When these 

 craters are formed the next thing that happens is that the lava 

 will begin to rise up in them, and it would continue to do so tUl 

 it filled them to the top, if it were not that the lateral pressure 

 of a fluid mass of such high specific gravity becomes too great 

 for walls of such loosely aggregated material to sustain, so that 

 almost invariably one side gives way, and the lava bursting 

 through deluges the surrounding plain. Now, in some of these 

 cones we have a means of estimating the exact height to which 

 the lava rose before the side of the cone prolapsed, for, according 

 to Scrope, in the crater of the Puy de la Vache this lava level is 

 still marked by a projecting ridge of light scoriaceous matter of 

 a reddish yellow colour, rich in specular iron, apparently part of 

 the frothy scum which formed upon the surface of the ebullient 

 lava, and adhered to the side of the vase at the moment of its 

 being emptied. We verified his observation as far as this ridge 

 of slag is concerned, and it certainly had all the appearance of 

 having been formed in the way he suggests. 



Now, this line is about 30 or 40 feet below the present ridge 

 of the crater, and allowing twice this amount for denudation we 

 may safely say that the lava rose to within 90 or 100 feet of the 

 edge of the crater before it burst its way through. But it must 

 be remembered that the density of these basaltic lavas was very 

 high, i.e. S'O or SI. If we suppose them to be replaced by a lava 

 like the Domite, having so much lower a specific gravity, it will 

 be readily seen that in that case the lava might completely fill 

 the crater before it began to overflow, which it would then do 

 regularly and evenly at every point of the circumference, without 

 any tendency to break down one of the sides of the crater. This 

 viscous semi-solid mass, permeated with gas bubbles, and with a 

 strong tendency to solidification, would then creep slowly down 

 the sides, forming those inverted bowl -shaped mountains which 

 are so peculiar and distinct in form from the ordinary craters. 



