OF VOLCANIC ROCKS. 57 
demonstrate in another chapter that the changes of level which have been connected 
with the ejection of rocks, appear to confirm these suppositions. Suffice it here to 
draw the necessary consequence, that the increasing tension must finally have had the 
effect of rupturing the newly consolidated masses. If in the intermediate time the 
masses filling the fissures of the first epoch had, at least partially, been consolidated, 
then a new system of fractures would be opened, within the limits of, though not coin- 
ciding with, the first. These fractures of the second epoch would descend to greater 
depth than those of the first, and in allowing the access of water to masses situated in 
lower regions and being of a more basic composition, would open for these the way to 
the surface. By the repeated occurrence of this or similar processes, the theater of 
action from which the rocky masses were conducted to the surface, might descend, by 
steps, into considerable depth within a comparatively short period, and thus there 
could be produced a great diversity among the rocks emitted through one system of 
fractures, though this diversity would be regulated by definite relations in regard to 
the nature and succession of the rocks ejected. The process would come to an end 
when the solidification of matter and the formation of fractures had descended to those 
masses, the state of aggregation of which was such as no longer to allow them to crys- 
tallize when the pressure was diminished, and in this way any further increase of 
volume would be prevented. The matter filling the fractures would now solidify, and 
the communication of the interior with the surface be cut off, with the exception of 
the voleanic channels. The resistance offered by the crust of the globe would hence 
be greater than it had been before, and there would follow another era of repose, 
longer than that which had preceded the era of eruptive activity. 
The application which may be made of these processes suggested by theory, to 
the explanation of the actual correlations of eruptive rocks in regard to their age, 
chemical composition and geographical distribution, is obvious. We may, indeed, 
venture to deduce @ priori the history of eruptive action, in its main features, from the 
hypothesis of Sartorius, and the assumption that silicates will increase in volume by 
perfect crystallization. We should have to conclude that ma remote period, when 
the crystallized crust and the sedimentary shell of the globe were inferior in aggregate 
thickness, fractures and eruptions of rocky matter would have been of frequent occur- 
rence, and that highly silicious compounds should have prevailed among the ejected 
masses. Little, if anything of them, is probably visible at the present surface, as the 
rocks of the Azoic and Paleozoic formations are probably the monuments of an already 
far advanced stage of the development of our planet. All the distinguishing features 
of the eruptive rocks of these periods (including of the Palsozoic age only the Silurian 
and the first part of the Devonian), such as the great number and individualization of 
the granitie districts, the independence of each of them in the subtler differences 
regarding the commencement and further development of the eruptive activity as well 
as the peculiar nature of the rocks ejected, the great preponderance of highly silicious 
compounds, the common association with them of small quantities of basic rocks, and 
the slight increase of the proportion of the latter in the Devonian period—all these 
phenomena are easily understood, without further explanation, in the hght of our 
hy pothesis. 
(99) 
