﻿Vol. 64.] QUANTITATIVE METHODS TO THE STUDY OF BOCKS. 229 



wanted is the careful study of more and better specimens, unaltered 

 by infiltration, chemical action, or surface-weathering, collected so 

 as to be specially suitable for this subject. 



It will thus be seen that, in some cases, the results for different 

 specimens agree sufficiently well, and therefore the mean may be 

 relied upon as approximately correct. In other cases, there are 

 great differences, sometimes certainly due to the presence of too 

 much carbonate of lime, and sometimes to the specimens being 

 too much swollen and weathered. Collecting together those cases 

 which seem most trustworthy for calculation, we may compile the 

 following list : — 



Table IX. 



Coal-Measure shales 13"2 



Slate-rocks, Moffat 3-6 



Slate-rocks, Hele 24 



Slates, Westmorland 0*49 



Slate, Penrhyn 0*24 



Alluvial clay, etc o2'0 



Tertiary clay 28-8 



Bouldei-Clay 24-4 



Liassic clay 24'4 



Gault 24-0 



It will thus be seen that there is a fairly-uniform decrease in the 

 amount of cavities, in passing from clays which have been subjected 

 to very little pressure, down to the oldest rocks ; and a most marked 

 decrease in those with well-developed slaty cleavage. The question 

 then arises whether the compression is due to age or to pressure. 

 Although quite prepared to believe that mere age may have some 

 effect, when combined with pressure, yet taking all into con- 

 sideration, and bearing in mind my experiments with clay, it seems 

 to me more probable that the chief cause of the compression was 

 the pressure of superincumbent rock, or of that which developed 

 slaty cleavage. 



Though it seems almost certain that the amount of interspaces 

 varies in some way inversely as the pressure, their exact relation is 

 unknown. Possibly it could be learned by experiments with a 

 testing-machine, by means of which suitable pressures could be 

 applied continuously for a long time. It seems, however, unde- 

 sirable to delay the publication of this paper, and better to make 

 use of the data now known. It is clear that the law must be 

 of such a kind that the effect of a given increase in pressure is 

 much greater on material which has been slightly compressed and 

 contains a large amount of cavities, than on highly-compressed 

 rocks. It is, of course, not the compression of the cavities, but the 

 deformation of the solid particles involved in fi.lling them up which 

 is" of prime importance; and this must be relatively greater and 

 greater as their amount becomes smaller and smaller. This must 

 have been brought about by the slow giving- way of minute grains 

 during long geological periods, and not by the sudden fracture 

 of large objects, as taken into account in studying the strength 

 of materials for engineering purposes. 



The percentage of cavities in fine-grained clays of great anti- 

 quity but never exposed to the pressure of more than a few feet of 

 superincumbent material, is about 33 per cent. ; and, after careful 



