JOINTING. 73 



bert* also draws a contrast between the divisions of a mass due to 

 shrinkage cracks and those known as jointing. 



There can be no question that jointing is due to mechanical causes ; 

 for joint planes cut through conglomerates with almost the same regu- 

 larity that they divide the most homogeneous rocks. They must, there- 

 fore, be due either to tensile stresses or to compressive stresses. In the 

 former case the joints must gape when first formed. I am fully satisfied 

 that Professor King was correct in asserting that the surfaces are usually 

 in close contact immediately after rupture. In very many cases disloca- 

 tion has taken place on jointed surfaces and, where this has occurred, 

 any irregularity in the surface will force the walls asunder. 



It has been shown in the preceding pages that the columnar structure 

 of lava is easily accounted for, but I know of no way in which such a 

 S} r stem of divisions as occurs in jointing can be accounted for by ten- 

 sion, f 



On the other hand, it is not difficult to show that most of the phe- 

 nomena of joints are fully accounted for by pressure, direct or inclined ; 

 but to this statement there is one exception. If joints are produced by 

 pressure, they are due to a tendency of the rock to move in opposite 

 directions along the joint plane; or, in other words, to a tendency to 

 faulting. Hence, if pressure is the cause, there is no distinction, except- 

 ing one of degree, between joints and faults. 



There is no doubt whatever that faults are often met Avith on joint 

 planes, yet this association is no proof that the two phenomena are not, 

 as they have often been assumed to be, independent of one another. 

 But the study of many thousand divisional planes which would cer- 

 tainly be regarded as joints by almost every geologist has led me to the 

 conclusion that the jointing and the faulting are concomitant. The 

 faults are often extremely small, but it is very rarely that in a system 

 of joint planes throws of an eighth of an inch or less cannot be detected ; 

 and where the rock is hard, slickensided surfaces will lie found even 

 when the relative movement is much less than an eighth of an inch. 

 Few geologists have been in the habit of looking for faults of such tiny 

 dimensions, and I believe that the distinction between faults and joints 

 has arisen from this omission. Partings due to tension would be free 

 from slickensides. 



It might be thought that a refutation of this conclusion is to be found 



'6 



* Am. Jour. Sci., vol. 24, 1882, p. 50. 



t Divisional surfaces produced by pressure differ from those produced by tension in a manner 

 which is distinguished even in ordinary parlance. Rupture under tension is only another name 

 for tearing, while division under pressure always involves as an essential feature that sort of cutting 

 which is performed with scissors or shears. Thus the very similes employed in describing rock 



fractures often indicate an instinctive perception of the mechanical pr sses involved, even when 



an attempt is made to reconcile phenomena with a less natural theory. 



