DEFORMATION OF ROCKS 1 89 



and yet accomplishes the same mass deformation, the term distrib- 

 utive fault is proposed. There may be distributive normal or 

 tension faults and distributive thrust or compressive faults. 



Relations of joints to bedding. — Joints have been classified 

 into tension joints and compression joints.' Tension joints ordi- 

 narily form nearly normal to bedding. This results from the 

 fact that the layers act as transmitters of forces, and that at any 

 given place one of the principal directions of stress is ordinarily 

 nearly normal to the layer, and the other two principal directions 

 of stress lie in the plane of the layer. That this tendency to thus 

 decompose the forces exists cannot be doubted. That it would 

 be the controlling tendency in the majority of cases could not be 

 asserted from analysis alone. However, examinations of vari- 

 ous regions during the past season has shown me that this is 

 often a controlling tendency, and that tension joints ordinarily 

 do form nearly at right angles to the bedding. It has already 

 been explained that in regions of simple folding there is one set 

 of tensile joints,^ and that in regions of complex folding there 

 are two sets of tension joints. The reverse statement may be 

 made, — that is, that where there are two sets of intersecting ten- 

 sion joints at right angles to each other normal to bedding, these 

 are evidence that the region is one of complex deformation. 



Compressive joints, in contrast with tensile joints, because 

 formed in shearing planes, are ordinarily inclined to the bed- 

 ding.3 This also results from the fact that one of the principal 

 directions of stress is usually nearly normal to bedding. Sup- 

 posing the maximum stress to be in the direction of the arrows 

 of Fig. 4, the mean stress at right angles to the plane of the 

 figure, and least stress in the plane of the figure and at right 

 angles to the maximum stress, in other words normal to the bed- 

 ding, the position of the joints will be as shown, their planes 

 being normal to the figure in which they are represented in sec- 

 tion. This results from the fact that the direction of least stress 



' Loc. cit., (Aj, pp. 668-671 ; (B), pp. 609-613. 

 = Loc. cit., (A), p. 669; (B), pp. 609-610,. 



M.oc. cit.. ( A ). ]) 671; (H). p. 613. 



