Crosby.] 72 [October 4, 



on the classification and origin of joint- 

 structures. 



BY W. O. CROSBY. 



The attention of geologists has been recently attracted to the 

 subject of this paper through the interesting articles by Mr. Gil- 

 bert and Professor Le Conte in the American Journal of Science for 

 January, March and July of the current year. In his second paper, 

 where the origin of joint-structure is more particularly discussed, 

 Mr. Gilbert has made it pretty clear that, notwithstanding the re- 

 markable efficiency of dessication in producing approximately ver- 

 tical fractures in recently de23osited sediments, as in .the examples 

 cited by Professor Le Conte, these cracks are essentially unlike the 

 planes of division in the Post-tertiary clays of the Great Salt 

 Lake Desert, and do not possess, as the latter unquestionably do, 

 the characteristics of ordinary joints. And it seems, further, as 

 if his arguments to show that the intersecting systems of continu- 

 ous and parallel joints, so universally present in sedimentary for- 

 mations, can not be due to either shrinkage or compression, must 

 carry conviction to all minds. In short, Mr. Gilbert has conclu- 

 sively demonstrated that parallel jointing must be quite distinct 

 in its origin, as it usually is in its characteristics, from both shrink- 

 age-cracks and slaty cleavage. 



Having reached this point, Mr. Gilbert follows Professor Dana in 

 so restricting the definition of joint-structure as to exclude all 

 shrinkage-cracks, as well as slaty cleavage. This classification 

 appears to the present writer of doubtful utility, chiefly because 

 of its impracticability. A very large proportion of the planes of 

 division in eruptive rocks are due to shrinkage, this cause being 

 clearly recognized by all in the columnar structure of basalt; and 

 it is in many cases simply impossible, at least with our present 

 knowledge, to distinguish, in this class of rocks, the cracks due 

 to shrinkage from those produced by other causes. Similarly, 

 there can be no question that the dessication, consolidation and 

 crystallization of sediments develop a tendency to contraction ; 

 and, as I have attempted to show, in the Geological Magazine for 

 September, 1881, if not immediately, yet sooner or later, this con- 

 traction must take place. When it does occur, fractures are 



