48 I Slher — Origin of Jointed Structure. 
Walling’s paper appears in the proceedings of the Montreal 
meeting of the American Association for the Advancement of 
Science. Mr. Crosby’s, which is the more elaborate, was read 
to the Boston Society of Natura] History in October, 1882. 
Mr. Crosby shows, first, that the earthquake vibration, by 
subjecting the imperfectly elastic material through which it 
passes to alternate stress and tension, is competent to produce 
fractures; second, that such fractures would be normal to the 
direction of wave propagation, and therefore vertical, except in 
the immediate vicinity of the earthquake focus, where they 
might be oblique; third, that such fractures, although actually 
curved, would have generally so great a radius of curvature as 
to be sensibly plane; fourth, that they would be parallel; and 
fifth, that the suddenness of the earthquake shock would tend 
to produce smooth fractures even in heterogeneous material. 
All these points sustain the hypothesis, and their collective 
effect is to give it great strength. ‘There are, however, two 
features of joint structure with which the theory does not prima 
facie consist. 
In the first place the angle of intersection of two co-existent 
systems of joints is usually bigh. This is recognized as a diffi- 
culty by Mr. Crosby, and he says in explanation, ‘that, after 
the rocks have been broken by one set of joints, the layers or 
sheets thus formed possess a strong natural tendency to break 
at right angles; and, under such circumstances, oblique vibra- 
tions may give rise to rectangular fractures and blocks.” If I 
rightly understand him, he refers by the expression, “ strong 
natural tendency,” to the comparative ease with which an elon- 
gated body of amorphous material may be broken in the direc- 
tion of its least diameter, a property evidently dependent on 
the fact that fracture hringh the smallest diameter involves 
the overcoming of cohesion through a surface of minimum area. 
If this were the true explanation of the high angle assumed by 
secondary joints, each of the layers between primary joints 
would be affected independently and the planes of cross-joint- 
ing would be discontinuous. ith continuous planes of cross- 
jointing, the total cohesion overcome is not diminished by any 
modification of attitude. 
existing joints, and would initiate a series dividing the terrane 
into rectangular blocks. The defect of this explanation is that 
