32 



causing the fracture in the interior is due purely to longitudinal 

 push in the outer part of the column. That outer part, therefore, 

 will not be subjected to the pull at all, and so the enlarging circular 

 conchoidal fracture should be expected to stop short without 

 penetrating to the outside of the column, especially at the angles. 

 On the event of the central part cracking, and so ceasing to bear a 

 pull, the outer part being less resisted than before, would increase in 

 length in the immediate neighbourhood of the new internal fissure, 

 and so would bring parts nearer the circumference than before into 

 the condition of being subject to a pulling stress. Also, the reverber- 

 ation or tremor at the instant of the cracking might, it seems reason- 

 able to suppose, carry the advancing circular edge of the fissure some- 

 what farther out than the region which would be subjected to a pull 

 if the action were slow, instead of being by a start. The appear- 

 ances of the cross-joints, with the central area of each like a circular 

 or oval flattish face, or like the convex or concave form of a watch- 

 glass, but not extending out quite to the angles, and usually not 

 quite out to the sides of the columns, seem to be in accordance 

 with the suppositions here made, and to give considerable corro- 

 boration to them. 



The cracks, if formed as supposed, without extending quite to the 

 outside of the column, would constitute places of weakness, from 

 which, under the shattering influence of earthquakes or other causes, 

 fresh fractures would readily proceed quite to the outside, severing 

 the columns completely across ; but these fresh fractures occurring in 

 ways quite different from those in which the original circular ones 

 had done, could not be expected to be in continuity with the 

 supposed original circularly terminating fissures. Thus is accounted 

 for, the approximately circular outer boundary to the flattish or 

 lunette-shaped middle part of the cross-joint, which is very com- 

 monly to be seen. 



On a recent visit to the Giant's Causeway, in the Summer 

 of 1869, Professor Thomson had noticed some phenomena 

 tending to confirm his views. He met with several instances 

 in which a small mass of stone, different in texture and in hard- 

 ness from the rest of the basalt, showed itself in the cross-joint 



