CRYSTALLINE STRUCTURES OF ROCKS. 189 



parts of the volcanic rocks. I have shown (Geol. 

 Trans.) that silica can he sublimed by heat ; and the \r/* 

 same fact has been affirmed to occur*in the volcanic 

 products of Vesuvius, by observers whose testimony 

 cannot be questioned. It is possible that compound ^ 

 minerals may be subject to the same laws ; and it is 

 also perfectly intelligible, how in a fluid or tenacious 

 rock containing the cavities produced by inflation, 

 those minerals which have sometimes crystallized in 

 the general mass, should have also protruded them- 

 selves into the cavities. 



There are probably, or possibly, therefore, two 

 origins to be assigned to the amygdaloidal nodules, 

 both of the trap rocks and the volcanic products ; 

 however the mode of explaining the igneous method 

 may here differ from that which has been adopted by 

 those who were more anxious to believe than able to 

 explain. Admitting them both, the question respect- 

 ing the igneous origin of the amygdaloidal bases of 

 the trap rocks, rests precisely on the same foundation 

 as before ; as the essential circumstance consists, not 

 in the presence of the nodule, but in the formation of 

 the cavity which contains it. 



Of the Nature of the concretionary Structure. 



It remains to see if any light can be thrown on the 

 general nature of this mysterious process. That it 

 differs essentially from crystallization, was already 

 noticed. It is not concerned^ either in the disposition 

 of original and similar molecules, or in arranging them 

 into geometric forms. Yet its phenomena bespeak a 

 tendency in the particles, or finer fragments consti- 

 tuting stones, to arrange themselves by a predominant 

 attraction, into certain forms rather than others ; 

 however irregular, or uninfluenced by geometrical 



