Desquamation of certain Rocks. 247 



the bed, but it is actually divided into columns as regular as 

 those which are found in the trap rock, and bearing an average 

 diameter of about two feet. On examining the transverse sec- 

 tions of these columns, it is found that many of them are marked 

 by concentric lines of red and white, alternately preserving a 

 circular form near the axis of the column, but gradually losing 

 it so as to become angular towards the margin where the ap- 

 proximate columns begin to interfere with each other. Even in 

 some part of this sand-stone bed, where no actual division into 

 columns has taken place, the same sets of concentric groups of 

 lines are to be found; each set deviating from the circular form 

 as it begins to interfere with the surrounding ones, and indi- 

 cating a future possible division of the whole into prismatic 

 shapes. Here no action of the atmosphere can be supposed to 

 have produced this effect, which, on the contrary, marks an 

 original concretionary disposition in the rock, analogous to that 

 which, in all probability, exists in the columnar traps, but here 

 rendered visible by the differences of colour, as it is there 

 concealed by the uniformity of tint. 



Having thus proved that the spheroidal exfoliation of trap 

 may proceed from a peculiar concretionary structure, as has 

 been generally received among geologists, I was desirous of 

 ascertaining whether cases might not also exist in which, as in 

 the artificial forms of granite before described, it was due to the 

 action of the atmosphere alone, and independent of any internal 

 arrangement of the parts of the rock. It was evident that this 

 could only be done by examining blocks which had received 

 their forms by art, or by accidental fractures, and after no long 

 search with this view, numerous opportunities were afforded 

 in several places where such blocks of basalt and greenstone 

 had been broken for the purposes of building agricultural 

 dykes. 



In many cases of trap, as of granite, no exfoliating tendency 

 seems to exist, and therefore it may often be fruitless to search 

 for illustrations of this subject. But in those to which I now 

 allude, it invariably happened that the blocks used in the 

 buildings, which, from their antiquity, or other causes, had sub- 



