on Basalt , Sec. 
3°5 
in which the process of arrangement described in the early part 
of this Letter, appears to yield a probable explanation of some 
of the peculiarities of basalt. The general disposition of basalt 
to divide into globular masses, in decomposing, is too remarkable 
a fact to have escaped the attention of naturalists ; though, as 
far as I am informed, no satisfactory explication of it has been 
given. The common effects of decomposition are obviously ina- 
dequate ; for it is common to see a large block of amorphous 
basalt separate into numerous balls, after a few months or years 
exposure to the weather ; and, rapid as the process of decom- 
position has been in the intervening portions, these balls resist 
its farther progress with uncommon obstinacy. May not this 
be attributed to the formation of the radiated spheroids, whose 
occurrence in my experiment I have already mentioned ? and 
may not their greater resistance of weather simply arise from 
their aggregation being more perfect than that of the incoherent 
molecules which have filled the intervals between them? 
Though the radiated structure has disappeared to the eye, these 
portions of the stone retain the superiority of more perfect 
internal arrangement; and, if my pigmy experiments could 
yield spheroids of two inches diameter, there can be no diffi- 
culty in supposing that the grand operations of nature may 
produce them of several feet. The separation of the decomposed 
fragments in concentric coats, seems easily explained; for I 
have already pointed out the facility with which the radii of the 
spheroids separated at nearly the same distances from their 
centres, and the form of the fragments which resulted, resem- 
bling fragments of bombs. * 
* Even granite has been frequently observed to affect globular decomposition, and 
division into fragments of concentric coats. This mode of decomposition extends to 
R r 3 
