aad other fphe- 

 roids formed 

 above them 

 would produce 

 articulated co- 

 lumns. 



1*4, v OBSERVATIONS ON BASALT, &C 



into prifms with well defined angles. In a ftratum compofed 

 of an indefinite number in fuperficial extent, but only one in 

 height, of impenetrable fpheroid^ with nearly equidiltant 

 centres, if their peripheries (hould come in contact on the 

 fame plane, it feems obvious that their mutual action would 

 form them into hexagons ; and, if thefe were refilled below, 

 and there was no oppofing caufe above them, it feems equally 

 clear, that they would extend their dimenfions upwards, and 

 thus form hexagonal prifms, whofe length might be indefi- 

 nitely greater than their diameters. The farther the extremi- 

 ties of the radii were removed from the centre, the nearer 

 would be their approach to parallelifm ; and the firuclure 

 would be finally propagated by nearly parallel fibres, (till 

 keeping within the limits of the hexagonal prifui with which 

 their incipient formation commenced ; and the prifms might 

 lhus (hoot to an indefinite length into the undifiurbed central 

 mafs of the fluid, till their ftructure was deranged by the 

 iuperior influence of a counteracting caufe, which would be 

 provided by the action of the atmofphere on the upper 

 furface of the bai'alt. If this arrangement exifted, the fame 

 caufe that determined the concentric fractures of the fibres 

 1 of the fperoids, would produce convex articulations in the 

 lower joints of the prifms; and, in proportion as the centre 

 from which they were generated became more remote, the 

 articulations would approximate to planes. If the gene- 

 rating centres were not equidiftant, the forms of the pillars 

 would be irregular ; and the irregularity would be in propor- 

 tion to the diverfity of diftance between the centres. If the 

 difference was great, the number of fides would be altered, 

 and they might be found pentagonal, tetrahedral, and trihedral. 

 As the comprethon of the fibres would be greatelt m the level 

 of the generating centres, the lower part of the mifms would 

 be molt compact. 

 The fa£ls at- ALl thefe conditions feem to be fulfilled, in the actual con- 



tending the pro- formation of bafaltic columns; for, in every inftance I am 



ducYion of bafal- , . . . „, . . r , • , 



tic columns agree acquainted with, tney appear to have been formed in the 



with the fore- tranquil bofom of the mafs, as they have been originally maft- 



* oinfi * ed by amorphous trap, and their prifmatic ftructure is only 



difplayed by the removal of this covering. This has been 



variouily effected, fomelimes by the apparent difrupture of 



rocks, fometimes by the exterior portions of the mafs being 



thrown 



