13.2 Mr. R. Mallet on the Origin and Mechanism of 



up) of every prism, so that each prism is in process of cooling 

 from its sides as well as from the top, where it will be coldest, 



VISCOUS BASALT 



and hottest at the bottom where the mass is yet united but 

 almost on the point of fracture. Thus the progress of cooling 

 within this length of any one prism at a given instant may .be 

 indicated, as in fig. 4, by lines which are the vertical sections of 

 isothermal couches or planes taken at successive intervals between 

 the surface and the axis of the prism — which is of course hottest 

 along the axis, and the exterior and succeeding couches of 

 lowest temperature thickest at the upper part of the prism and 

 thinnest at the lower part, where the time of cooling has been the 

 shortest. We have now to consider the mode of production of 

 the cross joints or articulated cup-like fractures observed in 

 these prisms, and to show that they too are fully accounted for 

 by the play of contractile forces. Before proceeding to this, it 

 should be remarked that the cooling, effects due to the pro- 

 gress of splitting asunder of the prisms, will be far more marked 

 than might be at first supposed. Geological facts tend strongly 

 to the conclusion that most basalts overflowed from fissures 

 under water. Whether the liquid basalt flowed directly into 

 contact with water or spread itself out laterally between a more 

 solid bottom and a mass of loose matter resting upon it, or was 

 poured out sub dio and exposed to rain &c, water would cer- 

 tainly find its way into all the joint-fissures opened between the 

 prisms, and with an energy proportioned to the depth or head 

 of water above it. Nor would the repulsion known to exist be- 



