33 



of the column ; and in which the joint presented to his view 

 the appearance as if the cross fracture had originated at and spread 

 out from this spot of irregular quality. When this extraneous or 

 irregular lump happened to be near the middle of a column, there 

 appeared to emanate from it, in all directions, approximately 

 straight but roughly-formed rays ; and when the lump happened to 

 be near one side of the column, the rays emanating from it spread 

 out from it in curved forms like a brush, and the several rays in pro- 

 ceeding outwards seemed to bend gently somewhat towards the 

 nearest external face of the column. This seemed as if they had 

 tended to run so, as at each moment, to be advancing in a direction 

 approximately perpendicular to the advancing circular or oval edge 

 of the enlarging fissure. If a fracture originating at one side of a 

 column, were to advance across to the other side, and in so doing 

 were to cut across any irregular lump in the mass, that lump would 

 leave a kind of tail extending from itself forward in the direction of 

 propagation of the fissure ; but the part of the fissure formed before 

 arriving at the lump would be scarcely at all influenced by the 

 presence of that irregularity. A tail emanating in this way from 

 an irregular lump or a vesicular cavity, and extending forward in 

 the direction of advance of the crack, is continually to be noticed 

 in the breakage of flints, glass, basalts, and other brittle substances. 



But the cases noticed at the Giant's Causeway, in which, from 

 an included lump, the lines radiated out in various directions, 

 and were curved when the lump was eccentric, tend to corroborate 

 the supposition that the fissure had its beginning at the irregular 

 lump, where some local weakness or overstraining might exist, 

 and that it flashed out from thence towards the circumference of 

 the column. 



In the discussion which ensued on the reading of the paper, an 

 interesting case was adduced by Mr. Gray in support of Professor 

 Thomson's facts and arguments for proving that the spheroidal 

 structure in successive coats, resembling those of an onion, so 

 frequently manifested in decaying basalts, is not an original con- 

 cretionary structure, but is due to decomposition penetrating from 

 without inwards, in blocks or fragments, into which the rock has 



E 



