438 



Walter Keeping — On Columnar Sandstone. 



" metal " on the road of approach. On entering it, evidences of 

 former volcanic activity are seen in some rubbly basalt-ash on one 

 side. A closer search is necessary, in the present condition of the 

 quarry, to find the mass of the basalt in the floor of the quarry. 

 This proves to be a very hard, close-grained olivine basalt, excellent 

 for road metal, and, as recorded by Gutbier, columnar in structure. 



I 



I 



Fig. 1 . Columnar Sandstone as seen in a quarry at Gorischstein, 

 West of Schandau, " Saxon Switzerland." (A walking 

 stick is placed against the bank as a scale of height.) 



V 



We may pause for a moment to consider what are 

 the petrological aspects of this mass of basalt : whether 

 it was a dyke, sheet, or " neck." There can be little 

 v% J doubt that it was an old neck ; for, judging from the 

 contour of the quarry, the horizontal section was 

 circular, and the basalt is a pipe-like mass. At the 

 j sides of the quarry ashy materials are still to be seen. 

 A small cinder-cone probably once capped the ground 

 above. 



On one side of the pit some loose sandstone prisms, 

 small and regular like those in the Dresden Museum, 

 Fig. 2. are scattered over the slope, but these are only part 



A Single Column of of the debris fallen from the pit sides higher up. In 

 the Natural Size, ^he undisturbed quarry-face above they are seen in 

 astonishing perfection, as indicated in the annexed 

 woodcut (Fig. 1), which represents a drawing of part of this exposure. 

 Here, facing the spectator, is a series of tiers of colonnades made up 

 of regular prisms. Those nearest the front lie nearly horizontal, 

 whilst as we recede into the sandstone they approach more and more 

 towards the vertical, till, at last, about two or three feet from the 

 outer series, the columns stand perpendicular ; so forming altogether 

 the regular curve seen in the woodcut. Now this curving is not 

 produced to any important degree, by a binding of the prisms them- 

 selves, but is due to the gradual and regular variations in the 



