154 ANNUAL OF SCIENTIFIC DISCOVERY. 



mass just before solidification is completed. In iron foundries, the 

 moment -when the crystals are forming is often indicated by a rising 

 of the metal in the gates of a mould. 



It is known that the crystalline structure of metals is deranged by 

 rolling, stamping, forging, wire-drawing, and other processes, but it is 

 not commonly known that this effect is temporary, that they have 

 power within themselves to recover their pristine formation. This we 

 have noticed in drawn wire and pipes of block tin. When of pure 

 metal, they are soft almost as lead, and yield to flexure as silently ; 

 but if laid aside a few years, they give out when bent the crackling 

 noise by which bars of the metal are characterized. 



A very interesting fact is mentioned by Scoresby. He found in 

 the Arctic regions that water congeals there in an almost endless va- 

 riety of geometrical figures, of which he enumerates five classes the 

 lamellar, the stelliform (which is most general, and occurs chiefly when 

 the temperature is near 32), the regular hexagon (which becomes 

 thin and diminishes in size as the cold increases), aggregates of hexa- 

 gons (which occur chiefly at low temperatures), and, lastly, combina- 

 tions of hexagons with spines or radii. Arctic Regions, vol. i. p. 

 432. Phil. Journal Franklin Institute. 



GKOUND ICE. 



Ground ice is the ice found under the surface of the water in rivers. 

 It has engaged the attention of men of science on account of its appa- 

 rently unnatural position, and also the attention of practical men be- 

 cause of the mischief it may occasion by accidental obstructions, such 

 as a branch of a tree in a mill-course, when the water is charged with 

 icy particles. Mr. Richard A die has published a paper in the Journal 

 of the Chemical Society on this subject. He believes that he was the 

 first to state that ground ice is formed in the coldest part of the 

 stream, and that the small crystals, as soon as formed, are carried along 

 by the current and submerged and entangled by plants, etc. In De- 

 cember and January, 18GO-1, he searched for ground ice where he 

 had previously found it ; but, although the frost was severer than it 

 had been for sixty years past, he found it only in one locality, viz., on 

 a stone covered over by the water of a rivulet at Duddington, near 

 Edinburgh. Other observations have led him to the opinion that the 

 position of ground ice is one of lodgment merely, in opposition to the 

 notion that the water has frozen in the bed of the river, the current 

 preventing its freezing in its natural place the surface. In a note 

 on Mr. Adie's paper, the eminent chemist, Dr. E. Falkland, gives his 

 opinion that the formation of ground ice, which takes place only in 

 rapidly-flowing streams, depends upon the fact that ice, like other 

 crystalline bodies, deposits itself more readily on rough surfaces 

 (freezes, in fact, at a higher temperature) when in contact with such 

 surfaces than within the mass of liquid itself. Hence when a rippling 

 stream is cooled to 32 ice crystals attach themselves to the pebbles 

 at the bed of the river and form nuclei for further deposition. 



