136 ICE AXD GLACIERS. 



of this phenomenon has been much controverted ; I 

 have detailed to you that which I consider most satis- 

 factory. 



This freezing together of two pieces of ice is very 

 readily effected by pieces of any shape, which must not, 

 however, be at a lower temperature than 0°, and the 

 experiment succeeds best when the pieces are already in 

 the act of melting.^ They need only be strongly pressed 

 together for a few minutes to make them adhere. The 

 more plane are the surfaces in contact, the more com- 

 plete is their union. But a very slight pressure is suffi- 

 cient if the two pieces are left in contact for some time.^ 



This property of melting ice is also utilised by boys in 

 making snow-balls and snow-men. It is well known that 

 this only succeeds either when the snow is already melt- 

 ing, or at any rate is only so much lower than O'' that 

 the warmth of the hand is sufficient to raise it to this 

 temperature. Very cold snow is a dry loose powder 

 which does not stick together. 



The process which children carry out on a small scale 

 in making snow-balls, takes place in glaciers on the very 

 largest scale. The deeper layers of what was originally 

 fine loose neve are compressed by the huge masses rest- 

 ing on them, often amounting to several hundred feet, 

 and under this pressure they cohere with an ever firmer 

 and closer structure. The freshly-fallen snow originally 

 consisted of delicate microscopically fine ice spicules, 

 united and forming delicate six-rayed, feathery stars of 

 extreme beauty. As often as the upper layers of the 

 snow-fields are exposed to the sun's rays, some of the 

 snow melts ; water permeates the mass, and on reaching 



' In the Lectnrf^ a series of small cylinders of ice, which had been pre- 

 pared by a method to be afterwards described, were pressed with their plane 

 ends against each other, and thus a cylindrical bar of ice produced. 



* Vide the additions at the end of this Lecture. 



