CLOUDS AND RIVERS, ICE AND GLACIERS. 177 



neat, the reflexion of solar light from the sides of the 

 little fissures producing the observed appearance. 



449. In like manner if you freeze water in a test- 

 tube by plunging it into a freezing mixture, the ice 

 produced is white. For the most part also the ice 

 formed in freezing machines is white. Examine such 

 ice, and you will find it filled with small air-bubbles. 

 When the freezing is extremely slow the crystallising 

 force pushes the air effectually aside, and the result- 

 ing ice is transparent ; when the freezing is rapid, the 

 air is entangled before it can escape, and the ice is 

 translucent. But even in the case of quick freezing 

 Mr. Faraday obtained transparent ice by skilfully re- 

 moving the air-bubbles as fast as they appeared with 

 a feather. 



450. In the case of lake ice the freezing is not uni- 

 form, but intermittent. It is sometimes slow, sometimes 

 rapid. When slow the air dissolved in the water is 

 effectually squeezed out and forms a layer of bubbles on 

 the under-surface of the ice. An act of sudden freezing 

 entangles this air, and hence we find lake ice usually 

 composed of layers alternately clear, and filled with 

 bubbles. Such layers render it easy to detect the planes 

 cf freezing in lake ice. 



451. And now for the bearing of these facts. Under 

 the fall of the Geant, at the base of the Talefre cascade, 

 and lower down the Mer de Glace ; in the higher re- 

 gions of the Grindelwald, the Aar, the Aletsch and the 



