Lake-glaciers embosomed in Mountains. 



439 



mean temperature at the surface, where the old ice is in contact with the new 

 ice forming, may be 32 degrees. In the ordinary regelation experiment the wet 

 surfaces of ice are put close together, and a portion of the water is evaporated 

 into the pores of the ice, and heat is abstracted, so that the remaining quantity 

 of water is frozen uniting the pieces of ice together. Regelation does not occur 

 when there is a thick film or sheet of water between the two surfaces of ice, 

 unless the ice is much below 32 degrees, or there is pressure producing currents 

 and evaporation. Ice, although not exactly a colloid; may have cells which 

 admit of exosmosis and endosmosis in the gaps between the internal crys- 

 talline surfaces.* There may be one-tenth of the whole bulk of the ice 

 occupied by such cavities . This could be proved by freezing an hydraulic press 

 of excessive strength containing a small film of water and an observing tube 

 When a mass of water was frozen in a mortar, a shot of 3 lb. weight was ejected 

 415 feet.f 2 No pressure gauge was introduced in this Canadian experiment. 



Mr. Ruskin, on the nth of this month, explained his views of glacier action, 

 and interested you very much. The subject to-night is almost a parallel one, 

 but I cannot treat it, unfortunately, with the power that Mr. Ruskin applied 



'M: 



to it. Mr. Ruskin's views on the subject of the viscosity of glaciers agree with 

 those of Prof. James Forbes, and are different to those I shall state to-night. 



(4) Glaciers are but frozen rivers, and they obey the same great laws of motion 

 that rivers follow, although the advance forward of a large river is as much in 

 a second as that of a large glacier in a week. No one has constructed a large 

 glacier artificially to experiment upon ; while Darcy and Bazin experimented 

 on the flow of water in many kinds of channel. See their " Recherches Hydrau- 

 liques," Paris, 1866. 



* On page 40, Note 1, the action of ice almost resembling a colloid body during the act of 

 regelation is described. t Ganot's Physics, page 261. 



