170 Messrs. R. M. Deeley and P. H. Parr on 



Bliimcke and Hess are of opinion that their results are 

 extremely accurate, and that the low temperatures they 

 found cannot be ascribed to experimental errors. It is clear, 

 therefore, that either the theoretical and experimental resnlts- 

 of Thomson and Kelvin are too small, or that the tempera- 

 ture of the glacier is slightly below the freezing-point to a 

 depth of at least 148 metres. 



This could only come about by the temperature of the 

 snow falling on the neve being much below the freezing- 

 point : so much below the freezing-point that the heat 

 liberated by friction has not been able to melt any of it 

 down to a depth of 148 metres. 



Inf?rna< H>-:r of friction. — There is also a further reason 

 to suppose that water may be formed in the interior of 

 glaciers. During the descent of the ice from higher to 

 Jower regions viscous shear takes place and the internal 

 friction generates heat. If this heat could not escape, some 

 of the ice would thaw. 



The heat liberated by friction in the ice is greatest near 

 the bottom, where the rate of shear is greatest, and decreases 

 rapidly towards the surface. 



The heat liberated in a column of ice 1 metre square and 

 200 metres high (which is about the thickness of the 

 Hintereis Glacier) in one year = weight x fall -i- mechanical 

 equivalent of heat. 



The weight of 1 m.- of ice 200 metres high 



= (100 - x 20000 cm. 3 x -9 x 981, 

 =1-765 x 10" dynes. 



The average velocity of the ice for a surface speed of 

 21 metres per annum is 11 metres per annum. With a 9 per 

 cent, grade this gives a vertical fall of 126 cm. per annum. 



Work, therefore. 



= 1-765 x 10 11 x 126. 



= 2-2-2 x 10 13 erg^. 



= 5o4.900 gram calories. 



Floic of Heat in Glaciers. — The only means by which 

 water in a glacier can be frozen, other than by a low 

 external temperature, is by the presence of a temperature 

 gradient in the ice. Such a gradient would exist in a wet 

 glacier, and it remains to be seen if this gradient is sufficient 

 to get rid of the water produced by friction. 



It was shown by J. Thomson from theoretical considera- 

 tions that the freezing-point should be lowered 0°-000231 C. 

 per kg. of pressure, and Kelvin obtained experimentallv a 



