310 POPULAR LECTURES AA7) ADDRESSES. 



" tion has reached, and therefore none is to be 

 " expected at greater depths, at all events so far 

 " as 50 or 100 miles down." 



This question was very carefully discussed 1 in 

 A of the British Association at its Glasgow meet- 

 ing in 1876; and I showed that instead of there 

 being reason to expect in an equable rate of increase 

 of temperature downwards in the greater depths, we 

 have on the contrary very strong reason to believe 

 that at a depth of 30 kilometres the rate of increase 

 downwards is considerably less than it is at the sur- 

 face, and that at the greatest depths the tempera- 

 ture is not high enough for fusion of the material. 

 The state of the case then is, that we have no 

 reason from anything that has been observed, or 

 that could have been observed, at depths to which 

 we have already penetrated, for concluding that the 

 rate of increase of underground temperature, ob- 

 served at small depths beneath the surface, continues 

 uniformly as we proceed inwards ; and that any 

 argument for intensely high temperature within. 

 based on the assumption that the rate of increase 



1 See pp. 240. 244 of present volume. 



