Heat of Granite and Basalt at High Temperatures. 71 



These values agree fairly well with the previous deter- 

 minations on another granite cylinder, the chief point of 

 difference being the straightness of the K : 6 line. The 

 initial high results are somewhat higher than any previously 

 obtained, but this is easily explained, as in the former work 

 the rocks were initially heated to a fairly high and unknown 

 temperature to bake the cement then used, so that even the 

 first determination represented rock whose conductivity had 

 already been permanently lowered by heat. In the present 

 instance the cement used set without heat. 



Determination of Specific Heat. 

 The equation 



AL27T *' ri 4 dt y 



i • i • i d& • n r m dd , 



which is true so long as -j- is small compared with -j~, has 



at at 



been used above, an approximate value of C being employed 



in the small correction for any slow T temperature variation. 



If, however, we make Q = 0, i.e. turn off the central current 



and arrange that —.- is large, the equation becomes 



4A ■ " ' dt' 



This equation, however, neglects the effect of the central 

 hole. When this is taken into account the equation 

 becomes 



where H is the radius of the hole and H is the thermal 

 capacity per cm. of the lagging and central wire. In the 

 small correction for H we can substitute an approximate 

 value of C without serious error. Substituting for R its 

 value 2-15 mm. and making an estimation for H (which is 

 of very small importance) we find that 



K= -0-3698 x C 



A dt 



or 



dt 



the minus sign occurring since A is always negative, i. e. the 



