Heat of Granite and Basalt at High Temperatures. 75- 



Anotlier set of determinations of the specific heat was also 

 made, the procedure being the same as before. In working 

 these out K was taken as (4*77 -000198 0) x 10~ 3 . Even 

 so the mean value obtained 



8 = 0-154 + 0-335 x 10" 3 0-0-25 x 10" 6 2 



is somewhat lower than the previous one. This difference 

 was chiefly due to Ioav values obtained with heating experi- 

 ments which on this occasion gave lower results than the 

 cooling experiments. The latter gave values only slightly 

 below the previous cooling determinations. 



We may take as the mean of the determinations on the- 

 conductivity of this sample of granite 



K = (4-66-0-00195 6) x 1Q- 3 . 



This refers to granite that has been heated to 500° C. 



As regards specific heat, the mean of the two formula? 

 already given is 



S = O-158 + O-37xlO- 3 0-O-35xlO- 6 6> 2 . 



This gives a value 0*178 at 56° C. The mean of the specific 

 heats of three granites obtained by Dr. Joly with the steam 

 calorimeter between 12° 0. and 100° C. is 0*192, so that the 

 above figures are probably about 8 per cent, too low; the 

 method is probably more suitable for finding the variation of 

 the specific heat with temperature than its absolute value. 

 The rise of the specific heat of granite with rise of tempe- 

 rature accords well with the known rise for quartz and 

 felspar. 



Second Basalt Experiment. 



The arrangement of the rock cylinder adopted in making 

 another set of determinations on basalt was somewhat 

 different. The cylinder, 18*2 cm. long by 3*6 cm. diameter, 

 was divided longitudinally instead of* transversely. Grooves 

 were cut along the axis of each half cylinder as shown in 



Fijy. 6. 



~L 



JU 



c ' _r 



fig. 6, thus forming a fine accurately centered hole when the 

 two were cemented together, through which the central wire 



