of some Mocks at High Temperatures. 57 



or when written in the most convenient form for calculation 



log [K-0-003835] =3-4130-0-00397 0. 



The next 9 results, beginning with that for a temperature 

 of 361°, lie roughly along a right line 



K= [4-48-0*00275 ff\ x 10~ 3 



between the limits of temperature 160° and 390°. 



The last 19 results, beginning with that for a temperature 

 of 370° C, which, however, seems to be of a somewhat 

 transitional nature, lie approximately along the lower curve 



K = (3-14 + 1-99 £-0-00914 e) x 10- 3 , 



or log(K-0-00314) = 3'-2989-0-00397<9. 



It will be noticed that the coefficient of 6 is the same in the 

 first and last curves. 



The progressive lowering of the conductivity is due to 

 baking of the stone, which was found to be greatly altered 

 when it was finally removed from the furnace. The first 

 member of the second and also that of the third series are 

 high temperature determinations, which probably caused 

 permanent alteration in the rock, though in the case of the 

 third series a higher temperature than 370° had previously 

 been attained. The fact that so many results lie approxi- 

 mately on the curve seems to indicate that the condition of 

 the rock had become approximately constant. The first and 

 third series are the most interesting, showing respectively 

 the fall in conductivity when the rock is first heated and the 

 variation of conductivity with temperature when the rock 

 has been several times raised to over 350° 0. The transi- 

 tional results in the second series seem to be less interesting. 



Slate. 



An attempted determination on slate failed as the rock 

 split when heated. 



Granite. 



The arrangement adopted with granite is identical with 

 that used with limestone. The cylinder of medium-grained 

 Mourne granite is 13*90 cm. long and 7*86 cm. in diameter. 

 The diameters of the inner and outer grooves are 1*833 cm. 

 and 6*156 cm. respectively. The specific heat and density 



