■64 Mr. H. H. Poole on Thermal Conductivity and Spetific 



The interior of the tube-furnace beyond the ends of the 

 cylinder was filled with asbestos millboard soaked and made 

 into a pulp with water, and the ends of the furnace were 

 •covered with a layer about 5 cm. thick of the same material. 

 When this pulp dries it sets nearly as firm as the original 

 millboard ; unfortunately it contracts considerably in doing 

 -so, and was very liable to break the couple wires. 



It was shown in a previous paper that 



A L-7T ° /'! 4 dt J 



very approximately, where 



K = Thermal conductivity. 



A = Mean temperature difference indicated by the couple. 



Q— Heat flux in calories per second per cm. of the axis. 



r ± — Radius of the circle containing the inner junctions. 



r 2 = „ „ outer „ 



C = Thermal capacity per c.c. of the rock. 



0=i Average temperature of the rock. 



The last term is only a small correction for any slight 

 variations of temperature. The diameters of the grooves 

 were 0*736 cm. and 2* 60 cm. respectively. The distance 

 of the single junction groove from the centre was 0*79 cm. 

 This is so near the geometric mean of r { and r 2 that the 

 temperature indicated by a couple in this groove may be 

 taken as the mean temperature without appreciable error. 

 For basalt C = 0*64 about. The distance between the voltage 

 leads on the central wire being 10*14 cm., 



Q W 



10-14x4*18 



where W is the product Current x P. D. Putting in these 

 figures we find 



w 



K = 4*73xl0- 3 x-^- 



where W'^W- 0*875*, 



«. being the rate of rise of temperature in degrees per minute. 

 It will be noticed that the correction for variation of tem- 

 perature is very small, a generally being of the order of 

 0°*01 0. per minute ; in fact, in the majority of cases it was 

 negligible. 



In arranging the rock cylinder in the furnace the single 



