Heat of Granite and Basalt at High Temperatures. 67 



being 7*5 cm. and diameters 3'6 cm., so that the complete 

 cylinder was 15*0 cm. x 3*6 cm., which was considered a 

 sufficient ratio of length to diameter. The arrangement of 

 the couples was identical with that used for the basalt. 

 The arrangement of the P.D. leads connected to the central 

 wire is shown in fig. 3, which shows a section of the rock- 

 cylinder and furnace-tube. A is the platinum wire, B an 

 asbestos lagging which fills the hole and keeps the wire 



Fig. 3. 





approximately central in it, and insulates A from the leads 

 everywhere, except at the junctions. This lagging was 

 removed from an asbestos insulated heating wire and threaded 

 onto the platinum wire. A more convenient method adopted 

 when the cylinder was being put together a second time was 

 that of unlaying a three-strand asbestos cord and relaying it 

 carefully round the wire. The lagging was made as uniform 

 as possible to prevent errors due to inequality of temperature 

 of the central wire. The insulation was tested by passing a 

 very small measured current through the central wire and 

 measuring the P.D. between the leads C, hence the resistance 

 of the part of the central wire between the leads was found ; 

 and as the current was too small to cause appreciable heating, 

 this ought to agree with the known resistance of that length 

 of wire at room temperature. This it was found to do almost 

 exactly. Any fault in the insulation would thus have been 

 detected. By this arrangement the P.D. junctions were 

 inside the cylinder as required, while the extra end cylinders 

 were dispensed with. 



D and E are asbestos cords cemented to the cylinder as 

 shown. The fine couple leads were twisted round these, and 

 where they passed through »the asbestos ends of the furnace 

 another asbestos cord was wound round them. This arrange- 

 ment entirely prevented breakage in the furnace ends and 

 enabled the latter to be opened when necessary without 

 breaking the wires. The difficulty of doing this previously to 

 the adoption of this plan had greatly hindered the location of 

 breakages in the couples. The couple grooves are also shown 

 in section in the figure. 



F 2 



