464 PROCEEDINGS OF THE AMERICAN ACADEMY. 



May 14, 1898. 



Summary of Results. 



There are certain errors which enter into these results which will be 

 eliminated or corrected in subsequent experiments; namely, conduction 

 through holes cut for lead wires, etc., variation of current and voltage, 

 heat generated by rotation of fans on interior of sphere, uncertainty 

 of galvanometer readings due to lack of sensitiveness of the instrument, 

 lack of homogeneity in the shell. 



Cast iron was used because it could be obtained and prepared with 

 facility for the experiments which were designed as a test for the capa- 

 bilities of the method, which is to be applied to the determination of the 

 conductivity of pure metals. 



Rogers Laboratory of Physics. 

 May, 1905. 



