528 



Mr. 0. J. Lodge on a Mechanical Illustration 



from Granotj art. 364, for the three metals, silver, copper, and 

 platinum (gold is left out because its results are discordant); 

 the second column contains the relative atomic weights ; the 

 third column an approximation to the relative electric conduc- 

 tivity of each metal, obtained by taking a mean between the 

 value given by Matthiessen and that given by Becquerel 

 (Miller, vol. i., being my authority); and the last column con- 

 tains the numbers which ought to be constant. 





9:6. 



on. 



1 

 r 



m9 



ToTe' 





27 



4-9 

 10-1 



108 

 63-5 

 197 



100 

 96 

 13 



2916 



2987 

 2586 



5985 





Platinnm 



Gold 



4-3 



196 



71 





Taking 2850 as a rough mean value of this constant, one can 

 calculate backwards to the radiating-power for any metal on 

 the same scale. The following Table contains the radiating- 

 power given by the hypothesis for a dozen common metals, 

 most of whose resistances at 0° Centigrade were ascertained 

 by both Matthiessen and Becquerel. The exclamatory mark 

 against gold signifies that its atomic weight has been called 

 98 because the correct value 196 gives 2*0, which is apparently 

 too small a result. 



Metal. 



Mean conduc-l R adia ^g- 

 Centigrade. 



Silver ! 100 



Gold! 71 



Copper 96 



Cadmium 241 



Platinum 13 



Zinc 265 



Lead 8-3 



Tin J 132 



Iron j 14-6 



Antimony j 4*6 



Mercury j 17 



Bismuth 1*25 



Concerning this Table, I have to remark that the hypothesis 

 cannot be expected to give correct values for the last three 

 metals ; for the electrical resistance was ascertained by ex- 

 periment on large masses of material, whereas it is the mole- 

 cular resistance which is concerned in radiation ; and though 



