328 EXPERIMENTS umn MAGNETISM. 



that the adion of the magnet upon the filings, at equal 

 diftances, was fenfibly lefs, than when the tube was full 

 of atmofpheric air. The want of a tube of fufficient dia- 

 meter prevented me from making the experiment in fo 

 fatisfadory a manner as I wifhed. It appears, however, 

 worthy of being repeated by thofe who may poflefs the 

 neceffary means. If the magnetic power fliould be ob- 

 ferved in fuch a vacuum, then the above conje£lure will 

 merit and receive the fate which has generally attended all 

 reafoning in phyfics not founded on accurate experiment. 



I am, very refpedlfully, 



Yours, &c. 



J. MADISON, 



William Ss* Mary College 



ThermometricaJ 



