8G 



to expectations, however, the sensitiveness with tlie nicliel core ap- 

 peared to be the greater in strong fields and with the iron core in 

 wealv fields. Both showed a maximum of sensitiveness at a short dis- 

 tance from the magnet, the maximum for nickel being the farther re- 

 moved. The nickel core proved to be more sensitive than the iron 

 core for distances up to 2.5 cm. 



When the detector was worked with the mixed core of iron and 

 nickel wires the deflections of the galvanometer increased as the magnet 

 approached the core, even up to the point of contact. The curve (Fe 

 & Ni, Fig. 2) lies above the Fe curve at all points and above the Ni 

 curve at most points, showing that a mixed core consisting of annealed 

 piano wire and hard-drawn nicliel wire produced a more sensitive de- 

 tector than was obtained l)y using a core of piano wire only. 



The detector gave small deflections of the galvanometer when I 

 used an antimony core; also when I used a core of iron filings contained 

 in a thin-walled glass tube. In both cases deflections were obtained 

 only when the magnet was near the core. A core of bismuth gave 

 no deflection. 



It is probable that the form of the curve of Figs. 1 and 2 depends 

 upon other points than those considered in this papei", as for instance, 

 the frequency and intensity of the oscillations sent out by the trans- 

 mitter and the annealing of the steel wires used in the core. 



Since electric oscillations appear to "have the power of reducing 

 the effects of magnetic hysteresis," it has occurred to the writer to test 

 their effect upon the hysteresis loss of transformers, armatures, etc. 

 Some experimental worlc on tliis subject has been done, but I am 

 not yet ready to announce results. 



Physics Laboratory of Indiana University, April, 1903. 



