84 



upon both the distance and direction of motion of tlie moving magnet. 

 When the magnet is near the core the detector is more sensitive when 

 tlie magnet is approaeliing, Init when some distance from the core the 

 detector is more sensitive when the magnet is receding. Both cnrves 

 indicate a maximum of sensitiveness at a distance from the core, the 

 distance being less when the magnet is approaching than when receding. 



Removing the magnet and operating the transmitter tended to de- 

 magnetize the core. Then when the magnet was placed in position and 

 the transmitter again opei'ated, as in Curve C, there was a relatively 

 greater change in the magnetism of the core than was obtained under 

 the conditions of Curves A and B. Hence the deflections in column 

 C are greater than those in A or B. It is evident that the relative 

 change in the magnetization of the core would be greater still where 

 the magnetic held is reversed after each reading, as in Curve D. 



Since nickel is more susceptil)le than iron in weak magnetic flelds, 

 and less susceptible in strong fields, it occurred to the writer that 

 a more uniform sensibility for varying distances between the moving 

 magnet and core might be obtained by making the core of nickel. 



Four cores were made, each one being 5 cm. long, approximately .4 

 cm. in diameter, and being wound Avith two hundred turns of No. 

 36 copper wire. 



Core 1 consisted of 26 pieces of piano wire, .063 cm. in diameter. 



Core 2 of 10 pieces of piano wire and 10 pieces of nickel wire, .082 

 cm. in diameter. 



Core 3 of 2 pieces of piano wire and 13 pieces of nickel wire. 



Core 4 of 14 pieces of nickel wire. 



Table II gives the deflections at various distances between the 

 magnet and each of the four cores, the magnet being moved one space 

 at a time and having its poles reversed after each reading. The data 

 for three of the cores is plotted in Fig. 2. 



