254 



A, Durward — Temperature Coefficients of 



Table XL 



Table XII. 



Jessops Flat Tool Steel. 











Dimensions of 



cross section, 



Tubular magnets of Crescent 





l-6 cm x 



0-32 cm . 





Polished Drill Rod. 





Wt. 



per cm. of length, 3-9 grams. 













Stl 



•23 *« 









O-O £« 



CO 



p. 



a o+z 



sg§ 



o a . 



a*" 



Boo 



ree of magnel 

 magnet is heat 

 to 100° C, e 

 )f the moment 

 ure. 



m 



3 



to 



t- 



a> 



o> 



a 



a 

 o> 

 o 



_a 



o> 



0) 



a 



a 



o 

 a 



of the magnet 

 magnet is heat* 

 to 100° C, e 

 f the moment 

 ure. 



a 



a 



o> 

 o 



B 



Oco a 



i*« 



fi <B W 

 ©5 g 



per degi 

 when the : 

 out 15° 0. 

 in terms ( 

 r temperat 



1 

 a 



o 

 o 



a 



a 



03 



6 



3 



■3 



per degree 

 when the 

 out 15° C. 

 in terms o 

 r temperat 



B 



So X 





J3 



be 

 3 



5 



'33 

 a 



9) 



'»2 



~sa£° 



i-3 





2 



rJ 



O 





Sop?* 

 3 



6 



11-0 



0-00089 



6 



0-95 



0-70 



0-00087 



8 



15*0 



0-00071 



7 



0-95 



0-70 



0-00075 



10 



13-5 



0-00058 



7-5 



1-28 



0*94 



0-00063 



15 



27*5 



0*00034 



10- 



1-28 



0-94 



0*00061 



20 



36'5 



0-00028 











Tables Y to XII show how the temperature coefficients of 

 magnets made of different specimens of steel and their losses 

 of magnetic moment between room temperatures and 100° C. 

 varied with their lengths. It is evident that in the case of 

 magnets made of a given kind of steel, the temperature coeffi- 

 cients are functions of both the diameters of the magnets and 

 of the ratios of their lengths to their diameters. Very long 

 slender magnets have in general relatively small temperature 

 coefficients, but it does not seem possible to reduce indefinitely 

 the coefficient of a magnet made of a stout rod of tool steel 

 by increasing the length. Figures 4, 5, 6, 7, and 8, in which 

 the lengths of the magnets are used as abscissas, show graphi- 

 cally the results given in Tables Y, YI, YII, YIII, and X, and 

 recall Cancani's experiments upon magnets made of fine steel 

 wire. 



The Jefferson Physical Laboratory, 



Cambridge, December, 1897. 



