456 A. Kendrick — Damping of Bell-magnets ; etc. 



section, and fig. 6, horizontal section and side elevation (with- 

 out perspective) of cylindrical boxes for the bell-magnets. 

 Similarly the bell-raagnets are shown, p. 461, by horizontal sec- 

 tions and side elevations. The suspension was a raw silk fiber 

 whose torsional effect can practically be ignored, though it 

 was quite appreciable in the cases of the smallest magnets. 

 The restoring force, unless otherwise stated, is understood to 

 be the earth's field, which has at the point where the needles 

 were suspended a horizontal intensity of about .16 dyne per 

 unit pole. 



As a measure of the. damping that occurs in a succession of 

 vibrations, I will use the average of the ratios of each ampli- 

 tude to the preceding amplitude taken from a number of 

 successive swings and call this A. Its reciprocal is approxi- 

 mately the logarithmic decrement, but this ratio is more con- 

 venient, especially for plotting, as the values must lie between 

 and 1, instead of between oo and 1. The lower limit signi- 

 fies " dead-beat " vibration, and the upper limit, no damping. 

 A value of A less than 0*1 means for most work very good 

 damping. Since some of the following comparisons are made 

 by referring to the number of swings (half vibrations) that the 

 needle, after disturbance, makes before coming to rest, it may 

 be useful to write a table giving approximately the values of 

 A corresponding to these. The scale used in reading deflec- 

 tions was divided into millimeters (scale distance about 65 cm ), 

 and the readings made to two-tenths of a millimeter. Then 

 with a moderate initial disturbance of the needle, say about 

 200 mm , if the vibrations decease to less than two-tenths of a 

 millimeter in 



5 swings, A is about 0*20 



6 " " " 0-25 



7 " " " 0-35 



8 " " " 0-40 



10 " " " 0-45 



11 " " " 0-50 

 15 " " " 0-60 

 20 " " " 0*70 

 33 " " " 0-80 



Let any copper box be designated by the number of the 

 figure that represents it, and any magnet by its letter. 



Magnet Q was used under a larger variety of conditions than 

 any of the others. These varying conditions were : five dif- 

 ferent moments of inertia of relative values 1*0, 1*15, 2*4, 2*9, 

 3*7, the first being that of the magnet, mirror and suspension, 

 and the others were produced by the addition of inertia discs ; 

 three magnetic moments of relative values 0*44, 0'68, 1'0, the 



