A. Kendrick — Damping of Bell-magnets, etc. 459 



fourthly with the bottom bored through. The A's as abscissae 

 and thicknesses of bottom as ordinates give the first curve on 

 the right in fig. 1, and the second curve is for the values of A 

 when the needle was at the same heights and the bottom bored 

 through. The advantage of having at least a slight thickness 

 of copper in the bottom is evident from the curves. The 

 effect of a thick plate of copper alone such as box 3 inverted 

 under the needle is slight, A being about 0*75. 



An idea of the relative effect of various outside and inside 

 diameters of the damping boxes is given by fig. 5, where each 

 pair of adjacent segments represents sections of those boxes 

 that give nearly the same damping effects. Thus, A and B 

 give A=0-04, C and D give A=0-12, E and F give A=0\L8. 

 J) is box 1 (2d), C is 4 (c) ; B is 3 and 4 (a), and A is a box of 

 outside diameter equal to D but smaller inside diameter. E is 

 box 2, described in the following paragraph, and F is box 

 A turned down to smaller outside diameter. The thickness of 

 the bottom of all these is about the same. The space for the 

 magnet in A and F is too small for convenient leveling, in C 

 it is larger than necessary. 



Box 2 is of like dimensions with 3 but planed down on two 

 faces. This gave A=0 - 18, and the direction of the longer 

 transverse axis was immaterial. It should not be inferred, of 

 course, from this that the same effect would be obtained from 

 a circular cylinder of the radial dimensions of the shorter axis. 

 In fact very nearly the same damping was given by a circular 

 box of 2 mm thickness and inside diameter of 2 mm less than that 

 of box 2. This is shown by E and F of fig. 5. 



Fig. 6 gives horizontal section and vertical elevation of a 

 box like 3, but slit nearly through vertically, aa and a' a' are 

 positions of the slit respectively perpendicular and parallel to 

 the needle's magnetic axis when at rest. In the position aa A 

 was 0"60, and in the position a' a' A was 0*04, the same as 

 when there was no slit. This box was afterward slit com- 

 pletely through. With both halves in position a' a' the needle 

 came to rest in 4 swings, with one-half in 7 swings. In posi- 

 tion aa, both halves, 19 swings ; one-half, 37 swings. 



With a brass box of same outside diameter as 3, bored 

 through, inside diameter 2 mm less than that of 3, A was about 

 O80. Comparing this with 1 (4th), brass is seen to be de- 

 cidedly inferior to copper as a damping material. 



With a stronger field the A's are of course larger, with a 

 weakened field smaller, and for Q in all of the above men- 

 tioned boxes, except the brass one, the swing may be made 

 " dead-beat." The time of swing is, however, lengthened with 

 increased damping, and this is the more noticeable the smaller 

 the moment of inertia. This seems to show that the effect of 



Am. Jour. Sci.— Third Series, Vol. XLVII, No. 282.— June, 1894. 

 31 



