432 Dr, Joule on an improved Galvanometer. 



magnetic action, negative with a very different strength, and 

 nothing for an intermediate one. Such might be the case 

 without a change of sign occurring within the Umits of ordinary 

 magnetic forces ; but it is highly probable that under these cir- 

 cumstances, and within the limits of experiment, the salts of 

 nickel would at least be observed to depart sensibly from the 

 general law of variation of magneto -rotary powers. In order to 

 test this, I compared the rotation produced by a solution of sul- 

 phate of nickel with that produced by water under very various 

 magnetic actions; and as I found the proportion between the 

 two rotations absolutely constant, I was obliged to abandon my 

 hypothesis. 



LIII. 0)1 an improved Galvanometer, 

 By J. P. Joule, LL.D., F.R.S. ^c* 



THE important experiments required in carrying out the 

 gigantic projects of electric telegraph engineers having 

 rendered a delicate and portable galvanometer an essential piece 

 of apparatus, I am induced to hope that the following descrip- 

 tion of one recently made from my design will interest at the 

 present time. In the figure, A A represents the frame on which 

 the wire forming the coil is wound. It is inserted in a semi- 

 circular groove cut in the block of wood B B. Another similar 

 block, not shown in the figure, is fastened to the first by hasps, 

 so as to hold the coil firmly in its place, c c shows the section 

 of the graduated circle, enclosed in a box whose glass lid d is 



fitted with a glass chimney e, surmounted by a cap and roller /. 

 Over this roller is thrown the filament of silk which supports the 

 needle (a piece of magnetized sewing-needle a quarter of an inch 



* Communicated by the Author. 



