494 M. Verdet on the Optical Properties developed in 



placement of the image of the scale observed in the small tele- 

 scope. By suppressing the elliptic frame of red copper, we should 

 increase without doubt the amplitude of the elongations, but we 

 should fall into the inconvenience which makes the use of gal- 

 vanometers with a single needle, such as the sine and tangent 

 boussoles, so restricted. The needle deviated by the impulse 

 of a current, only rests again at the expiration of some minutes ; 

 the least accidental cause would communicate to it a motion 

 which would be equally long in disappearing, so that the suc- 

 cessive observations would necessarily be separated by a con- 

 siderable interval. On the contrary, in M. Weber's apparatus, 

 the influence of the induced currents in the copper frame weakens 

 the oscillations of the needle, and fixes it in its position of equi- 

 librium with a promptitude surprising to those who observe it 

 for the first time. The effect of small accidental oscillations is 

 destroyed almost immediately, and nothing hinders the observa- 

 tions from succeeding each other at very close intervals*. 



It is now easy to perceive how each experiment was made. I 

 commenced always by making two or three observations on the 



* This advantage is rendered much more sensible by means of an inge- 

 nious modification suggested by M. Ruhmkorff. This consists in cutting hol- 

 low the magnetized bar so as to dimmish considerably its moment of inertia 

 without sensibly altering its magnetic moment. It follows evidently, that 

 the damping of the oscillations ought to he much more rapid ; the formula; 

 developed in the preceding note give the mathematical expression of this 

 difference, and experiment shows very evidently that it is correct. M. 

 Ruhmkorff had joined two different magnetized bars to the galvanometer 

 which he had constructed for my experiments ; the one was solid, and the 

 other hollow ; both were magnetized to saturation, and had the same ex- 

 ternal dimensions. "With the solid bar, the amplitudes of the oscillations 

 decreased according to a geometric progression whose ratio was equal to 

 - 646 ; with the hollow bar the geometric progression had for ratio - 477- 

 The following are the two series of observations relative to the two bars : 

 n is the number of the order of each elongation observed, first on one and 

 then on the other side of the position of equilibrium; a the amplitude of 

 the elongation ; r the ratio of one elongation to the preceding. 



Mean . . 0646 

 Thus the movement of the hollow bar is damped in three oscillations as 

 much as the movement of the solid bar in five oscillations. As, besides, 

 the oscillations of the hollow bar are more rapid than those of the solid bar, 

 the advantage of the hollow bar is made again much more sensible. 



