PROCEEDINGS OF THE POLYTECHNIC ASSOCIATION. 451 



resistance. No. 3 is a simple plate of copper, whose resistance is 

 nul, or so small that it need not be taken into account. No. 1 is 

 for intensity, No. 3 for quantity, and No. 2 for common mixed 

 currents. 



I now employed a current from four cups of Hill's battery; first 

 through coil No. 1, and then through coil No. 2, against different 

 resistances, from 4.1 to 151.1 miles: the resistance of No. 1 l)eing 

 greater than that of No. 2, I was careful to switch in rheostat 

 coils, so that the sum of resistances of the galvanometer, and the 

 rheostat coils in the circuit should be always equal, thereby secur- 

 ing isodynamous, or equally intense currents. 



The resistances introduced in five observations were 4.1 — 11.1 

 — 41.1 — 81.1 and 151.1 miles. The tangents of the several 

 deflections by No. 1 being divided by those by No. 2, give the 

 following quotients: 4.4 — 4.3 — 4.4 — 4.44 and 4.3. 



The deflections by No. 4 were from 75 deg. to 8 deg. 30 miu., 

 and by No. 2, from 40 deg. 10 min. to 2 degrees. Such results 

 give indisputable evidence of a very true tangent galvanometer. 

 At the same time I noted the deflections of another galvanometer, 

 whose needle is four inches long, coil one-half inch wide, and 

 resistance .9 mile, under the influence of the same isodjaiamous 

 currents. Dividing the tangents of No. 1 by those with this 

 instrument I obtained the following quotients : 2.41 — 1,45 — 

 .84 — .51 and .46. Showing how much more jDowerfully the 

 needle of the old o-alvanometer was influenced when near the 

 meridian, and how the effect diminished as compared with that of 

 coil No. 1, when it deflected so as to carry its extremities more 

 and more outside of the narrow coil. 



Galvanometers have been constructed of large circular coils, 

 open within, fifteen to twenty inches diameter, with a very short 

 needle in the centre, which nearly fulfill the condition required ; 

 but the deviations obtained by a given current are small compared 

 with those of an instrument whose needle is close to the coil, and 

 a coil of much greater resistance is necessary. Such galvanome- 

 ters aie large, cumbersome, and inconvenient, and the changes in 

 the deflections are too minute to ensure great accuracy in the 

 observations. 



Poggendorff, Melloni, Ampere, and others have published inge- 

 nious methods of determining the relative intensities of currents, 

 by any common galvanometer, which may perhaps be sufficiently 

 reliable for ordinary purposes, but in every case laborious compu- 



