550 Prof. Wartmann on Electric Irldiictidri. 



tubes, of equal dimensions, and made from the same tube; 

 corks fixed from distance to distance separate the two wires 

 of each pair. The test-tubes, united by a leaden plate, in- 

 tended to keep them vertical and parallel, were immersed in 

 a tinned copper reservoir half filled with a solution of chloride 

 of calcium in water. Lastly, a double current lamp keeps 

 this bath at a constant temperature, indicated by an excellent 

 standard thermometer graduated on its own tube. fvjUn sna 



QQ. All the communications from the extremities of the 

 pairs with the rheometer are effected by means of M. Pog- 

 gendorff's clamps, which have the advantage of securing an 

 excellent contact without springs or mercury. They are 

 massive parallelopipedons of pure copper, in which at intervals 

 three parallel cylindrical apertures had been bored, which are 

 traversed perpendicularly to their axis by three screws, a? Jib 



67. To elongate the circuit of the pair of variable length, 

 I employ Mr. Wheatstone's rheostat. The one I have used 

 is formed of two cylinders of 0'"*06 diameter; their clamp re- 

 ceives the brass wire No. 3, which makes sixty revolutions, and 

 the serviceable length of which is 1 1™-3. This additional wire 

 must be of an homogeneous texture, perfectly cylindrical, and 

 of a very good conducting nature, without which it would oc- 

 casion perturbati9i}8'An:'t4e notations, which would obscure 

 the final result. •) <:-r;tff biU!' . (i :;i jL.';mJ?. 



.^^j^8. The arrangement adopted for the experiment consiiSts 

 j||, placing the rheostat in the circuit of one of the pairs, and 

 in compelling the current of the other pair to traverse the 

 rheometer in an inverse direction to that of the first*. Each 

 current then possesses two ways of communication ; either by 

 the large, short and homogeneous wire of the rheometer, or 

 through the heterogeneous and comparatively longer circuit 

 of the other pair. Experiment and calculation prove that 

 the derivation produced by this second method is so small 

 that it need not be regarded. For example, the current of 

 the second pair giving l^" deviation at -f 114° Cent., the 

 addition of the circuit of the other pair, not heated, causes 

 the needle to fall to 13° 30'. The first pair having produced 

 26° at -f- 114° Cent., the derivation caused by the other re- 

 duced the index to 25° 40', and no longer allected it sensibly, 

 when the deviation was reduced to 4° or 3° by the interposi- 

 tion of a sufficient length of the wire of the rheostat. As the 



,* The following is a numerical example, intended to prove the delicacy 

 winy measuring apparatus. The thermometer marking -|- 143°, — 

 ,b9ii<<i!The first pair made the needle deviate from 0'^ to 39° east 

 rfji// j-JThe second ... ... ... ^0 ...18 west 



«jjjr-|5 And tlie opposition of their currents •••loij.'i'ltl'" 29 east. 



