4 Mr. Barlow on the Electro-magnetic conducting Power 



limits, beyond them the galvanometer must become an ineffi- 

 cient instrument. 



In my experiments I employed twelve copper wires, each 

 two feet long, varying in weight from 17 grains to 1590 grains, 

 and twelve brass wires varying from 38 grs. to 3770 grs. ; 

 the squares of the diameters of course varying in the same 

 ratio. One would naturally think, therefore, had such a law 

 had place, it must have been rendered sufficiently obvious in 

 these experiments ; but so much was it the contrary, that I 

 could perceive no increase of power in any of the wires which 

 exceeded in weight 180 grs. for a length of two feet. 



These results are therefore so decidedly at variance with 

 the law above stated, that I shall be excused bestowing a few 

 lines in describing the manner in which the experiments were 

 conducted. This was as follows : 



Two stout wires about $ of an inch in diameter were bent 

 into the form shown in the diagram, their ends m m } being bent 

 down about an inch, properly amalgamated, and inserted into 

 two wooden cups containing mercury, these cups being about 

 20 inches apart. Two notches were cut in the sides of the cups 

 m m\ in which were laid the specimens to be experimented 

 on, these being placed so that the wire m m' was exactly in the 

 magnetic meridian, and therefore parallel to the needle placed 

 between them but considerably below the wire. These speci- 

 mens it should be observed, were all also bent down an inch at 

 each end and amalgamated, to be, 

 like the poles of the battery, in- 

 serted in the mercury. The notches 

 in the sides of the cups m m' were 

 for receiving each successive spe- 

 cimen, so that no change should 

 take place throughout in the rela- 

 tive position of the wire and com- 

 pass. 



The only remaining point to * 



guard against was the variable 



power of the battery; and to provide against this, a standard 

 wire, weighing 470 grains, was inserted in the cups, and the 

 deviation produced by it observed and registered before and 

 after that produced by each successive specimen ; the mean of 

 the former deflections was then considered as indicating the 

 mean power of the battery for that experiment, and it was 

 thus easy afterwards to reduce all the deflections to those 

 which would have been produced by a constant power of the 

 battery. 



These observations and reductions are stated in the follow- 



[\ 



