in Chemical Combinations. 487 



these stakes, so that the floor of the room forms a platform en- 

 tirely independent of the instrument. 



Although the glass pointer weighs, as I have already ob- 

 served, only 7 or 8 milligrammes, the resistance to its motion 

 presented by the air was so great as to bring the magnetic 

 needle to which it was affixed to a state of perfect tranquillity in 

 ten seconds after the circuit was shut. Therefore I found it 

 quite unnecessary to employ the oil recommended by Lenz as an 

 additional resisting medium. Nor did I find it necessary to 

 make use of any verniers or micrometers ; the only assistance to 

 the naked eye being a glass prism, to look through horizontally 

 when the pointer coidd not be well viewed from a vertical posi- 

 tion. I found no difficulty in reading off the angles of deflec- 

 tion to 2 or 3 minutes of a degree. Those who are accustomed 

 to work with galvanometers will admit, that it would be useless 

 to attempt to arrive at greater accuracy by the employment of 

 means which must necessarily increase the time occupied in the 

 observations. 



The galvanometer was adjusted in the plane of the magnetic 

 meridian, by changing its position until a current passing in one 

 direction could produce a deflection of the needle, exactly equal 

 in extent to the deflection on the other side of the meridian 

 occasioned by a current of the same intensity, but passing in the 

 opposite direction. After the galvanometer had been thus adjusted 

 with very great care to the magnetic meridian, it was found that the 

 glass pointer stood 30' from zero. This error arose from the diffi- 

 culty of cementing the pointer to tbe needle so as to be exactly 

 in the same plane with the magnetic axis of the latter ; but it did 

 not give rise to any serious inconvenience, as it only made it 

 necessaiy to affect the observed deflections with an increase or 

 diminution of 30', according to the direction of the. current. I 

 may mention in this place, that in every observation the position 

 of both ends of the pointer was noted. 



Since the force of torsion of the filament of silk is so trifling 

 that six complete twists of it only produce a deflection of the 

 magnetic needle amounting to 12', and since the length of the 

 magnetic needle is only g^th of the diameter of the coil, it could 

 hardly be doubted that the tangents of the angles of deflection 

 represent pretty accurately the intensity of the transmitted elec- 

 tricity. I have nevertheless made experiments in order to prove 

 my galvanometer. Sixteen large cells of Darnell's pile having 

 been arranged in a series of four, the deflection of the needle 

 under this voltaic force was ascertained for both sides of the me- 

 ridian. The mean of the deflections produced by each separate 

 fell was also noted. It is evident that the resistance of sixteen 

 cells in a series of four is equal to that of a single cell; conse- 

 quently the deflections produced by the above arrangements 



