TANGENT GALVANOMETER. 215 



The inexactitude of the law of tangents in this apparatus was 

 first made clear by the experiments of Despretz,* and Blanchet 

 calculated the correction (746) which should be introduced into 

 the readings. 



Dr. Joulef appears to have been the first to use a tangent gal- 

 vanometer in conditions more in agreement with theory. The needle, 

 consisting of a small bar 5 mm. to 6 mm. in length, is suspended 

 by a silk fibre in the centre of a coil 15 cm. in diameter. We have 



then - = , and the correction does not exceed even with a 



a 30 800 



deflection of 45. To read the deflections, the needle is provided 

 with a glass thread, the end of which moves over a graduated circle. 

 The friction of this thread in air produces a powerful damping ; and 

 the mean of the readings, after reversing the current, gives a measure 

 of the deflection with an error which, according to Dr. Joule, does 

 not exceed 2'. 



In accurate experiments it is generally preferable to give to the 

 galvanometers a smaller degree of sensitiveness by diminishing the 

 number of windings, of increasing the diameter of the coil, and 

 measuring the deflections by the mirror. The term of correction y 

 is then sensibly constant. 



Gaugain's coil (748), conically wound, abolishes the term of 

 correction to the second order, when the needle is at the apex 

 of the cone; but this mode of construction presents practical 

 difficulties, and a dissymmetry which soon led to its use being 

 given up. 



The two equal coils of Von Helmholtz (749) get rid of the 

 term of the second order when the distance of the coils is equal 

 to the mean radius ; and even the term of the fourth order, which 

 depends on the coil when the dimensions of the channels are in a 

 convenient ratio. Lastly, coils with four sets of windings (750), 

 and such with three (751), give other means of obtaining a 

 sensibly uniform field ; but these arrangements seem to have been 

 but rarely utilised. 



Another arrangement, used by Weber, J consists in placing the 

 magnetised needle outside of the coil and in a principal position on 

 the axis or in the mean plane, the axis being always perpendicular to 

 the meridian. 



* DESPRETZ. Comptes rendus, Vol. xxxv., p. 449. 1852. 



t JOULE. Brit. Assoc. Report. Cork, 1843. Scientific Papers, Vol. i., p. 404. 



% WEBER. Electrodyn. Maasbestimmungen, Vol. i., p. 16. 1846. 



