INCLINATION. 591 



observations more rapidly, by noting the division which corresponds 

 to the central wire. 



It is, finally, well to know the torsion couple of the wire. For 

 this it is sufficient to observe the deflection 8 produced by a con- 

 siderable torsion for instance, half a turn and we have 



O = HMsinS, or = ; 



HM 7T 



the wire is not suitable if this ratio is not very small. 



Very different forms have been given to declination compasses. 

 Whatever be the mode of construction, the apparatus should be 

 provided with all the necessary means of correction, and the course 

 of the observations is always the same. 



In the compass which we have described, the point at which the 

 stirrup is attached to the wire of suspension is so high above the 

 centre of gravity of the movable system, that the couple arising 

 from the vertical component does not make the bar incline to 

 any appreciable extent ; but if the needle is supported on a pivot, 

 it is sometimes necessary to allow for this source of error, and 

 in different latitudes we are obliged to bring the needle into the 

 horizontal by small counterpoises. 



1163. INCLINATION. We may determine the inclination of a 

 needle which moves about an axis perpendicular to the magnetic 

 meridian, either by the apparent inclination in a plane which makes 

 a known angle with the meridian, or in two rectangular planes. 



In inclination compasses, or inclinometers, as in that of Gambey, 

 long needles were formerly used to facilitate the readings. Here 

 again, besides that the apparatus is of too great dimensions, the 

 time of oscillation makes the observations far too slow, without 

 adding anything to their exactitude.. 



Fig. 242 represents an inclination compass of very small dimen- 

 sions, in which the needle is only 7 centimetres in length. It is a 

 steel plate in the form of a very acute lozenge, traversed in the 

 centre by a steel axis the ends of which are truly cylindrical, and 

 which rest on two agate knife-edges which are in the same hori- 

 zontal plane. The agates are supported by a frame which rotates 

 on a horizontal circle, and the needle moves in front of a vertical 

 circle. The axis of the needle, in rolling on the agates, may leave 

 the centre of the circle ; it is brought back by raising by a screw 

 a V-shaped stirrup, which lifts the axis and then lets this down 

 gently on the agates. 



