9S 



SIXTH REPORT — 1836. 



for the effect of the momentum of the needle, unless the centres 

 of gravity and of the axle should be strictly coincident, which is 

 a nicety of adjustment rarely if ever attained. The amount of 

 this correction is learnt by comparing the dip observed at some 

 one station with that needle, and with others of the ordinary con - 

 struction in which the poles are changed in each determination. 

 The following observations, made at Limerick, furnish this cor- 

 rection. 



]. Needle S(2). 



2. With other needles. 



Whence it results that — 12' is the correction for the dips ob- 

 served with Needle S (2) ; and from the small amount of this 

 correction it may he regarded as constant within the limits com- 

 prised by the observations in Scotland. The dips inserted in the 



* In Meyer's needle, by the use of spheres of different magnitudes in the 

 different sets, the results are obtained from arcs difTering very widely from 

 each other, and in which the needle rests on very different parts of the axle. 

 The avoidance thereby of any constant error, caused by the imperfect cun'ature 

 of some particular parts of the axle, is one of the advantages of a needle on this 

 construction which ought not to be lost sight of, or unattended to in observing 

 with it. The partial results may be wider if the axle be not very truly ground, 

 but the mean is more likely to be free from error. 



