Introduction. xv 



to the west and east respectively : before that date they were in the reverse posi- 

 tions ; consequently, 



Joint effect of Bifilar and Balance Magnets on the Declination Magnet 



before July 30. 1841 = - 0-30 Sc. div. 

 after = + 0-30 Sc. div. 



14. Effect of the Copper Ring. 



No observations were made to determine the error due to the copper ring till 

 1843. Previously, however, it had been placed over the Balance Magnet which 

 indicates immediately^the presence of the smallest quantity of iron ; but the needle 

 remained motionless. 



In 1843, series of observations were made by reading the position of the De- 

 clination Magnet with the copper ring, in its place and away ; the results were con- 

 tradictory, being, for the effect of the ring, from — 1.82 to + 1.89 Sc. div., and lead 

 to the belief that the effect must be small, if anything. The differences, it was pre- 

 sumed, were due to the generation of currents of air by lifting and shutting the box, 

 to the changes of Declination occurring at the time, and to the greatly increased arc 

 of vibration when the ring was removed. 



In July 1843, after an extra Declinometer had been obtained, the results were 

 equally contradictory, the differences were attributed to aerial currents as before, — 

 the result of the best observations being nearly zero : the ring was therefore allowed 

 to remain. 



In October 1844, a careful series of observations was made with all the guards 

 of double boxes, &c., when it was found from the consistent results, that the effect 

 of the copper ring when in its usual position was equivalent to about — 1*0 Sc. div. 

 It is now believed that the differences of the partial results obtained in 1843 were 

 to some extent due to slight changes in the position of the ring in the different ob- 

 servations.* 



The observations for the effect of the copper ring are reserved for the Intro- 

 duction to the Volume for 1843. No correction has been applied for its effect in 

 1841 and 1842, as the position of the ring must have varied slightly during that 

 period. It is believed that the effect must have been between about — 0'*5. 



15. Error of Collimation of the plane glass in the south side of the Declino- 

 meter-box. 



No observation was made to determine this error till 1843. The magnet rest- 

 ing on blocks, the glass being in its usual position, reversed, and away, no difference 

 could be detected in the scale readings. 



* It should be remarked that the method of taking away and replacing, recommended in the 

 Report of the Royal Society, does not seem that best fitted for determining the existence of a disturbing 

 cause. During the Observations of 1844, it was found, that, by slightly varying the position of the 

 ring, the effect might be much magnified, or reduced to zero. 



