574 Dr. G. E. Allan on 



and testing the compensation in the new position of the 

 needle. 



The presence of these H-disturbing components may be 

 more easily tested and their magnitude gauged by timing 

 the period of oscillation of the needle with no current, and 

 with direct and reversed currents in the magnetizing and 

 compensating coils after compensation has been made in the 

 usual way. 



This method was used in the previous work, and indicated 

 a change in the value of the horizontal field at the needle of 

 over 5 per cent, when the magnetizing current was reversed. 



The effect was neutralized in the later experiments by the 

 addition to the magnetizing-coil circuit of a small auxiliary 

 compensating coil. The latter consisted of seven turns of 

 wire 9 cms. in diameter, and was placed to the north of the 

 needle at about 20 cms. distance, with its axis in the meridian 

 through the needle. The coil was in series with the magne- 

 tizing and compensating coils, and the current was passed 

 through it in such a direction as to neutralize forces intro- 

 duced by a want of alignment of the coils of the nature 

 shown in fig. 1. 



Fi2. l. 



b C 





i 

 i 



I 



A is the magnetizing coil ; B, the compensating coil ; and C the 

 small auxiliary compensating coil. 



A position of the coil C (fig. 1) was found such that no 

 change took place in the period of the needle when the 

 magnetizing current was reversed. One coil was found to 

 be sufficient to overcome the disturbing effect, and when 

 adjusted the compensation was good at all parts of the scale, 

 the variation of the field-intensity being reduced to less than 

 0*5 per cent. 



The Basalt Bars. 

 A preliminary examination of the fiYe new tars was made 

 as to their permanent magnetism, and the bars, when placed 



