NipJier — Magnetic Disturbances. 161 



negative electrons moving in the same direction as the ex- 

 plosion. In other cases the explosion was not complete. 

 In many cases not more than six or eight inches of the 

 column would explode, and the results were conflicting. 

 In the two cases above referred to the deflection was in 

 one case 51 minutes of arc and in the other 54. 



It was found that vibrations of the trough, or breaking 

 of the bars into long and large fragments which were 

 not detached from it resulted in disrupting the column 

 of dynamite. It was found that large rocks hung from 

 the trough had a tendency to keep it in stable position 

 and prevented the vibrations. 



It has since been found that when the dynamite is 

 slowly poured into a long rubber tube, which is gently 

 tapped in order to make the column continuous, the ex- 

 plosion can be made complete. The tube must be hung 

 in a vertical position while being filled. This work on 

 the convection current is in an unfinished condition. 



