668 Professor W. G. Adams [June 3, 



of currents during tliis storm was from south-west to north-east, or 

 from south-south-west to north-north-east, with varying intensity, 

 the agreement being very close between the disturbances of 

 the declination needle and the Blackheath and Greenwich photo- 

 graphic record. From Mr. Preece's record also earth currents were 

 violent from 10.30 a.m. on the llth (i. e. they were noted within ten 

 minutes of the beginning of the magnetic storm) to about 2 . 30 p.m., 

 and again from 9 to midnight. They were very violent on August 

 12th, beginning at 11.30 a.m., the beginning of the second storm, 

 and quieting down about 4.30 p.m., then beginning at 7.30 and 

 lasting until 9 . 30 p.m. 



Again on the 13th, they are strong for 1^ hour from about 5 in 

 the morning, i. e. just about the end of the second magnetic storm. 



The general direction of the earth currents as observed at Derby 

 or Haverfordwest, as well as at Greenwich, was from north-east to 

 south-west. 



Again on January 31st last another violent magnetic storm, in 

 which, Mr. Preece tells me, the currents were even more violent than 

 in the August storm. 



Intimately connected with magnetic disturbances and earth 

 currents is the phenomenon of the aurora or polar light, which is an 

 electric discharge in the upper regions of the atmosphere. During 

 the August and January storms the aurora was well seen in England, 

 it was also seen at St. Petersburg and as far east as Siberia. 

 It does not appear to have been seen, although it was looked for, 

 at Zi-ka-wei, in China, by M. Dechevrens, the Director of the 

 Observatory, although the magnetic storm was so violent there that 

 the horizontal force was suddenly changed by one hundredth part of 

 its total amount. 



We may arrive at some idea of the character of the aurora by 

 studying electric discharges in vacuum tubes, and Dr. De La Eue 

 has already brought this subject before you in his Friday evening 

 lecture. 



We may gradually pass from electric discharges in air of ordinary 

 density, in which we get the well-known electric spark between two 

 surfaces, to air of loss density but better conducting power, and then to 

 air of less density still, but of such high resistance that no electricity 

 will pass. Dr. De La Rue has shown that with 11,000 cells of his 

 battery the striking distance between two points is about six-tenths 

 of an inch in air of ordinary density of about 760 mm. pressure. 

 When the pressure in a hydrogen tube is reduced to 21*7 mm., 8937 

 cells will cause a discharge to take place through 30 inches. When 

 the pressure is reduced to • 642, about six-tentbs of a mm., 430 cells 

 will cause a discharge through the tube. When the pressure is still 

 further reduced to -0065, it requires 8937 cells to cause the dis- 

 charge. So that the spark passes more readily at a pressure • 642 mm. 

 than it does at a higher or a lower pressure. 

 This is also the case with air. 



