November 17, 1923] 



NA TURE 



727 



Thunderstorms and Globular Lightning. 

 By Dr. G. C. Simpson, F.R.S. 



''T^HERE is no real boundary between pure science 

 J- and applied science, and it is inconceivable that 

 any one whose life's work is the practical application 

 of electricity should not be interested in all thin<;s 

 electrical. One might, therefore, expect an electrical 

 engineer to show at least a dilettante interest in 

 atmospheric electricity, but one is surprised — although 

 equally gratified — to find that the president of the 

 Institution of Electrical Engineers devoted a large part 

 of his inaugural address on October 18 to the discussion 

 of the electrical potential gradient in the atmosphere 

 and the mechanism of thunderstorms. There has been a 

 great deal of work done on these subjects in recent years, 

 but it cannot be said that the results have yet reached 

 far beyond the small band of workers who are actually 

 engaged in making the investigations. Dr. Alexander 

 Russell has, therefore, done a good service to his fellow 

 engineers in summarising for their benefit our present 

 knowledge and indicating problems still unsolved. 



Dr. Russell accepts the breaking drop theory for the 

 origin of electricity in thunderstorms, but he appears 

 unable to give up entirely the old idea that free electrons 

 form nuclei for condensation in the atmosphere. There 

 are certain ideas which once they have appeared in 

 scientific literature cannot be eradicated no matter how 

 conclusively they are shown to be wrong. C. T. R. 

 Wilson in his classical work on the condensation of 

 water on to ions showed two things : first, with great 

 supersaturation water will condense, in the absence of 

 other nuclei, on positive and negative ions ; and 

 secondly, that no condensation takes place on even 

 the negative ions until fourfold supersaturation has 

 been reached. This latter point is nearly always for- 

 gotten, and until some one has shown that fourfold 

 supersaturation does exist in the atmosphere, meteor- 

 ologists cannot recognise that ions play any role in the 

 proces.ses of atmospheric precipitation. 



The breaking drop theory of thunderstorms has met 

 with very wide acceptance ; for it gives such a simple 

 and complete account of the origin of the electricity 

 and explains so many of the observed facts, such as 

 the part played by ascending air currents, why the 

 lightning flashes are mainly between the base and the 

 top of the cloud, and why the rain carries .sometimes 

 a positive and .sometimes a negative charge with the 

 former preponderating. 



The phy-sical basis of the theory has been examined 

 in great detail by Lenard in Germany and McClelland 

 and Nolan in Dublin, and there can now be no doubt 

 that the breaking of drops does produce a separation 

 of electricity. There was, therefore, every justification 

 for Dr. Russell to give the breaking of drops as the chief 

 source of electricity in thunderstorms, but this is only 

 a part of the complete theory of thunderstorms, which 

 takes into account the part played by hail and explains 

 also those winter thunderstorms in which there appears 

 to be no drop formation. 



The brea!dng drop theory was put forward as tlie 

 result of work during the mon.soon in India, and in the 

 original paper it was said that there had been no 

 opportunity to examine the electrical phenomena con- 

 nected with ordinary rain or with snowstorms. That 



NO. 2820, VOL. I 12] 



opportunity has since occurred, and has given the data 

 for rounding off the theory so that it can now be applied 

 to all kinds of atmospheric precipitation. 



The separation of electricity on the violent disruption 

 of a body is not confined to liquids, but occurs, probably 

 more strongly, when solids are rapidly separated. 

 Rudge's work on the electrification of dust clouds 

 threw much light on this subject. When dust is blown 

 up into the air, the dust particles are found to be 

 highly charged. This is not an effect of frictional 

 electricity as usually understood, because two different 

 substances do not come into contact ; for example, 

 highly charged particles are obtained when sand con- 

 sisting of pure silica is used to make a dust cloud. 

 The effect appears to be exactly the same as in the 

 case of the breaking drops ; a violent separation of 

 parts takes place, the substance obtains one kind of 

 electricity while the other kind passes into the air 

 probably in the form of large ions. 



Rudge's work was undertaken to explain the high 

 potential gradient observed in tropical regions during 

 dust storms, but similar electrical effects are observed 

 during blizzards in polar climates. There is physically 

 no difference between a dust storm and a blizzard 

 accompanied by much driven snow, and in both cases 

 the particles of solid matter become charged in con- 

 sequence of their frequent collisions. This is then the 

 origin of electricity in snowstorms. One difficulty, 

 however, must be faced. If the electrification takes 

 place by collision, how does a sufficient separation of 

 electricity take place to give a lightning flash, for this 

 can only occur after some process has widely separated 

 the electricity set free by the collisions ? The answer 

 is that so long as the cloud contains only snow which 

 settles very slowly through the air, there is no thunder- 

 storm ; it is only when soft hail accompanies a snow- 

 storm that thunder and lightning occur. As the soft 

 hail falls through the snow flakes, electrification takes 

 place on each collision and the falling hail carries away 

 with it large charges of electricity. Thus the fall of 

 the hail effects the separation of electricity which gives 

 rise to the large electrical fields necessary for a thunder- 

 storm. Compared with the electrical effects of a tropical 

 thunderstorm with its heavy rainfall, the electrical 

 effects of a snowstorm are almost insignificant, and 

 during the polar winter, when there is no soft hail 

 as.sociated with the snowfall, thunder and lightning do 

 not accompany the most violent snowstorms. 



Dr. Russell in his address also gave considerable 

 time to discussing globular or ball lightning. He came 

 to the conclusion, which is now very generally held, 

 that this is a real natural phenomenon with an objective 

 existence. The chief characteristic of ball lightning 

 may be summed up as follows : 



(i) The body or ball itself, which is able to retain 

 its individuality as it moves through the air, 

 appears to be composed of gas or matter in 

 some novel luminous condition. 

 (2) The balls appear to exist independently of any 

 large electrical intensity, for they have been 

 observed within closed rooms where large 

 electrical fields are impossible, and have also 



