LIGHTNING PEEK 177 



When lightning strikes the earth at some sandy spot its path is 

 often defined by a ghisslike tube. This tube, which may have 

 branches like a tree, is, in fact, sand fused into " glass " by the 

 lightning current. Such tubes are called '' fulgurites," and can be 

 produced by artificial lightning. 



Because of the transient and explosive nature of lightning some 

 quite unexpected phenomena frequently happen. Data will now be 

 given on typical types of gafis, the needle gap representing the non- 

 uniform field type, the sphere gap the uniform field, and the insu- 

 lator representing the surface discharge. 



Figures 10, 11, and 12 give characteristic sparks. 



The lightning generator consists of high-capacity condensers just 

 as in the case of tlie clouds, only the insulation is glass instead of 

 air and it is relatively more comijact.^ As in the case of the cloud 

 lightning, the electricity is stored at a relatively slow rate and dis- 

 charged at an enormously rapid rate in a few millionths of a second. 



The lightning generator has been of considerable help in gaining a 

 knowledge of natural lightning. In fact, it has afforded a means of 

 estimating the voltage of a real lightning stroke. The means was 

 very simple and will be cited as an example of the general method of 

 the research, which was to observe natural lightning and tlien ob- 

 serve artificial lightning on models built to scale in the laboratory. 

 Actually, the function of the model is to solve electrostatic problems 

 that are difficult to calculate. 



THE VOLTAGE AND ENERGY OF NATURAL LIGHTNING 



It was shown in the experiments at the very high voltages that 

 approximately 150,000 volts maximum are required for every foot of 

 spark. While this rule Avas found to apply up to the ver^'^ highest 

 laboratory voltage of about 3,000,000 maximum, it in itself gave no 

 assurance that it would apply indefinitely, so that the voltage of 

 lightning coidd be found by multipljnng the length of the flash by 

 the factor. However, an experiment Avhich seems to offer an almost 

 direct means of measuring lightning voltages will be described. 

 When a lightning flash occurs within a certain distance of a trans- 

 mission line, a certain percentage of the voltage of the bolt is in- 

 duced on the line. The voltage of the bolt can not be measured, but 

 its distance from the line and height of cloud can be estimated. 

 The actual voltage induced on the line can be measured by gaps or 

 estimated from insulator flash-overs. The author has measured 

 lightning voltages on transmission lines in Colorado as high as 



*A full teclinical doscriptiou of the lightning generator \Till he found in the papers 

 referred to in Note 1. 



