THE FIELD OF EXPERIMENTAL KESEARCH. 129 



come largely from the apparent al)ility to secure a representation of 

 the eiiects of lightning discharges upon a moderate scale, and the pos- 

 sibilit}' of studying the action of air and other gases, as well as liquids 

 and solids, at varying temperatures and pressures under high electric 

 stresses. Broadly considered, however, the similarity of the efl'ects to 

 those produced in a thunder cloud is more apparent than real. The 

 olobules of water constituting the electrified cloud do not possess 

 charges of millions of volts potential, the effects of which are seen in 

 the stroke of lightning. The individual globules may possess only a 

 moderate charge. When, however, they are massed together in a large 

 extent of cloud the virtual potential of the cloud as a whole, with 

 respect to the earth, may be enormous, though no part of the cloud 

 possesses it. The cloud mass not being a conductor, its charge can not 

 reside upon its outer surface or upon its lower surface nearest the 

 earth, as with a large insulated conductor. The charge, in fact, exists 

 throughout the mass, each globule of water suspended in the air hav- 

 ing its small effect upon the total result. 



When the cloud discharges, the main spark branches within and 

 through the cloud mass in many directions. The discharge can at best 

 be onh' a very partial one, from the nature of the case. These are 

 conditions which are certainly not represented in our experimental 

 production of high-potential phenomenon, except perhaps upon a very 

 small scale in the electrified steam from Armstrong's hydro-electric 

 machine, a type of apparatus now almost obsolete. Yet if we wish to 

 reproduce as nearly as possible, upon a small scale, the conditions 

 of the thunder cloud, we shall be compelled to again resort to it. In 

 volcanic eruptions similar actions doubtless occur and give rise to 

 the thimder clouds which often surround the gases sent out from the 

 crater. 



Considering, then, that the conditions in the thunder cloud are so 

 different from those in our experiments with high potentials, we can 

 easily understand that the study of lightning phenomena may present 

 problems difficult to solve. Two forms at least of lightning discharge 

 are quite unknown in the laboratory — namely, globular lightning and 

 bead lightning, the latter the more rare of the two. Personally I 

 can not doubt the existence of both of these rare forms of electric dis- 

 charge, having received detailed accounts from ej^ewitnesses. On 

 one occasion, while observing a thunder storm, I narrowly missed see 

 ing the phenomenon of globular lightning, though a friend who was 

 present, looking in the opposite direction, saw it. The explosion, 

 however, was heard, and it consisted of a single detonation like the 

 firing of a cannon. According to the testimony of an intelligent eye- 

 witness, who described the rare phenomenon of bead lightning within 

 an hour after it had been seen, it is a very ))eautiful kuuinous 

 appearance, like a string of beads hung in a cloud, the beads being 

 SM 99 9 



