32 



SCIENCE. 



[N.S. Vol. XVII. No. 418. 



rauch the most likely direction of assault 

 from a lizard.' According to this writer, it 

 is when the butterfly requires to rest that it 

 settles, not on the under side of a leaf, as do 

 most other butterflies, but ' on the bare trunk, 

 or one of the larger boughs, of a tree.' 



Nat. Sci. 



SEORTEB ARTICLES. 



DATA WITH A POSSIBLE BEiiRING ON THE CAUSE 



OF LIGHTNING. 



1. Lenard inferred from his experiments 

 that it is necessary for the water jet to strike 

 a solid obstacle to generate the electricty ob- 

 served in the surrounding medium of air. I 

 find that a surface of water is also efficient, 

 and I place the electricity as a charge on the 

 water nuclei produced, because the charge in- 

 creases with the number of nuclei computed 

 from their coronas. In other words, the mere 

 attrition of water by water is sufficient to 

 charge the nuclei. 



In a forthcoming paper in the American 

 Journal of Science I show that if each nu- 

 cleus carries one electron, there must be at 

 least 10^ nuclei per cubic cm. ' At least,' 

 because much of the charge is lost in the 

 tube which conveys the nuclei into the con- 

 denser, and I have not yet allowed for this. 



From the coronas simultaneously produced 

 I find that about 5,000 nuclei are present per 

 CO. Hence each nucleus carries 200 elec- 

 trons, while its potential is below five volts. 

 Thus there are a million electrons in each 

 c.c. of the air which I examine, or in a cubic 

 kilometer there would be 10^1 electrons, or 

 7 X lO'^^ electrostatic units of charge, or 

 about 200 coulombs. 



2. Let this region be spherical with a 

 superficial capacity, which would then be 

 .62X10^ cm. The potential* at the sur- 

 face of the region would be eleven million 

 electrostatic units of potential, or over three 

 thousand million volts, if the nuclei were all 

 of the same sign and were transferred to the 

 surface. Every time the region is emptied 

 of its nuclei, the surface acquires a charge of 



• For a mile flash of lightning 70 coulombs at a 

 million volts are usually conceded (Lodge). 



the enormous potential stated, and there is no 

 reason why the nuclei may not be continually 

 produced by attrition while they are being 

 transferred. 



3. Now regarding the transfer of nuclei, 

 we may note that when they are produced 

 from pure water, positive charges are usually 

 in excess; when produced from dilute solu- 

 tions, negative charges are usually in excess; 

 but I find that the bulk of the nuclei are 

 symmetrically positive and negative. 



The velocity of the nucleus of charge e, in 

 an electrostatic field of the potential gradient, 

 E, is v=^Ee/Qi'^H-U, where B is the radius of 

 the nucleus and F- (.0002, say), the viscosity 

 of air. Put, therefore, in this equation the 

 values which I have here and in other places 

 adduced, R = 1Q-^ cm., e = 200 X 7 X 10-^0 

 electrostatic units, whence i; = 37 cm./sec, 

 or over four fifths mile an hour, for the unit 

 electrostatic field; about .003 mile per hour 

 of a field of one vqlt/cm. 



Thus there is considerable mobility in these 

 nuclei. With a strong electrostatic field at 

 least locally in action, the nuclei of one sign 

 would thus be driven outward, warmer nuclei 

 into colder regions of continually increasing 

 conduction or rarer air, where their charges 

 would be dissipated. The other nuclei would 

 be driven earthward, colder nuclei into 

 warmer regions of continually decreasing 

 conduction to be discharged, if at all, by a 

 flash, particularly if, on growth of drops, 

 gravity steps in as a final motor. 



Whether there is sufficient commotion in 

 thunder-storms to give rise to the attrition 

 of water, whether comminution will not suf- 

 fice if accomplished in other ways, whether 

 an earth electrostatic field is an adequately 

 permanent or localized occurrence, whether 

 indeed separation of nuclei is needed if there 

 is enough excess of charges of definite sign, 

 must be left for further consideration; but 

 it seems to me that data bearing on the occur- 

 rence of lightning are here suggested which 

 deserve serious scrutiny. 



0. Bar09. 



Bbown Univebsity, 

 Pbovidence, R. I. 



