Professor Leslie on Electrical Theories. 
li 
rity with which the particles recede from A will therefore gra- 
dually diminish with their distance, and will become totally in- 
sensible, in ordinary cases, a few inches from the electrified ball. 
But the stream which flows from A, will at first obstruct, in 
some degree, the motion which the affluent current has acquired. 
It will, however, diverge exceedingly, and soon convey its par- 
ticles out of the direction of that current. There is another 
cause that will mightily contribute to this dispersion of the re- 
pelled particles. For we have found, that the air which flows 
towards A, is rather denser than the common atmosphere, while 
that which flows Jrom it is somewhat rarer ; and hence the af- 
fluent particles will continually press upon the refluent particles, 
and will gradually deflect their motion. And, if the particles be 
again attracted with sufficient force by the original electrified 
body, and return to it, their retrogade course must lie on the 
outside of their efflux, and envelope it. 
It is indeed impossible to estimate the precise effects that will 
take place ; the forces exerted are variously modified and com- 
bined, and the motions of the two opposite currents give an in- 
tricacy to the whole. From the preceding investigation, it ap- 
pears, however, that the velocity and quantity of the aerial 
stream depends, not on the distance of AG, at which the body 
sensibly acts, but on the greatest intensity BS of the attraction. 
Hence it will be easy to determine the difference of effect oc- 
casioned by the shape of the body. Let P (PI. I. Fig. 2.) be any 
particle placed extremely near two unequal balls, AGCI and 
AFBK, and acted on by their surfaces. From P, with any ra- 
dius PG, describe the circles GOI and FQK on the two spheres ; 
and from the same centre describe the other circles o i and Jo 7c, 
contiguous to the former. It is well known, tliat the surfaces 
GAI and FAK, are equal to the cylindrical rings, which have 
AGCI and AFBK for their circumferences, and AD and AE 
for their altitudes ; consequently these will be as AC X AD, is 
to AB x AE, or as AG 2 to AF 2 . But the nearer P is taken 
to A, the ratio of AG to A F, will approach nearer to that of 
PG to PF, and, therefore, to that of equality. Whence the space 
GAI will ultimately be equal to FAK; and, for the same rea- 
son, the space GAI will ultimately be equal to FAK. Conse- 
quently, the zones GOI iog, and FOK fc o f on the surfaces of 
