W. A. Norton—Molecular and Cosmical Physics. 3838 
waves that fall upon it, from any side; and in so doing * 
intercepts a minute fraction of the wave-force, which 
passes back into the ethereal medium as a reflected wave. 
t us now take two atoms m and n, Fig. 2, and let 
c'nd’ bea conical surface circumscribing n. It is plain, 
from what has just been stated, that a certain portion 
of the impulsive wave-force propagated to n, from the 
ether included within this conical surface, will be inter- 
cepted by n, and thus prevented from passing on to m. 
The impulsive action upon m received from the ether with- 
in the cone cmd, will then be in excess over that received 
from the ether posited within the cone e’md’, by the amount 
thus intercepted by n. The atom m should then gravitate c 
toward n, by this differential force. Let a denote the 
amount of wave-force that falls upon m, from the cone emd; 
a’ the amount that falls upon n, from the conical frustum e’nd’ ; 
is} 
a 
a 
> the amount of impulsive action on m due to the wave-force 
a E 
a;— that which takes effect on n, from the wave-force a’. The 
: ‘ a eee ; 
impulsive forces, > and —, will be intercepted by m and n, re- 
. , . 
spectively. Let —=b ; and the wave-force from c’md’ that is 
propagated to m will be represented by a—3, and the impulsive 
action it exerts on m, by — Then —— —, or —, will be 
the effective force by which m will gravitate toward n; for the 
entire gravitating force will be due to the conical portions of 
ether considered, provided all the points of m are taken into 
account. It is obvious that n will gravitate toward m with an 
ee force, if the two atoms are exact counterparts of each 
other. 
It may be imagined that the deficiency, z in the impulsive 
action received from c’md’, may be made up by the wave-force 
reflected back from n toward. m, but this initial wave-force is 
diffused over the whole extent of the outspreading wave, and 
Proportionally weakened at any one point, as m, which it 
Teaches ; and hence the portion that falls on m is buta minute frac- 
tion of the initial impulse, — reflected at n toward m, and 
? 
it is on] 1 rong ne : ke effect 
only the — part of this minute fraction that can take e 
by 
impulsively on m. Suppose = of <= is propagated to m, 
