1485 
We further follow the method applied by Enystem and Hoer *) to 
another problem, by Mrs. pe Haas—Lorenrz*) to the Browman 
movement, but with a slight abbreviation. We substitute /, for 
m . ; ae Ee 
——— V,cosa, and write equation (2) in the form : 
M + m . 
U, — Guy Ue 
u, — Or == es 
This gives on summation over v intervals t, when f= rt: 
vl y v 
Sup — Oyu, = >R, 
2 1 1 
or 
JN S75 
aen An 
If we increase f, the absolute value of the first term of the first 
member increases in general, u,,; — u, however, does not. If ¢ is 
sufficiently long we may neglect these terms and write : 
v 
DING 
4 i 
tT 10 
If we raise both members to their square and take the mean for 
a great number of fs, this becomes : 
Sica) 
mt (1—64))? 
The A's are independent of one another, hence : 
(ER) =v R? 
& is still unknown, but can be calculated in the following way. 
We raise both members of equation (2) to their square, after we 
have first written it in the form : 
up = Oyrup—1 + Br. 
If we now also take the mean for a great number of times r, 
we have: 
(3) 
— yw +26y Ui Rey IR? 
Uji and Ap, are mutually independent, and equally often positive 
as negative, hence 
wer Bra = 9 
and 
1) Einstein and Hoer. Ann. der Phys. 33 p. 1105 (1910). 
2?) G. L. pr Haas—Lorentz. Die Brownsche Bewegung. (1913) p. 51 et seq. 
