1006 
IL. No reaction is possible, in which all phases of the invariant 
point may participate. 
When e.g. the phase /, cannot take part into one single reaction, 
then in (1) and (2) a, becomes =O. Then we have an invariant 
point with m+ 1 phases, for which the considerations sub J are true. 
Leiden, Znorg. Chem. Lab. (To be continued). 
Physics. — “Krperimental Inquiry into the Laws of the Brownian 
Movement in a Gas.” By Miss A. SNernLace. (Communicated 
by Prof. P. Zeeman). 
(Communicated in the meeting of Feb. 24, 1917). 
1. In a former paper’) some objections have been advanced 
to Ernsrein’s formula for the Brownian movement by Prof. Van 
DER Waars Jr. and me. According to this formula: 
AE eds Ae: (1) 
in which A? represents the mean square of the displacement which 
a “Brownian particle“ obtains per second in a definite direction. 
Equation (1) has been derived on the supposition that the particle 
meets in its movement with a resistance of friction. Accordingly B 
is the inverse value of the factor of resistance which is found when 
the particle travels with constant velocity under influence of an 
external force. Statistical mechanics, however, teaches that a particle, 
in equilibrium with the surrounding molecules, does not experience 
a force dependent on its velocity, hence no ordinary friction. We 
have written the equation of motion in the form: 
ia Ri ne Sag eee Py 
and derived a value for A*, which does not lay claim to great 
accuracy, but leads, at least for the Brownian movement in a 
ae Fowl 
gas, to <A? being proportional with =; when a represents the radius 
of the particle. 
According to Stokes’ formula with CUNNINGHAM’s correction : 
if 
Orban. tan es ee See eee 
pi (3) 
in which § represents the coefficient of friction of the medium and 
À mi 
k= (2 — 45) 
a 
1) These Proc. 18, 1916, p. 1322. 
