= 
Ur 
No 
lep} 
ek 
168? c 
Ee 44u? —dau? + B 
Je 
nin je + 6u! + utd 2a(u—! Hu?) +e | EE 
. . Cust 
in which we have put a=— 
The form (9) yields the force of resistance by multiplying by 
P . 5 3 5 
Ee coskt and from this we further find the energy used up. The 
T 
quantity p,, however, still remains unknown. We can find this by 
considering that the refractive index u also depends upon the vibra- 
tions performed by the electrons. Thus we find e.g. for the critical 
point where a=0O and F#—1: 
mv (u’—l)? e+ | 
fe ra eae Gye 
Ne°2? 16u | 5 Ee re) u—l 
being an extinction proportional to 4°, and depending directly upon 
the quantity ¢, just as was found I. e. for the opalescence. Now I 
especially wanted to find the way in which the extinction-coefficient 
increases on approaching to the eritical point. For this it will suffice 
to consider the variability of the expression between square brackets 
in (9). The quantity 
je & 
may be used as a measure of the ‘distance to the eritical point.” 
For large values of a? we may expand the logarithm in that 
expression into descending powers of a?. On reduction it is seen that 
the term with a? is the first that has a coefficient differing from 
zero, and that for large values of a? we can take for the whole 
expression 
64 64 
5 Terr rr oe Aeon. (C110) 
Only preserving the first term, this will duly give an extinction 
inversely proportional to 1—F. 
In my thesis I have graphically represented extinction-meas- 
urements of a liquid mixture, by plotting the reciprocal value of / as 
a function of 7— 7). The points lay rather accurately on a straight 
line, which cut the temperature-axis below 7. Using this same 
method with the quantity (10), we find for the development of the 
reciprocal value of (10) 
