339 
From (a) and (f) (Perreau) we 
find A = 2,83 
... (b) and (e) (Perreau) 
2,92 
... (c) and (d) (Perreau) 
2,98 
... (b) and (e) (Mascart) 
3,02 
...'(c) and (d) (Mascart) 
2,54 
If we take D = 5230 equation (12), § 16., gives 
K = 1,000574; 
the experimental data are 
K = 1,000590 (Boltzmann) 
586 (Klemencic) 
If we were to try to adopt Klemencic’s value as the exact one, 
D would become =5120 and A <C 0 would be found. 
K. Scheel’s results are given below; some values (corresponding 
to X = 5,016; to À = 5,048; to À = 5,460) have again been omitted. 
a) À = 4,358 ... v= 1,0002954 
b) 4,712 .. . ' 2946 
c) 4,922 . . . 2937 
g) 5,780 . . . 2918 
l) 6,152 . . . 2912 
i) 7,056 . . . 2904 
From ( a ) and (i) ... A — 2,68 
... ( b) and (l) ... 3,19 
... (c) and (g) ... 2,93 
(6) 
Scheel’s empirical constants (see § 16.) are as follows 
a = 2870,5.10“ 7 ; b = 16,23. 10“«; 
substituting these values we get from (8), § 16.: 
D = 5226; 4 = 2,96. 
( 7 ) 
( 8 ) 
These results and the preceding ones, derived from Kayser and 
Runge’s, Perreau’s and Mascart’s data, show a close agreement. 
§ 19 . Nitrogen. It is surprising that so little attention has 
been paid to the dispersion of this substance. This is particularly 
unfortunate in view of the fact that the behaviour of nitrogen con¬ 
stitutes a conspicuous exception to a general rule to be considered 
below (see § 24.). 
