182 
0,00015 n. KOH, 0,00030 n. KOH, 0,00050 n. KOH and 
0,00065 n. KOH). 
Especially with K,.SO, we see that these tops are 
arranged all but symmetrically, we can namely draw an 
axis of symmetry along the line, denoting a concentration 
of 0,00040 n. KOH. For KNO, (fig. 5) the symmetry can 
also be well observed. The axis of symmetry here cuts 
the curve at 0,00030 : KOH. It does not run horizontally 
a TT 
ae À 
RS BE A 
LH 
Zee: 
ane 
ee ct 
Se NE 
GC) EE re 
= de-Copis 
= asanau 
en 
LESERE 
SRRRRRRE 
Deere ee 
ef ae le 
SSSR 
2 ASP 5 
De Ë ; 
LE > 
A4 
8 0 ûl 1 3 # 5 6 L 8 9 10 n nm LE] 
conc. K NO. 001n. 
Fig. 5. 
here. Of KCI (fig. 4, page 56) the point of intersection 
of the axis of symmetry lies (at any rate for the topmost 
part, which is lengthened with the removed lower part) 
at 0,00045 n. KOH. The axis of symmetry does not run 
horizontally here either. 
The symmetry in question is for us an indication, that 
the capriciousness of the figures cannot be attributed to 
inaccuracy. The sixfoldedness of the experiments, besides 
the fact, that for the denoting of a point of the curve 
we generally did 18 or still more experiments, are a very 
good warrant for that. The meaning of the symmetry 
