12 Messrs Paine and Evans, The Conductivity of 
which corresponds to 14:5 x 10~° + 0:340 = 42°7 x 10~ gr.-equiv. Oo 
HCO, (and of the corresponding H ions) per litre. This occurs 
at a point of the curve for which the abscissa is 43-4 x 10~* gr. 
equiv. of H ions added in the H,SO,. Hence the quantity ol 
H ions still in solution as such is the difference of these, viz, 
0-7 x 10-* gr.-equiv. per litre (= m,)*. | 
The fact that these two values of m, are the same is very 
strong evidence that we are dealing with the same chemical on 
physical change in both cases, that is to say, that the initial) 
irregularity in the sulphuric acid curve is due to this particular) 
action proceeding. ; 
Ve 
These experiments and calculations point with considerable | 
certainty to the conclusion that the impurity constituting the | 
main part of the residual conductivity of ordinary distilled water 
provides a sufficient explanation of the irregularity observed | 
with extremely dilute acid solutions. They also point to the con- | 
clusion that this impurity consists mainly of a carbonate. 
Similar measurements on alkali solutions should throw some 
light on the nature of this carbonate. As already mentioned, | 
* The recombination of the ions of carbonic acid is probably more complicated . 
than what we have assumed above, taking place in two stages, “a 
Ht+C0O3;--=H CO;-, 
and Ht+HCO0O;-=H,COs. 
Let m,,m,., and mg represent the same quantities as before, and let m, represent || 
the number of gram-equivalents of HCO, ions present. Then we have for equili- | 
brium 
M4 My=P1-™M4, 
and M1 M4=Py+ Ms, 
where p; and p, are constants; | 
) 
m 2 
whence 8 . 
m2 PiP2 
an equation identical with the previous one. F 
As for the conductivities, assuming the velocity number of the HCOs ion to be © 
about the same as that of the $CO3 ion (which is probable) the ‘loss of conduc- — 
tivity’ at Y(=XY) will be that due to the disappearance of (m3+ 4) gr.-equiv. of 
H and CQs ions, and the loss of conductivity still possible (=YZ) will be the © 
XY _m3+m ait 
conductivity of (m, +4) gr.-equiv. of H and CO; ions. Hence —— 
where this ratio is unity (for which case the above numerical calculations are made), | 
YZ ing +g” 
. XY ms : : : 
we have again YZ Sale Also, (m3+ma) gr.-equiv. of H ions have disappeared, 
ilo 
1.e. aon gr.-equiv. The calculation we have made on the simpler assumption 
still holds therefore. (Note that the total loss of conductivity possible is due to 
the disappearance of (mg+m3+2m,) gr.-equiv. of H,CO, from the ionised state, 
m, gr.-equiv. of HCOs ions giving rise to 2m, gr.-equiv. of H,CO3.) ; 
