SIE B. C. BRODIE ON THE ACTION OF ELECTRICITY ON GASES. 
4G5 
The number of these experiments is thirty-three. 
Hence the probable error of the result = 0 • 6 7 4 5 
= 0'02 cub. centim. ; 
and also 
the probable error of a single experiment =\/ 33 X the probable error of the result 
= ^33x0-02 
=0T1 cub. centim. 
It appears, therefore, from these experiments that it is an equal chance that the true 
value sought of the ratio B lies between the values 2-01 and 1-97. The value of this 
ratio indicated by chemical theory is 2 ; the preceding experiments, therefore, entirely 
agree with this theory. We may also infer, from the calculated value 0T1 of the error 
of a single experiment, that if we should proceed to make another experiment by the 
same method it is an equal chance that the value of B in that experiment will lie between 
the values T88 and 2T0 ; and half the values of B found by the preceding observations 
should lie between the same limits. As a matter of fact, out of these 33 experiments 
17 experiments are within these limits, and 16 experiments are outside these limits. 
The experiments, therefore, are in perfect accordance with the calculated value of the 
probable error of a single result. 
These experiments are fully confirmed by certain experiments of Meissner'* previously 
referred to, which really throw great light upon the subject, although this chemist has the 
art of singularly misinterpreting his results. The object of the experiments was to effect 
a comparison between the weight of a volume of oxygen equal to the contraction of the 
electrized gas and the increment of weight of a solution of iodide of potassium, acidulated 
with sulphuric acid, when the same electrized gas was passed through the solution. The 
contraction was (as I understand the experiment) estimated by comparing the volume of 
the oxygen before and after its passage through the induction-tube, which was an induc- 
tion-tube of the form devised by Von Babo ; the weight corresponding to this contraction 
was then calculated, and the increment of weight of the acid solution of iodide of potas- 
sium was determined by weighing the apparatus in which it was contained before and after 
the experiment. I shall not pretend to criticise these experiments, the uncertainty of 
w 7 hich is fully admitted by the author. Among other sources of error, however, the total 
increment of weight of the acid solution of iodide of potassium is comprised in the several 
experiments between a maximum of 0016 gramme and a minimum of 0‘ 005 gramme, 
quantities which it is evidently very difficult to estimate with precision in this way. I 
would rather direct the attention of the reader to the mean result of these twelve expe- 
riments, which gives 2 '02 as the ratio of the increment of weight of the acid solution of 
iodide of potassium to the weight of a volume of oxygen equal to the contraction of the 
electrized gas. From these observations, taken in connexion with the Table of experi- 
ments before given, we may consider it as conclusively proved, not only that the oxidation 
* Neue Enter suchungen iiber clen elektrisirten Sauerstoff, 1869, p. 95, Tabelle F, 
