460 Scientific Proceedings, Royal Dublin Society. 
The acidulated solution is finally run off into a porcelain 
dish, and titrated either with the permanganate or bichromate 
solution. 
It remains to determine the value of the ferrous sulphate 
solution ; and in the authors’ experiments this was always done 
under similar conditions to those obtaining in the dissolved oxygen 
determination itself, that is to say, the separating vessel was filled 
with the water under examination, 7 c.c. removed, and the remainder 
poured into a porcelain dish, where it was mixed with the same 
volume of sulphuric acid and the same volume of ferrous 
sulphate solution as were employed in the dissolved oxygen deter- 
mination, and then titrated. In this way the error arising from 
oxidizable substances in the water is eliminated even when their 
amount is considerable. 
In practice it was found convenient to standardize the iron 
solution during the interval required for the absorption of the 
dissolved oxygen, and consequently as a preliminary to the latter 
process, to first of all fill the separating vessel with the water 
under examination and to transfer the latter (after removing 7 c.c.) 
to a porcelain dish ready for the addition of the ferrous sulphate 
and sulphuric acid. 
Calculating the Results—With the stronger solution of 
either the permanganate or bichromate, it is obvious that 
the amount of dissolved oxygen contained in the volume of 
water operated on is expressed in cubic centimetres at N.T.P. 
by the difference between the burette readings for the blank 
experiment and those of the actual determination—with a 
correction for the dissolved oxygen contained in the 5 c.c. of 
the ferrous sulphate solution and the 2 ¢.c. of ammonia. Such 
a correction is, however, small, and in all the authors’ experi- 
ments, the assumption was made that the 7 c.c. of mixed 
reagents contained the same proportion of dissolved oxygen 
as the water under examination.! 
1 Thus, in one of the determinations, the difference between the burette readings 
for the blank and dissolved oxygen determinations was 2°35 c.c. 
Assuming that the 7 c.c. of reagent added contained the same amount of dissolved 
oxygen as the water under examination, then the latter contained per litre— 
2350 
832°5 
= 7°067 c.c. dissolved oxygen, per litre, at N.T.P. 
