Section B. 
CHEMISTRY. 
1.— VARIATION IN THE AMOUNT OP FEEE AND ALBU- 
MINOID AMMONIA IN WATERS, ON KEEPING. 
(illustrated by curves.) 
By A. LI VER&IDGE, M.A . , F.R.S., Professor of Chemistry , University of 
Sydney , JV.S. W. 
In the course of an investigation into the quality of the Sydney 
Water Supply, 1875, 1 had noticed variations in the amounts of ammonia 
after keeping, and in order to obtain satisfactory and trustworthy data 
with respect to such changes — i.e., as far as they were determinable by 
Mr. Wanklyn’s process — samples of water were put up with the utmost 
care, and the amount of ammonia determined from time to time ; a 
separate bottle being opened for each determination. 
When waters were artificially made impure, the admixture was 
effected in a large glass pan, well stirred and otherwise mixed, and 
then transferred to clean clear quart bottles which were filled to the 
top, and the excess squeezed out by the stopper. 
The results in the following tables show very clearly that, in some 
cases, the amounts of free and albuminoid ammonia increase and then 
decrease, and occasionally increase again ; these changes are much more 
strikingly show r n in the curves than in the tables. Until decomposition 
and other natural changes have taken place, the potassium perman- 
ganate process fails to set free a large proportion of the albuminoid 
ammonia from the organic matter, so that a water might wrongly be 
declared pure, if examined when freshly collected or after lengthened 
keeping. 
These following series of experiments were carried out as far back 
as 1876, with the aid of my then assistant, Mr. J. M. Muir, afterwards 
University demonstrator. 
I have hitherto refrained from publishing this investigation, as I 
had hoped to extend the results in other directions ; but as it may be 
some time before additional experiments can be undertaken, I present 
those already obtained. These, however, as far as they go, show how 
very important it is that the reports on the analysis of water should 
state how many days, weeks, or months have elapsed since the 
samples were collected. 
The temperatures given in the table were taken at 9 to 10 a.m. 
and 4 to 5 p.m. ; they were recorded, because more rapid changes might 
be expected in warm than in cold weather. 
1 may remark that the determinations were made as recommended 
by Mr. AVanklyn, and that the sodium carbonate used for the free 
ammonia, and the potash and potassium permanganate used for the 
albuminoid ammonia, were boiled until they yielded distillates free 
from ammonia, and the distilled water used was also distilled from 
potash and potassium permanganate until free from ammonia. 
