335 
ON THE DETERMINATION OF ORGANIC MATTER IN 
WATER. 
(Concluded from p. 296.) 
Determination of Nitrogenous Organic Matter. —Messrs. Wanklyn, Chapman, 
and Smith, in a paper recently published, state that the process of evaporation 
for determining the solid constituents of a water is liable to another source of 
error. 
Certain forms of organic matter, in fact, undergo slow alteration when heated 
with water, in such a manner that if a weighed quantity of such matter be in¬ 
troduced into water undergoing evaporation, it may be made to yield a residue 
of any desired weight by simply adjusting the length of time employed in the 
process, being almost entirely dissipated if the time occupied be sufficiently pro¬ 
tracted. They prove this by evaporating a gramme of urea, for example, with 
quantities of water varying from 50 cubic centimetres to 1 litre; the residue 
was found to vary according to circumstances from -98 to *007 gramme. The 
ordinary method of taking the residue of a water therefore does not yield the 
same percentage amount independently of the volume of water employed. 
The method which is proposed by these chemists in dealing with the organic 
matters present in waters is founded upon the estimation of the ammonia 
actually formed from the nitrogenous matter during distillation with carbonate 
of soda, caustic potash, and permanganate of potash. 
They state that the nitrogen of urea is easily obtainable in the shape of 
ammonia by boiling with carbonate of soda, which does not affect gelatin or 
albumen. Whilst a caustic alkali, they find, evolves from the latter substances 
only one-third of their total nitrogen in the form of ammonia, the other two- 
thirds being liberated by similar treatment with permanganate of potash. 
They thus establish a distinction between the ammonia obtained from urea, and 
that which they designate as “ albuminoid.” 
The following is an outline of the process adopted:— 
When possible, the ammonia in the water is determined by Nessler’s test 
without previously distilling. This may always be effected in waters which 
have very little colour, and contain more than the very smallest traces of am¬ 
monia ; for example, in most of the London well-waters. [The manner of ap¬ 
plying the test, as well as its preparation, will be found in the conclusion.] 
I. A litre* of water is placed in a retort, carbonate of soda, about two 
grammes, is added, and the contents distilled rapidly, the distillate being re¬ 
ceived in a 100 c.c. flask. As soon as the flask is full, it is emptied into a nar¬ 
row glass cylinder and at once replaced to be refilled. Nessler’s test is now 
added to the contents of the cylinder. Another cylinder is filled and treated in 
the same manner, and the operation repeated until no more ammonia can be de¬ 
tected in the distillate. Unless we have an unusually large quantity of am¬ 
monia or urea be present, no ammonia can be detected after the third 100 c.c. 
have come over. We may now mix the distillates, and determine how much 
ammonia the colour corresponds to. 
The ammonia thus found includes that which exists ready-formed in the water 
and that which results from the decomposition of the urea. If the existing am¬ 
monia have been determined previously in the water itself, the difference be¬ 
tween this and the quantity found by the distillation gives the amount due to 
the urea. If we have not been able to make this determination, we can still 
* A quart of water may be placed in the retort and 5 ounces distilled at once. 1 gramme 
= 15433 grains. 
