232 
NATURE 
[ Fan. 4, 1883 
for caution is shown; ¢.g. samples containing the refuse of 
canning tomatoes might have been erroneously thought con- 
taminated with animal matter; others, containing a watery in- 
fusion of human fzces, with vegetable matter, &c. 
Of the biological results under the same head, the most note- 
worthy is the well-marked pathological effect on rabbits of the 
injection of waters contaminated solely by such vegetable matter 
as would usually be thought harmless, e.g. peaty water. True, 
in the well-marked cases, the amount of organic matter present 
was large, but not beyond that in water sometimes used for 
drinking purposes. The Dismal Swamp water is an example ; 
it has often been chosen for ship-supply, and has been spoken of 
as a source of supply for the city of Norfolk. On the theory 
(which has much in its favour) of disease caused by drinking 
water being due to the presence and action of living organisms, 
there might possibly be safety in drinking a peaty water, or 
water filtered through dead forest leaves, when fresh; danger, 
when, after some exposure, bacteria had been developed ; and 
safety, again, perhaps, after the growth of these had fallen off, 
and more or less of the available organic matter had been con- 
sumed. Ship-captains say the Dismal Swamp water, after a 
time, becomes remarkably good and whulesome. 
As to the putrescent or non-putrescent character of organic 
matter in water, the chemical evidence goes to prove (in opposi- 
tion to Tidy’s opinion) that the proportionate consumption of 
oxygen from permanganate within the first hour is rather greater 
for those waters containing vegetable than for those containing 
animal matter. Dr, Smart has expressed the opinion that 
gradual evolution of albuminoid ammonia indicates the presence 
of organie matter (vegetable or animal) in a fresh or compara- 
tively fresh condition, while rapid evolution indicates that it is 
putrescent. His interpretations in this respect proved to be 
correct in a large proportion of eases, but not always. 
The biological results under this head accord, on the whole, 
with the general belief that putrescent organic matter is more 
dangerous than that in a fresh or but slowly decomposing 
condition. 
Prof, Mallet proceeds to state some general conclusions with a 
view to sanitary application as to the value, separately and collec- 
tively, of the different processes of water analysts which have been 
under examination, 
It is not possible to decide absolutely on the wholesomeness 
or unwholesomeness of a drinking water by the mere use of any 
of the processes examined for the estimation of orgamtc matter, 
or its constituents. Not only must such processes be used in 
connection with investigation of other more general evidence, as 
to the source and history of a water, but this should even be 
deemed of secondary importance in weighing the reasons for 
accepting or rejecting a water not manifestly unfit fer drinking 
on other grounds, 
There are no sound grounds on which to establish such 
general ‘‘standards of purity”? as have been proposed, looking 
to exact amounts of organic carbon or nitrogen, ‘‘albuminoid 
ammonia,” oxygen of permanganate consumed, &c., as permis- 
sible or not. 
Chemical examination may be quite legitimately applied, first, 
to the detection of very ross pollution” (as of a well from crushing 
of soil pipes), and secondly, to periodical examination of a water 
supply, so that suspicious changes from the ascertained normal 
character of the water may be promptly determined and their 
cause investigated. ‘In the latter connexion there seems to be no 
objection to the establishment of /oca/ ‘‘ standards of purity,” 
based on thorough examination of the supply in its normal 
condition. 
_ A careful determination of the nitrites and nitrates seems very 
important. 
If he had to watch a large city water supply, the author would 
use all the three processes; each gives information which the 
others do not. Where only simple means were practicable, the 
albuminoid ammonia and permanganate processes might be em- 
ployed together ; but in no case should one only of these methods 
be resorted to. 
Practical Suggestions as to the Use in their present form of the 
Chemical Processes Studied,—In general, water samples should 
be examined with the least possible delay after collection. 
Besides examination of a water in its fresh condition, samples of 
it should be set aside in half-filled but closed glass-stoppered 
bottles for (say) ten or twelve days, and one of these examined 
eyery day or two, to trace changes undergone. 
In the case of the combustion process, however skilful the 
| analyst, duplicate or even triplicate concordant results should 
be insisted on. To avoid the presence of ammonia from coal 
gas, in the atmosphere about the water-bath, the bath should be 
heated by steam brought in a small closed pipe from a distant 
| boiler (preferably in another room), and the waste steam and 
condensed water should be carried off to a safe distance. 
As to the albuminoid ammonia process, it would be well to 
adopt the rule that the distillation be stopped when, and not 
before, the last measure of distillate collected contains less than 
a certain proportion, say per cent., of the whole quantity of 
ammonia already collected. To diminish the loss of amines or 
other volatile forms of nitrogenous matter, a separate distillation 
should be made with alkaline permanganate added af once, 
besides the usual course of treatment prescribed by Wanklyn, 
and the results of the two distillations compared. The details 
of the evolution of ammonia should always be given. 
The Tidy form of the permanganate process is rather to be 
recommended than that of Kubel, if but one be used. Thetime 
during which the permanganate is allowed to act in the Tidy 
process should be increased to at least 12, better to 24 hours, 
several determinations, on different samples set aside at the same 
time, being made at (say) 1, 3, 6, 9, and 12 hours, to trace the 
progress of the oxidation. 
Suggestions as to possible Improvements on the Processes 
examined deserving further Investigation —Combustion Process.— 
The author proposes to evaporate the water in a closed vessel 
immersed in a water bath, and connected with a good (water 
jet) air pump, a condensing worm being provided for the aqueous 
vapour, the feed to be managed through a nearly capillary tube 
with a glass stop-cock. The evaporation would thus te effected 
within a moderate time at a fixed temperature much lower than 
the boiling point. The loss of organic matter by simple volati- 
lisation or oxidation would be greatly reduced ; much less sul- 
phurous acid would be required ; the tendency to formation of 
sulphuric acid would be reduced to a minimum, and absorption 
of ammonia from the atmosphere about the dish quite prevented. 
In te ting this last effect, two bulb tubes containing pure sul- 
phuric acid might be interposed between the vacuum chamber 
and the pump. 
For certain reasons it might be well to evaporate at first with 
the addition of a small excess of magnesia (as recommended by 
Lechartier), thus removing all ammonia, and then, the water 
having been brought down to a small volume, add a moderate 
excess only of sulphurous acid with a drop of a solution of 
ferrous salt (as directed by Frankland), and complete the evapo- 
ration to dryness—the whole in a jet pump vacuum, as suggested. 
Further experiments as to the Williams method (copper-zinc 
couple) for removal of nitrates, are desirable. 
From preliminary experiments, the author thinks nitrates and 
nitrites may be completely reduced by evaporating to a small 
bulk with no great excess of phosphorous or hypo-phosphorous 
acid, guarding against evolution of phosphuretted hydrogen by 
use of a low temperature, then adding magnesia in small excess, 
and completing the evaporation. The plan deserves to be care- 
fully tested. 
Albuminoid-ammonia Process, including Determination of free 
Ammonia.—To prevent (or at least largely reduce and make 
uniform) the loss of ammonia from imperfect condensation, the 
author would use a retort in a saline solution kept heated by 
steam (at say 102° or 105° C.), and condense in a glass worm 
surrounded by ice water, till the distillate should be brought to 
a uniform temperature not over (say) 5°C. It might be still 
better to distil in a completely closed apparatus, with a fixed 
difference of temperature between the retort and the far end of 
the condensing tube, with glass stopcock to draw off the distillate 
in successive measured portions, and a small safety-valve near 
the cold end. 
In determination of free ammonia, it might be well to try 
a closed distilling apparatus connected with a (water-jet) air- 
pump, So as to maintain a partial vacuum within, keeping tke 
retort at a fixed temperature much below 100° C., and collecting 
all the ammonia in a flask and one or two bulb tubes, with weak 
mineral acid placed between condenser and pump. There would 
be the disadvantage, however, that the progress of evolution of 
ammonia could not be easily traced by its collection in separate 
successive measures of the distillate ; and it would be necessary 
to ascertain whether the application of the Nessler test would 
be at all interfered with by the sodium salts formed from the 
acid used. 
To overcome, if possible, the most serious difficulty in the 
ee 
