66o 
THE TROPICAL AGRICULTURIST. [April 2, 1894. 
When the sample bottles were opened in 
Colombo the water from 12 feet depth had a 
slight smell, but wa^s almost quite free from colour. 
The other four samples from the reservoir had a 
marked smell. Their colour seemed to have 
deepened considerably during the twenty-four 
hours after they were drawn, and a quantity of 
ferruginous sediment had been precipitated from 
solution. The samples from tlie Wakoya stream 
and natural spring were clear, colourless, free 
from smell, and remained so till they were 
analysed. 
The analyses shew that tlie amount of free 
and saline ammonia increases to a very marked 
degree with the depth. At the depth of 12 feet 
there was the merest trace, (no donbt due to the 
fact that the samples were drawn in time of 
drought). At 24 feet the amount of this 
ingredient had risen to the proportion that is 
present in suspicious waters, while at 36 feet and 
all below this it had risen to the amount found 
in highly impure waters. The amount of albu- 
minoid ammonia did not vary to any marked 
extent with the depth. As the albunynoid 
matter decomposed, its nitrogen would be partly 
liberated in the form of ammonia, which would 
therefore increase the item which appears in the 
analysis as free and saline ammonia. Tiie amount 
of organic mattei as inferred from the loss by 
ignition of the total solid residue did not vary 
in a definite manner with the depth ; but the 
amount of oxidisable matter increased with the 
depth, as represented by the amount of oxygen 
absorbed from potassium permanganate. 
The amount of mineral matter increased 
with the depth. 
Nitrous acid was only detected in the sample 
from the depth of 48 feet. 
The microscopic examination shewed that the 
more highly-organised living forms, such as 
Cyclops quadricornis, were only to be found in 
the upper stratunr of water. In a deposit col- 
lected from the wire gauze strainers, through 
which the water passes before entering the main 
leading to Colombo, the prevailing organic forms 
were the same as found in the water of the upper 
stratum of the reservoir, viz., numerous cyclopes, 
and other members of the entomostraca, thousands 
of brown globular-shaped monads with very 
slightly protruding oral orifices. After keeping 
the deposit for some time, living forms of low type, 
such as were found in the lower depth.? of the 
water, made their appearance. It is worthy of 
note that the sample of water brought from the 
Wakoya stream, the main feeder of the reservoir, 
absorbed more oxygen from potassium perman- 
ganate than any of the five samples drawn from 
the reservoir. The sample brought was too 
small to investigate the cause of this ; but it 
was probably <liie to the iron Wing originally 
present in the state of a protosali, and pahsing 
by absorption of oxygen to the state of a 
persalt, whereas the iron in the water of the 
reservoir had already been exposed for a uiucli 
longer time to the oxidising influences of the 
atmosphere. It was quite otherwise with the 
small sample of water brought from the natural 
,=-j>ring in tlie scouring culvert. This contained a 
minimum of albuminoid matter, ami also absorb 
ed the lea>t amount of oxygen of any of the 
samples. If we may assume that this is the 
same w ater as that in the reservoir after lia^ ing 
been subjected to a process of natural filtration, 
which has removed the organic matter and the 
iron, it gives a fair idea of what Labugama water 
would Ik; like after pa**8ing through lilteriug hadf. 
During the liay time the temperature of the 
water is highest at the surface, and deereaset* 
towards the bottom of the reservoir : there cannot 
therefore be any convection cunents tending 
to bring the impure water from below to the 
surface Durin^ the night any convection 
currents caused l>y the cooling of the surface 
would only extend to a short depth, but these, 
together with the wind, must ensure the aera- 
tion of a certain stratum of water I am of 
opinion that the inferiority of the lower to the 
surface water is not due to impurity rising from 
the bottom, but to the want of aeration of the 
deeper water an<l decomposition of the organic 
matter in the water itself. Tht increase in the 
amount of free or saline ammonia must be due to 
the descent of dead and decaying organic remains 
continually going on, the ammonia set free by the 
process of decay pas.sing into solution. The 
deejwst water having been longest exposed to 
this process necessarily contains maijt anuuonia. 
The following is the analysis of a sample 
of water from the Kandy public water supply 
(December. 1891), and, in a ])aralled column for 
comparison, I put the analysis of the C"oloml>o 
public water supply for 1891 :— 
A nalysk of Kandy and Colombo Water Supplitji. 
Kandy. 
Colombo. 
Grains 
Grains 
per gallon. 
per gallon 
Free and saline auimonia.. 
•0014 
-0010 
Albuminoid ammonii 
•0035 
-0050 
Total solids 
3-6400 
1-8700 
Solids volatile on ignition 
-8400 
•9380 
Fixed mineral matter 
2-8000 
■9320 
Chlorine... 
•3685 
■1561 
Nitrous acid 
absent 
absent 
Nitric acid 
■1777 
-0114 
Iron 
trace 
•0340 
Oxygen absorbed from po- 
tassium permanganate, in 
presence of acid, in four 
hours, at the ordinary 
temperature 
-0080 
-0263 
Total hardness ... 
1-7000 
•5000 
Color 
Less than 
2-2° yellow 
-5° red 
Labugama 
water 
This sample of the Kandy water supply con- 
tained less albuminoid matter and less iron ; it 
also absorbed less oxygen than the Colombo 
water supply ; on the otiher hand the lower pro- 
portion of chlorine and nitric acid in the Colombo 
water indicates a gathering ground freer from 
animal impurities. 
