588 
MICROSCOPICAL EXAMINATION OF WATER. 
While making the experiments with urine and phosphates 
I also made many with sewage and effluent sewage water 
under various conditions. There happens to be near me a 
farm on which the sewage of Epsom is utilised, and I obtained 
various samples of the sewage as it entered the farm, and of 
the effluent sewage water, and also samples of the latter after 
it had become mixed with the water of a brook. 
In dealing with such samples as these in their normal con- 
dition, and examining them microscopically from time to time, 
the first thing that strikes an observer is their capability of 
producing a great variety of life without the addition of sugar 
or any extraneous substance, and when, after various degrees 
of dilution, the samples are submitted to the sugar test, they 
become less or more turbid, and contain variable proportions 
of fine filaments and bacterian bodies. 
In effluent sewage water, I have rarely met with the fila- 
ments in that beautiful network form in which they are found 
in water containing only urine or a phosphate; but there is 
no difficulty in deciding that both filaments belong to the 
same class of growth. 
The filaments met with are of at least two sizes ; one is so 
fine that the joints cannot be seen when magnified 1000 
diameters. The joints in the other filament can be observed 
when magnified above 400 diameters. Both descriptions of 
filaments are found in water containing sewage, but in some 
diluted samples of effluent sewage water I have only been 
able to discover the finer filaments, while in others only the 
larger. In mixtures of urine or of a minute quantity of 
alkaline phosphate with New River water, the larger filaments 
are found, and they occur in the network form. 
Both filaments soon break into pieces of various lengths 
and diffuse themselves throughout the liquid. In the case of 
the filament produced in the water containing urine, I 
observed that in its earlier stages of development the filament 
appeared to be surrounded by a sort of soft casing, which 
gradually separated from the filament, the jointed character of 
which was then readily distinguishable. 
While dealing with the subject, it is proper to point out 
that when some descriptions of ditch and other dirty water 
are submitted to the test, filaments similar in character are 
developed, and after these break up and become diffused 
throughout the liquid, they are hardly distinguishable from 
those produced in water containing sewage ; in fact the only 
difference apparently observable is that the broken filaments 
are much less numerous than in samples of effluent sewage 
waters, which undoubtedly produce the filamentous growths 
