486 
Fig. 1 
Proceedings of the Royal Society 
difference the weight of the drop. The nutrient gelatin is melted 
at a gentle heat in a Lister flask, and sufficient poured into the cell 
so as to mix with the drop of water and to form a layer of one- 
tenth of an inch deep. The glass cell thus pre- 
pared is placed in a special glass shade or cham- 
ber, the air of which is kept saturated with 
moisture. After from three to five days, the 
colonies developed are counted by the aid of a 
lens. To facilitate the enumeration, the plates 
are all ruled by a writing diamond into squares. 
The collection of ordinary samples was by 
means of small 2-oz. phials, washed out with 
boiling water, and then heated for many hours 
in an oven to about 250° C. 
In order to obtain samples of water at differ- 
ent depths, a very efficient and simple apparatus 
was invented for me by an ingenious friend.* 
To the disc of wood a (see fig. 2), weighted with 
lead, is fixed a frame of brass g ; in each side of 
the frame are slots, in which a clip x slides freely. A small stop- 
pered bottle is firmly attached to the disc, and the clip secured to 
the stopper ; the clip, and with it the stopper, is firmly held down 
by means of two pieces of elastic at s and s' ; the apparatus is let 
down to any required depth by means of a line attached to the 
stout copper wire w\ to take deep sea samples the line is best 
marked into fathoms like a deep sea line. The stopper is opened 
at any time by pulling on a string attached to the ring r ; on releas- 
ing this string the stopper is immediately pulled into its place by 
the elastics. 
Sewage . — I found it practically impossible to make a quantitative 
determination of the micro-organisms in sewage unless the liquid 
was very much diluted. A convenient dilution is 1 c.c. of sewage to 
a litre of sterilised water. Drops of this dilute sewage, weighing from 
40 to 50 milligrams, and added to nutrient gelatin, yield, after a few 
days’ cultivation, from 100 to 200 colonies — numbers the counting 
of which present no difficulty, and a simple calculation gives the 
* Mr Stafford L. Archer, whom I have to thank for the trouble he took in 
perfecting the mechanism. 
