99 
glucose lactose broth, the older is the pollution. Further, it 
is astonishing how really sensitive for surface waters this test is, 
and what a comparatively small volume of recent pollution 
brings the acid and gas line in the two media together. The 
standing example of this is the condition of the water 
at the Howrah Bridge. We have already stated that the pollu- 
tion is relatively small in amount, though gross in character, 
yet divergence in the acid and gas line is rare at any time of 
year. Of course it is not maintained that simply finding the 
acid and gas line in the two broths will tell the analyst every- 
thing he wants to know about a sample, but it will undoubted- 
ly give much valuable information. 
The next question we have to decide is :—What is the 
best laboratory method of estimating the relative | numbers 
of the two classes of or ganisms ? 
In the course of our investigations we have always observ- 
ed that, during the summer months, even when using a lactose 
broth and lactose solid medium we get a great preponderance 
of white colonies on the agar. (Investigation shows that 95 
per cent. of all these organisms are bacillus P.) But if a glu- 
cose broth is used for the first step, practically no red colonies 
can be isolated at this time. This being so it was thought that 
it might be possible to get a rough indication of the relative 
number of these two groups of organisms (lactose + glucose +, 
and lactose — glucose+) by counting the white and red 
colonies that grow on the agar medium. A large number of 
figures have been compiled during months of research and we 
find that this method is of very little value for the following 
reasons :— 
It was discovered that the relation between the number 
of white and red colonies in any given sample depended al- 
most entirely upon which tube of the series was used for making 
the plates. Thus with asample of a well stored water, suppos- 
ing in the lactose broth series the 20, the 10, the 5, the 1 and 
the ‘1 c.c. had all given acid and gas, if plates were made from 
the 20 c.c. red colonies would predominate, probably a few 
white colonies would be obtained; if from the I or ‘I c.c. 
(the last one of the series which gave acid and gas) practi- 
cally no red colonies at all would be found on the plate; 
