46 
(3) While the lactose fermenting organisms go on decreas- 
ing in number, the results show that the class glucose + lac- 
tose — remain stationary, or increase. 
(4) That on the whole fecal organisms of all kinds are 
more plentiful at the bottom than on the surface. 
II. On account of the objections that could be raised 
to the above series of experiments, we attempted to prove 
the same point on a large natural sheet of water which only 
required a heavy downpour of rain in order to pollute it. The 
place chosen was the Red Hills Lake in Madras which 1s about 
14 miles in circumference. For reasons which need not be 
given in detail here, a holiday of a few weeks’ duration was all 
the time that could be given to this work, and during the 
period only one shower fell. But the results are fairly satisfac- 
tory. This experiment will be referred to later on, as other 
points of very great interest are also demonstrated. The fol- 
lowing points should be observed in Table VI(/). 
(i) Note the analysis before the fall of rain ; the great in- 
crease in the number of organisms of all kinds due to the rain ; 
the coincidence of the acid and gas line after pollution. 
(ii) Observe the very rapid fall in number of bacilli in 
all samples for about 4 or 5 days and the purity of the surface 
sample in No. 7. 
(iii) From about this point the glucose + lactose —start 
to increase, and do so fairly steadily throughout the experi- 
ment. 
(iv) Note also ashort similar increase in the glucose 4 lac- 
tose +, this will be discussed fully later on ; it is due to one 
variety of the class growing in the water at certain times. 
Besides these points it must also be stated that an analy- 
sis on the 20th showed nothing but bacillus P ; on the 21st 
and 22nd other non-lactose fermenters were mixed with these ; 
but from 23rd onwards only bacillus P could be isolated. 
Ill. A third series of investigations have been carried 
out on great rivers in Bengal during the rains and dry weather, 
in order to observe what takes place during the process of clear- 
ing when the rains are over. A series of very careful weekly, or 
fortnightly, analyses have been taken of the river Ganges at 
various points. It is not an easy matter to demonstrate by 
