87 
In the table showing these analyses there is one point, 
which will certainly come as a surprise to anybody perusing 
it carefully, namely, that in a fair proportion of the samples 
taken, the number of water organisms and fecal bacilli is 
greater in the middle layers than it is either on the surface or 
at the bottom. The number of instances in which this occurs 
is too great to be explained by the supposition that some 
error occurred either in taking or analysing the sample. These 
were taken, in every instance but one, by the same man, and 
from as nearly as possible the same spot in the lake, a con- 
siderable distance (600—700 yards) from the edge, all usual 
precautions being observed. Further examination of this 
point is necessary before it can be accepted as constant, or 
before an explanation is sought for. 
One point that is amply demonstrated in these results, 
is the fact that there is an enormous preponderance of bacil- 
lus cloace in all the samples taken from the bottom of the 
lake. A glance at the table shows this is beyond doubt. 
Hence it would appear, that this organism is more resistant 
than others to the influences which are at work in this situation. 
The rarity of bacillus coli communis on the whole is also 
remarkable. Only on two occasions was it found, and in 
only one was it present in any considerable quantity (vide 
samples 6 a, b, c, taken on December 7th). They were prob- 
ably brought into the lake by a downpour of rain, which 
occurred in the very early part of December. Samples 
taken from some of the inflow channels, about this time, 
show that bacillus coli communis was present in them. It 
should be noted that this organism was present in all the 
samples, surface, middle and bottom, consequently they must 
have been very numerous. By December 13th (that is to 
say, one week afterwards) no coli communis could be found 
in thelake at all. Thesurface samples showed no fecal organ- 
isms in 50c.c., the bottom samples as usual showed a great 
preponderance of cloace, and the middle the usual lactis 
gerogenes, with some bacillus Grunthal. 
The other instance when coli communis was found, was in 
the sample taken from the bottom on October 31st ; none were 
found when the next sample was taken on November 11th. 
