23 
In Series No. VII human feces were plated direct on to 
solid medium. Out of 707 colonies from human feces on 
lactose solid medium none belonged to the class glucose+ 
lactose —. 
All these previous results have been obtained on_ bile 
salt neutral red lactose agar. In order to check the accuracy 
of these it was thought desirable to make a few estimations 
with medium of another nature. Consequently Conradi 
Drigalski medium was tried. | 
Series, VIII.—g separate samples of human feces were 
wiped direct on to these plates. The total number of colonies 
on all plates was not estimated, but whenever a plate contain- 
ing discrete colonies was obtained, every colony was put 
through glucose and lactose broth. In this way 159 colonies 
were obtained; out of these 141 or about 88% fermented 
glucose and lactose and 18 or 12% belong to the class glucose + 
lactose —. It should, however, be stated that 16 out of these 
18 colonies all were obtained from one specimen of feces. 
This was evidently one of those rare occasions when these 
particular organisms were unusually numerous in feces. 
Series IX.—Exactly the same work was carried out with 
5 different samples of cowdung in which case 99 organisms 
fermented glucose and lactose and one was glucose+lactose—. 
Series Nos. VIII & IX were done in the autumn of IgII. 
It would, therefore, appear that this class of organism 
elucose +lactose,—is rare in the feces of human beings and 
animals ; and that investigations carried out over a consider- 
able number of years show that they seldom exceed 5 
per cent. of all organisms that grow in bile salt and very fre- 
quently they are present in smaller percentage. The results 
obtained show that rarely, one sample in eighty-four, they 
may be rather numerous, that is to say, as much.as 20% or 
30% of the total coliform organisms. In the course of our 
experiments we have found several instances when no mem- 
bers of this class can be isolated from a particular specimen, 
even when very large numbers of colonies were investigated. 
Some study has been made of the fermentative reactions 
of this group which give glucose+lactose—. We have used the 
same sugar reactions that we always use for lactose ferment- 
