398 
THOMPSON YATES LABORATORIES REPORT 
Milk. Although, as Durham 9 points out, this medium may not be wholly satis- 
factory, nevertheless it does form a ready means of differentiating the groups of 
bacilli under consideration. Lubarsch's 25 statement that B. enteritidis clots milk has 
been shown by Durham, 8 Flexner, 10 and others to be erroneous. Cushing's 5 careful 
study demonstrates the uniformity of alkali production in milk by all the members 
of the B. enteritidis group to which he had access. It was by means of milk cultures 
that Bacillus ' L ' was shown not to be a typical member of the B. enteritidis group, 
for while the latter all turned milk alkaline after two days, Bacillus ' L ' continued 
to acidify it. Thus : — 
Acid formed after one week at 37 0 C. = 2*25 per cent, normal KOH. 
„ „ „ four weeks „ = 375 „ „ „ 
Taurocholate lactose agar. This very useful differential medium synthesized by 
MacConkey 26 brought out a characteristic difference between the B. coli group and 
the other two groups studied, i.e., in the formation of a definite haze or halo about 
the separate colonies of the former. As shown by MacConkey, this is due to the 
fact that bile salts are precipitated by the acid formed from lactose, B. coli alone 
(within these three groups) having the power to ferment this sugar. 
Neutral-red media. As pointed out by Rothberger, 51 Scheffler, 34 - and others, 
B. coli rapidly decolourizes glucose agar containing Neutral-red, while B. typhosus 
produces no change. Wolff, 42 and later, Rambousek, 51 have claimed that the de- 
colourization is a process of reduction, the addition of two atoms of nascent H 
forming a ' leuco-product.' It seems likely, however, that other agents as well must 
be able to decolourize Neutral-red, for in agar containing lactose, B. enteritidis and 
Bacillus ' L ' (organisms that do not ferment lactose) the characteristic decolourization 
and fluorescence were observed (see Table VI). If this were due to the presence of 
a small amount of glucose, either originally present in the agar or formed by over- 
heating of the lactose, gas bubbles should have been seen. Possibly the process is 
similar to the regularly observed decolourization of litmus in a litmus-lactose-peptone 
solution by B. typhosus (which does not ferment lactose). 
Makgill 28 and Savage, 55 after using the medium for routine water examination, 
came to the conclusion that if the characteristic reaction were not obtained, B. coli 
was probably absent. Hunter 18 noted that B. enteritidis reacted like B. coli in 
Neutral-red, and, therefore, came to the conclusion that B. enteritidis was a modified 
form of B. coli, a view expressed also in Lehmann and Neumann's Atlas of 
Bacteriology. The assumption seems unwarranted, for the following reasons : — 
1. The fundamental requirements of a true B. coli, as laid down by 
Gilbert 13 in 1895, are, that in addition to its motility and production 
of indol, (a) it shall ferment lactose ; (b) it shall clot milk. No 
member of the B. enteritidis group satisfies these two conditions. 
