358 Transactions of the Boyal Society of South Africa. 
My attention was first drawn to this variation by the occurrence of a 
strain of B. coli (Bl), which, when tested shortly after isolation, produced 
gas from dulcite only, and when re-tested after a month's culture was found 
to have attained the power of producing gas from lactose and saccharose 
though still non-gas-producing in glucose. Later it also acquired the power 
of fermenting glucose with gas formation. 
Though the absence of gas production in the case of B. typhosus and 
B. dysenteriae is known to be a stable character of these organisms, the 
question arose as to whether " anaerogenes'' types of coliform bacilli simply 
represented variant strains of aerogenes types. 
While certain strains have been noted which immediately after isolation 
produced no gas from a/^?/ of the sugars fermented (e. g. BlOl), it was found 
that in many cases gas production was only absent in certain of the sugar tests 
(e. g. strain of Bl quoted above) ; also, as a general rule, such organisms 
after a few subcultures quickly developed the property of gas production, i. e. 
the character seemed to be only in abeyance. On the other hand, the various 
strains classified as anaerogenes'' (sub-group D), even after repeated 
subculture and after being kept in artificial growth for long periods, still 
remained non-gas-producing. 
It has been shown by Penfold that by growing B. coli on monochloracetic 
acid agar a variant strain could be selected out which differed from the 
original in the absence of gas formation in certain sugars. With a view to 
determining the possibility of transmuting an aerogenes type into an 
anaerogenes variety, certain B. coli strains were submitted to Penfold's 
procedure. 
Method. — The monochloracetic acid was made up in a 10 per cent, watery 
solution, and after having been made slightly alkaline to litmus by adding 
sodium carbonate, was sterilised by filtration through a Maassen filter. The 
solution was then incorporated in a 2 per cent, peptone- water- agar in 
measured proportions. The percentages indicated below are expressed in 
terms of the amount of the acid in the quantity of medium used for plating. 
In the first experiment a series of plates each of 10 c.c. of monochloracetic - 
acid-agar, the acid being in the following proportions : 
12 3 4 
0-05 per cent. . O'l per cent. . 0'5 per cent. . 1*0 per cent. 
were inoculated with atypical B. coli communis strain. On plate 1 a normal 
amount of growth was obtained but the colonies varied considerably in size. 
On plate 2 the difference in the size of the colonies was more marked and 
many of the larger colonies showed papillae as described by Penfold. On 
plates 3 and 4 no growth appeared. A subcultivation on ordinary agar 
was made from a large colony on plate 2, and from this plates containing 
the following concentrations of monochloracetic acid were inoculated as 
before : 
