5 
Physiology of the Purple Sulphur Bacteria. 
colour. With such cultures one can observe the more rapid growth 
at the side of the vessel which is turned to the light, but not much 
more. It was therefore necessary to find some method, allowing 
of culture under conditions which could be controlled more exactly. 
In the first place I tried drop cultures such as Winogradsky 
used : these have unfortunately many disadvantages. A great deal 
of time is occupied by their examination : each one must be “ fed” 
with sulphuretted hydrogen water at least once every day: and 
when all is done it is impossible to control exactly the supply of 
food substance. Such cultures have scarcely given useful results. 
I then tried cultures in conical flasks. A few bubbles of 
hydrogen sulphide were passed into each flask every day. This 
supply is however too uncertain, and too difficult to regulate. But 
with a modification of the method I attained my end, and was able 
to cultivate the bacteria under approximately known conditions. 
I placed the conical flasks under a glass bell, which was fixed 
with wax and lard to a glass plate: the tubulus of the bell was 
provided with a double-bored stopper, through which passed two 
glass tubes: these could be closed by rubber tubing and glass 
stoppers. The tubes served for the admission of the necessary 
sulphuretted hydrogen. If such a bell of 3 litres content be 
provided with 50 to 150 ccs. of sulphuretted hydrogen (the measur¬ 
ing out of the gas was performed by a suitable piece of apparatus, 
into the details of which it is scarcely necessary to go) it takes 
several days before oxidation is complete. In general 50 to 100 ccs. 
were introduced into the bell to start with, and on the following 
days the contents tested by means of a small hand bellows, and 
lead acetate paper. When the reaction for sulphuretted hydrogen 
was given only very faintly, or not at all, then the gas supply was 
renewed. This renewal took place as a rule every second or third 
day. The method appears at first sight very rough, but in practice 
it gives excellent results. The bacteria developed as surely, and as 
rapidly as under the most favourable natural conditions. 1 
As regards the species which have been investigated it must 
be remarked that in the original infection material (from Kiel) 
many of the colonial forms described by Winogradsky were 
observed: the motile forms were all swarming conditions of these 
—typical Chronuitiuin was not seen. In my raw cultures 
Laniprocystis roseo-persicina, formed the chief constituent of the 
1 The first culture I made with this method is dated 13th July, 1912—that 
is about six months before the description of a similar method for Beggiatoa, 
bv Keil (1912). 
