IO 
Macgregor Skene. 
be thought that they are autotrophic. When I used one of these 
flask cultures for infecting Lieske’s solution, I obtained a good 
growth in the second generation : the third generation was equally 
satisfactory. That such a strong development could take place at 
the expense of the small amounts of organic material carried over 
with the infection, or of dust particles from the air, I do not think 
possible. Thus it would seem logical to infer an assimilation of 
carbon dioxide in the culture flasks. But as we are dealing 
with mixed cultures it does not by any means follow that the 
assimilation is carried out by the purple sulphur bacteria. In the 
present case the matter is complicated by the fact, that in my 
cultures an undoubted autotrophic bacillus is always present. 
After one or two weeks there constantly appears on the surface of 
the liquid a thick whitish or yellow skin, which consists of sulphur, 
the oxidation product of a minute bacillus oxidising hydrogen 
sulphide. This bacillus (which is mentioned by Lieske, 1912 a) is 
easily isolated by means of silica jelly plates. It grows excellently 
in the solution employed for the purple bacteria, in presence of 
hydrogen sulphide, and produces a characteristic deposit on the 
surface of the liquid. 
It is at all events a possibility that the carbon assimilation 
takes place solely through this organism, and that the purple 
forms live at its expense. A symbiosis in the strict sense of the 
word does not come into consideration: in the first place because 
the bacillus and the purple bacteria inhabit, in the main, different 
layers of the culture liquid : and in the second place because the 
bacillus does not require the others, and can live without them. Some 
looser connection, to the advantage of the purple forms, might, 
however, be quite possible. The solution of the problem can again 
be obtained only with pure cultures. Without these it was 
possible to try by comparative cultures, to find out in which 
direction the answer to the question lies. 
Eight flasks were infected with a pure culture of the bacillus, 
and allowed to develope for 12 days: at the end of that time when 
all showed a vigorous development, four were sterilised in the 
steamer, and four fresh uninfected flasks were added. Of each of 
these three series three flasks were infected with purple sulphur 
bacteria, and one was left as a control. Growth began in all the 
infected flasks about the same time. After three weeks the 
development was estimated : the control flasks showed no growth : 
the others were all very nearly equal, such small differences as 
