124 



crococcus from the air, a chance contamination, which grows 

 in pure white colonies. The same strain has been continually 

 used in the agar inoculations. For the sake of brevity it may 

 be referred to as „air coccus". 



That the symbiosis phenomenon may be indefinite or even 

 entirely absent, when the organism' which is to evoke it, is 

 brought into too intimate contact with Pfeiffer's bacillus, is 

 proved by a big series of experiments that I carried out in 

 19l8 to investigate the growth-promoting action of a large 

 number of different species of bacteria on some strains of 

 Pfeiffer's bacillus. The latter was inoculated in streaks and 

 the other bacteria were inoculated at certain points i n them'. 

 It was then found that the expected symbiosis effect was 

 very weak or quite absent in many cases. 



Fig. 8, plate I shows Pfeiffer's bacillus inoculated partly as a 

 streak and partly as a four-sided area. The air coccus was also 

 inoculated al 4 points outside the cultures and at 2 points in it. 

 It will be seen how the colonies situated outside the cultures of 

 Pfeiffer's bacillus produce greatly increased growth in the nearest 

 portions of these, while the colonies inoculated actually in the 

 cultures are quite inactive in this respect. It may be mentioned 

 that a corresponding experiment where a Streptococcus took the 

 place of the air coccus, gave precisely the same result, in spite 

 of the fact that the Streptococcus colonies in the cultures of Pfeif- 

 fer's bacillus were larger than the ones outside. 



In figs. 2—7, portions of 6 agar plates containing a small amount 

 of dissolved haemoglobin will be seen. The plates were inoculated 

 with Pfeiffer's bacillus, the two first so thickly that the individual 

 colonies could not be identified with the naked eye, the following 

 ones more or less scantily. The air coccus was also inoculated at 

 a point on the plate. Where the symbiosis effect did not come 

 into action, the growth of Pfeiffer's bacillus was always slight. The 

 irregular growth of the air coccus in fig. 3 is due to the presence 

 of condensation water on the plate. It will be seen that neither 

 the limited growth of the air coccus in fig. 2, nor the extensive 

 growth in fig. 3 were able to produce the slightest increase in 

 the growth of Pfeiffer's bacillus in their vicinity. On the next 

 plate (fig. 4) where the individual colonies could be distinguished 

 with the naked eye a marked increase in the growth of Pfeiffer's 

 bacillus took place, but only in quite a small zone around the 

 air coccus. On the succeeding plates the extent of the effect in- 

 creases obviously in proportion to the more and more scanty ino- 

 culation. 



Lastly, it will be observed in figs. 9 & 10, how the innermost 



