256 ANIMAL AGGREGATIONS 



moistened, and inoculated with B. suhtilis. There was no growth 

 when this was incubated. 



A control tube of gentian-violet agar was inoculated with this 

 organism. No growth occurred. Another similar tube was covered 

 with a thin layer of the killed and washed organisms just described. 

 After this layer had dried, an inoculation of living B. suhtilis was 

 made on top of it and growth occurred. Evidently the shield of 

 dead, washed organisms, without sufficient nutritive value to sup- 

 port growth in themselves, served as some sort of a barrier between 

 the poisonous dye and the living cells. Similar results were ob- 

 tained if dead bodies of Micrococcus aureus or of Bacillus coli were 

 used in place of dead B. suhtilis, so that the protection is not species- 

 specific. These results are suggested graphically in Figure 23. 



These experiments with agar media indicate that there is a real 

 difference between the behavior of a single cell and that of a group 

 of cells in the presence of gentian violet. This was tested further 

 by experiments upon the survival and growth of B. coli in a gentian- 

 violet broth at a dilution of i : 100,000. The strain used came from 

 one of the two colonies that appeared from single-cell transplants 

 onto gentian-violet agar. Not only did these cells come from a 

 strain known from several weeks culturing to be gentian-violet 

 tolerant, but they all came from a single organism of this strain 

 which had successfully produced a colony when isolated onto 

 gentian-violet agar. 



When single organisms of this strain were cultured into plain 

 broth, 80 per cent growth was obtained. When 30 or more cells were 

 cultured together in gentian-violet broth, almost 100 per cent 

 growth was obtained. The first 140 single-cell inoculations into 

 gentian-violet broth yielded no growth. In a final series of 8 inocu- 

 lations, delayed growth was obtained in one tube in the second 24 

 hours. The chances are at least 147 in 148 that growth will not 

 occur under these conditions. These results are to be compared 

 with those showing that there are almost 100 chances in 100 of 

 inocula of 30 such cells growing under similar conditions. 



It may be that 30 cells succeed in growing because these 30 cells 

 are able to produce some anti-dye substance in an amount sufficient 

 to destroy the harmful effect of the dye when a single cell is unable 



