124 Marine Microbiology 



a pH of 7.8, was autoclaved. The loss of some of the heat-labile 

 substances apparently did not disturb diatom growth. 



Soon it became clear that the bacteria, while presumably 

 exercising a beneficial action, were also responsible for cessation 

 of growth and death of the diatoms. When bacterized cultiu'es 

 of Chaetoceros dydimus were transferred in fresh medium, the 

 diatom developed abundantly, but when growth was completed, 

 the bacteria multiplied rapidly, and destroyed the diatom cells. 



Several bacteria were isolated from cultures of Chaetoceros 

 dydimus, Asterionella japonica and Coscinodiscus. They were 

 all Gram-negative rods, strongly saccharolytic, tentatively inden- 

 tilied as belonging to the family Fseudomonadaceae and Achro- 

 mohacteraceae. Since these bacteria could grow in the synthetic 

 medium to which 0.1 per cent glucose was added, it was con- 

 cluded that the bacteria were thriving, very likely, at the expense 

 of the sugars produced by the diatoms during photosynthesis. 

 The diatom medium without added glucose and without diatoms 

 did not support bacterial growth, when inoculated with the 

 bacteria. 



Antibiotics were used at the beginning in the attempt to 

 eliminate the bacteria. To select the most effective antibiotics, 

 the bacteria, plated on sea water peptone agar, were tested 

 with Difco antibiotic sensitivity disks (Table 2). 



Polymyxin B and Neomycin were selected as being the most 

 effective and the least toxic against the diatoms. Medium with 

 fifty /igm/ml of both antibiotics was effective in keeping the bac- 

 terial count down. However, at this concentration, resistant mu- 

 tants developed, which made impossible their eHmination with- 

 out damage to the diatoms. 



By using an alcoholic solution of iodine (2-3 drops of 0.1 

 per cent solution in 10 ml of diatom suspension for 1-2 minutes) 

 and by subsequent centrifuging and washing in sterile medium, 

 it was finally possible to eliminate the bacteria and grow the 

 diatom Chaetoceros dydimus in pure culture. 



The exposure time seemed to be very critical: an exposure to 

 the iodine solution of five minutes kills the diatoms by irreversibly 

 bleaching the cells. 



Sterility of diatom cultin-es was tested by inoculating sea 



