61 



Several times it was ol)served that Ihe supernatant fluid 

 from the tube containing extract and adsorbent at the neutral 

 point, showed more activity than tlie original fresh concen- 

 trated extract. This may either be due to a relate v^ely ex- 

 tensive adsorption of growth-inhibiting substances or to the 

 dilution of the extract, (ienerally it was observed that the 

 supernatant fluid was less active than the original native 

 solution on either side of the neutral point. 



By determining the nitrogen in the supernatant fluid, it was 

 observed, that from 50 to 60 percent of the total nitrogen of 

 the extract was adsorbed — furthermore it was found, that 

 about 30 per cent of the adsorbed nitrogenous substances could 

 be dissociated from the adsorbent. This was rather constant, 

 no matter which method was used for the dissociation. The 

 growth promoting power of the displaced substances wasi 

 very different in the different methods. The best way in get- 

 ting a good dissociation of growth promoting substances was 

 to add Ringer solution with 0,2 c. c. of a 1 per cent sodium 

 oleate and keep it for about two hours at 39 ' C. and 15 

 hours in the refrigerator. None of the dissociated solutions 

 showed approximately as activating a power as the original 

 extract. Absolute destruction of the activit}' of the extract 

 was the result of the method of dissociation at 53 ° C. for 

 two hours. 



The amount of nitrogenous substances which coidd be 

 dissociated was the largest when the adsorption had taken 

 place at Ph. 7,0 in spite of the fact, that the amount of 

 adsorbed was found to be the least at this reaction. 



The amino acids remained in the bulk and were not 

 adsorbed to any measurable extent. The determination of 

 the amino acids was made after the method of S o r e n- 

 sen by the formol titration. It is interesting enough to see 

 that the amino acids were not adsorbed in spite of the fact 

 that 50 per cent of the nitrogenous substances were adsorbed. 

 It is verj^ probable, that it is the largest molecules that are 

 responsible for the activity. The experiments of ultra fil- 



