BY R. GREIG-SMITH. 



151 



found to give the boiling- and dilution-reactions. The original 

 mixture of bacteria was plated out in levulose ammonium-sulphate 

 agar, and the flora was seen to consist of five organisms named 

 provisionally 8. B.1-5. These were tested with the same hay- and 

 dextrose solution, but as they showed no sign of toxin-formation, 

 it is unnecessary to give the details. The original experiment 

 was repeated, but the result was quite different from the first. 

 Experiment x. 



The differences between these tests is marked, and, while there 

 probably was a difference in the kinds of bacteria added in the 

 two cases, there is the suggestion that the time of incubation 

 may have an influence upon the result.* In the 12-days' test, the 

 toxin may have been formed on or about the 7th day, and have 



'■■" The experiment was repeated some months afterwards, with the fol- 

 lowing results : — 



The previous experiment was not confirmed, and there is no evidence of 

 toxicity shown by the dilution-curves. 



