292 BOTANICAL GAZETTE [OCTOBER 
in all the flasks. Connections were made by means of new rubber 
stoppers and rubber tubing, which had been boiled in a 10 per cent 
solution of NaOH until perfectly clean. In addition, rubber 
stopper connections were sealed with paraffin. There was every 
indication that all the connections were air tight. The medium 
used for these cultures was made up as follows: 
Conductivity water (ammonia-free)............... 1000.00 CC. 
NT a ee eh ki se ees wks ees 45.00 gm. 
Monopotassium phosphate...................00+ 3.00 gm. 
MN PETA oes ees ce 0. 20 gm. 
Se MUNN 0 sk sk eke os we es ew 0.10 gm 
I OTN io i eeicos iba bas bs cde ces 0.02 gm 
Of this medium 1oo cc. was supplied to each flask and then all 
apparatus except the U-tubes was sterilized in the autoclave at 
115°C. Open ends were plugged with cotton during the steriliza- 
tion and then connections were made as quickly as possible after 
removing these. Inoculations were made with the usual precau- 
tions, giving one culture for each of the four forms: Hormodendron, 
Myceliophthora, Fusarium, and Pachybasium. 'Two controls were 
prepared, one inoculated with M@ yceliophthora and the other wi 
Hormodendron, and both sterilized immediately after inoculation. 
The latter of these controls became contaminated during the culture 
period and is therefore not considered in the analytical results. 
These cultures were allowed to develop from February 24 to May 4. 
A perceptible growth occurred in all except the controls and 
examination, including microscopic, showed the cultures to be 
pure. The growth, while perceptible, was very scant. To the eye 
it appeared a little less than in those of investigation II, which 
stood in laboratory air. At the end of the development period 
previously mentioned, the cultures were disconnected and all 
Sterilized in the autoclave. As soon as possible analyses were 
made as before, for mycelium and filtrate separately. The one 
uncontaminated control was also analyzed. The results are show? 
in table VIII. 
From table VIII it is seen that no indication whatever is shown 
of nitrogen-fixation. The values for all the cultures are well within 
the limit of error of the analytical method. 

