[VoL. 3 
428 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
total nitrogen. To this was added the amount of nitrogen 
found in the felt. Lipman (711-7112) did not separate the mat 
from the medium, but transferred the whole to a digestion 
flask, and later to a distilling flask, determining the total nitro- 
gen in one lot. All the cases cited above involved one or more 
transfers of material, since the fungi were usually grown in 
Erlenmeyer flasks or similar receptacles, and the contents 
filtered or transferred before digestion, thence usually a sec- 
ond transfer to a distilling flask. It was with the end in view 
of eliminating the possibility of error in this direction that 
the method already described was employed, i. e., of growing, 
digesting, and distilling in the same flask and without transfer. 
Where the digestion of nitrates was involved in the culture 
solution, the previous investigators have used, almost with- 
out exception, the Gunning-Jodlbauer method—phenol or 
salicylic acid and zine dust being employed for the reduc- 
tion of nitrates. In our work the Forster modification was 
employed, since certain workers have found difficulty in ob- 
taining all the nitrate by the former method. Indeed, it was 
this difficulty which first led Förster to use sodium thio- 
sulphate as a reducing agent. If all nitrates are not reduced 
a serious error is, of course, involved, one which, moreover, 
makes for a difference between controls and inoculated flasks. 
The results may be presumably correct for the converted or 
assimilated nitrogen of the mycelium (or products excreted 
therefrom), low figures resulting for the nitrate of the cul- 
ture media. If an error were present, then, it would be 
related somewhat closely to total growth or to sugar consump- 
tion, factors determining nitrate consumption. It is equally 
true that the capacity to fix nitrogen by a fungus, if possible, 
might also be related to the capacity for growth under the 
particular conditions. 
In the use of the Forster method at first certain difficulties 
were experienced. In preliminary work the recovery of 
nitrogen from a water solution of KNOs was easily accom- 
plished within experimental error. When, however, a nitrate 
was added to a soil, compost, or plant tissue decoction the 
results were invariably low. It was found necessary to add 
