[VoL. 3 
420 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
for analysis on such media were almost insuperable. The de- 
sirability of growing the Basidiomycetes on solid media was 
obvious, but under such circumstances it would have been 
necessary for greatest accuracy either to analyze the entire 
contents of such bulky cultures or to take account of change 
in weight of materials due to loss of СО» and H20 produced 
in respiration. 
Lohnis (210) seems to report having found N-fixation in 
Torula, but his work was neither extensive nor reported in 
such way as to give the details of the methods employed. 
Pennington ('11) worked with two species of Penicillium, 
two species of Fusarium, Aspergillus niger, and Alternaria 
sp. A variety of experiments was arranged in liter flasks 
containing 100 ec. of solution. Employing aecepted methods 
he obtained no nitrogen fixation in a first series of experi- 
ments, although there was some growth on media practically 
without nitrogen, that is, with scarcely perceptible amounts, 
due, he believes, to impurities in the dextrose employed. In 
another extensive series no differences between the flasks con- 
taining the moulds and the controls were sufficient to indicate 
fixation. In one ease Penicillium gave an apparent fixation 
of .88 mg. In considerable part his work was planned to con- 
firm or disprove that of Latham (209). The results are 
therefore peculiarly interesting, especially as he has ap- 
parently observed great care to eliminate all possibilities of 
error, and in advance thoroughly tested his ability to deter- 
mine the nitrogen content of the eultures accurately. 
Medisch (710) observed some growth of the fungus Hypo- 
crea rufa in solutions to which no nitrogen was added. He 
reports a gain of 1.05-2.45 mg. nitrogen in 50 ce. distilled 
water. This preliminary experiment was followed by others 
in which, under purified air, the organism was grown on var- 
ious culture solutions. "These included solutions containing no 
nitrogen, nitrogen as ‘‘potassium humate,’’ and as NH4NO3. 
The results indicate that, whereas in the first case, with nitro- 
gen present only as an impurity, the fixation was 1.74—3.23 
mg. in 100 cc., in the humate solution the gain was 3.15-4.61 
mg., and in the solution containing NH4NOs Ше gain was 
