[Vor. 1 
172 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
equally elearly to the conclusion that the remaining forms 
experimented with do not belong to this class. 
Oes (25) made the observation that Azolla with its endophytie 
Anabena Azolle grew exceedingly well in mineral nutrient solu- 
tion free from combined nitrogen. Analysis showed a distinct 
fixation of nitrogen. Attempts to cultivate the Anabena in 
pure culture failed. While calling attention to the possible 
direct róle of the associated bacteria in the observed fixation, 
the author inclines to the view that Anabena Azolle is itself 
capable of fixing free atmospherie nitrogen. 
The preceding survey of literature shows that in all of the 
earlier investigations, and in a considerable number of the later 
ones, impure cultures were used. In experiments conducted 
under these conditions, it is evident that negative results are, in 
general, more reliable than positive ones. Attention should 
therefore be called to the negative results which have been ob- 
tained from investigations with impure cultures. These, as 
will be seen from the literature cited, include a large number of 
genera and species from both grass-green and blue-green alge, 
and indicate in many cases with a reasonable degree of cer- 
tainty that the faculty of elementary-nitrogen fixation is absent 
in а very considerable number of species of both the Chloro- 
phycee and Cyanophycee. In the former class, all investiga- 
tions conducted with pure cultures have led without exception 
to the conclusion that these forms are unable to fix free atmos- 
pherie nitrogen. 
As regards the Cyanophycee, it should be stated at the outset 
that while many observations are on record both affirming and 
denying free-nitrogen fixation in the group, it is questionable 
whether experiments have been conducted with more than a 
single species in pure culture. Bouilhac, it is true, claims to 
have isolated Schizothrix lardacea and Nostoc punctiforme in 
pure culture. From the meager account given of the isolation 
technique, it appears very improbable that the latter form was 
actually obtained in culture free from bacteria, although the 
former may have been. However, the work of Heinze, while 
not conducted with pure cultures, renders free-nitrogen fixation 
in Nostoc probable, and it appears especially desirable, there- 
