874 Mockeridge . — Some Conditions influencing the Fixation of 
and even 5 per cent., whilst the cultures have been allowed to grow for 
widely divergent periods of time before analysis has been made of their con- 
tents. The results reported by different workers as to the amounts of 
nitrogen fixed appear to be somewhat conflicting, but they are not strictly 
comparable one with another, since Gerlach and Vogel have shown that the 
activity of the organism varies with the concentration of the carbohydrate, 
and with the age of the culture, which latter statement is borne out by the 
results obtained in the course of the present work. 
Some attempt having been made to solve the foregoing problems, some 
consideration was given to the question of aeration. Considerable attention 
has been directed by various investigators to the discovery of some 
means of improving the aeration of organisms grown in liquid cultures. 
Freudenreich more particularly advocates the method of using gypsum 
plates, placed in Petri dishes and soaked in nutrient solution, and Krainski 
reports that he obtained the characteristic browning in from two to four 
days by the adoption of this method ; while strips of gypsum placed in 
test-tubes containing some of the solution are favoured by other workers. 
For solid media, Koch recommends the use of large Petri dishes containing 
a thin layer of mannite-agar, and this method has been universally adopted 
in cases where it is suited to the special requirements of the occasion. 
Whatever be the vessels used for the production of liquid cultures, the use 
of a layer of nutrient solution as thin as is consistent with convenience 
is always recommended ; and some workers, notably Hoffmann and 
Hammer, advocate the use of white quartz sand, which is shaken to one 
side of the flask, and forms a slope which rises well above the surface of the 
liquid. The use of these sand slopes proved to be by far the most efficient 
and convenient method of improving the aeration of the organisms in the 
course of the present work, for a suspension of the Bacteria could readily be 
obtained by adding water, well shaking the vessel, and allowing the sand to 
subside ; while when required for analysis, the whole of the contents of the 
flask could most easily be transferred to a Kjeldahl flask by means of 
a jet of distilled water from a wash-bottle. Accordingly some attempt 
was made to obtain an idea of the extent of the superiority of this method 
over that of the ordinary liquid culture, as far as the fixation of nitrogen 
per unit of carbohydrate consumed was concerned. 
In order to ascertain the best medium, as regards the inorganic 
constituents, for the growth of Azotobacter , a pure culture of the organism 
was first obtained on a mannite-agar plate. A colony was transferred by 
means of a sterilized platinum wire to about 30 c.c. of sterile water contained 
in a large test-tube, which was then well shaken, to ensure a uniform 
distribution of the Bacteria. One cubic-centimetre of this suspension was 
transferred by means of a sterile pipette to each of the flasks to be used for 
the experiment, for the purposes of which three nutrient solutions were 
