20 
Journal of Agricultural Research 
Vol. XXV, No. I 
manifestly unjustifiable to expect rigid control of atmospheric conditions 
by this installation. It was anticipated, however, that difference of evap¬ 
oration could be restricted while maintaining a fairly constant tempera¬ 
ture difference between the two culture chambers. It was evident that 
the plane of temperature in both chambers must be allowed to rise 
gradually as the season advanced into spring and the temperature of the 
lake water increased. 
culture chambers 
The culture chambers were constructed from cypress 3 cm. thick in 
the form of four-paned window sashes as to sides and top. The sashes 
were set with glass of single thickness on both sides, thus providing heat 
insulation by an air space 1 cm. deep. The effective size of panes was 
36.3 cm. square. Extra width of the bottom bars of the side sashes pro¬ 
vided for isolation of a subchamber 10 cm. deep. Both floor and ceiling 
of this compartment were of 2 cm. pine boarding mortised to the sash 
bases. The culture chamber proper thus took the form of a cube with 
inside depth of 85.5 cm. It was painted glossy white throughout to 
promote reflection of light and heat. 
A circular rotating table was provided in each chamber so as to facilitate 
uniform exposure of all cultures to varying degrees of temperature, 
humidity, and illumination. This was borne by a circular base of cast 
iron, 31.3 cm. in diameter, and projected upward in the form of a trun¬ 
cated cone. The base rested upon the bottom of the subchamber and 
its conical projection bore upon a ball bearing a cylindrical steel post 
2 cm. in diameter and 10 cm. high. The latter was thus protruded 
through a central hole at the bottom of the culture chamber sufficiently 
to bear the rotating table. To it was fixed in the subchamber a wooden 
sheave 21.5 cm. in diameter and 2.5 cm. thick, by means of a central iron 
ring with set screw. The sheave was grooved for a small belt which 
passed through holes in the chamber wall to a reducing gear and motor 
outside. The table was composed of three pieces of cypress 2 cm. thick 
glued together to form a circular piece 75 cm. in diameter. It rested 
upon the supporting post through a centrally placed iron socket. Despite 
thorough painting it warped badly after some time. To avoid warping 
under these trying conditions, metal has been substituted for the wood. 
One side of the chamber was supported upon hinges to allow access 
to both table and sheave. The conditioned air was conveyed from the 
heating cylinder outside to an opening beneath the edge of the rotating 
table by an extension of the sheet-iron cylinder projected diagonally 
upward on one side of the subchamber and converted to oval form as it 
opened into the culture chamber. To facilitate even distribution of the 
air upward, it w'as reflected beneath the rotating table by an arc of gal- 
vanized-iron sheeting erected close to the rim of the table from the base 
of the chamber. In addition, iron flanges were suspended radially at 
intervals from the bottom of the table. Air escaped from the chamber 
by a series of holes through the upper bars of the window sashes, whose 
total area was about 70 per cent greater than that of the intake. 
SOIL CULTURES 
On February 25, 1920, buckwheat seeds selected for uniformity of size 
from a Japanese variety w'ere planted in 3.5 kgm. portions of Miami silt 
loam in earthenware jars (one-half gallon crocks). The soil was watered 
to 25 per cent of saturation, covered with paraffined paper, and placed 
