558 
AMERICAN JOURNAL OF BOTANY 
[V ol. io. 
At the point of outlet from the capillary tube the solution drips directly 
into the small funnel I upon which the tube rests, and passes to the bottom 
of the culture vessel through the glass tube provided for that purpose. It 
is possible with a close-fitting rubber connection to raise or lower this funnel 
at will, thus regulating the length of the siphon and consequently the rate 
of flow. In the non-aerated cultures of series III it was found necessary to 
use special care in preventing the inflowing solution from entraining air 
bubbles and carrying them to the bottom of the culture vessel, thus setting 
up in a small way an aerating system. This was prevented readily by the 
use of an inlet tube of larger diameter and by receiving the drip on the side 
of the funnel rather than permitting it to fall directly into the opening. 
The overflow from the culture jar was provided for by means of an auto¬ 
matic siphon suggested and used by Dr. Shive. This siphon, shown in the 
diagram of figure i, consists of a bent glass tube having a bore of about 2.0 
mm., with one arm of the tube about 1.5 cm. longer than the other. The 
siphon is held in position by the paraffined cork stopper which supports the 
seedlings. The long arm of the siphon extends just below the surface of the 
solution, which lies in a horizontal plane passing through the short arm at a 
point just above the outlet. When the system is in operation, the siphon 
having been started, the solution surface in the jar may be maintained 
automatically at any level desired by raising or lowering the siphon. The 
overflowing solution may be caught in a waste jar or may be disposed of in 
any convenient way. 
Continuous renewal of solution in the sand cultures was accomplished 
by allowing the solution to drip directly upon the sand in the center of the 
vessel. A watch glass was inverted just beneath the surface of the sand 
at this point for the purpose of spreading the solution as it percolated down¬ 
ward to the outlet tube through the bottom of the vessel. The rate of 
solution flow in the sand cultures, as in the solution cultures, was main¬ 
tained at approximately one liter per day per culture. 
Seeds were germinated in moist sphagnum moss, and carefully selected 
seedlings when about 5 cm. tall were transferred to the prepared culture 
media (solutions and sand) on August 16, 1922. The plants were grown 
during a period of 62 days and were harvested on October 17. 
Experimental Results 
Sand Cultures 
In table 1 are given the relative dry weights of the plants grown in the 
sand cultures. Each value represents the average of plants or plant parts 
of two like cultures and is based upon the corresponding dry-weight value 
of the simple sand cultures as 1.00. Examination of table 1 shows that aera¬ 
tion has no beneficial effect upon the growth of tops as compared with that 
of the plants in the simple sand cultures. It does, however, produce con- 
