Mar. i 7> 1923 Influence of Soil Temperature on Seedling-Blight 
841 
ordinary telegraph relay. The temperatures below greenhouse tempera¬ 
ture were maintained by running a stream of cold water into the tank, 
lowering the temperature of the water slightly below that desired, and 
then heating the water to the temperature required by the thermostatic 
cally controlled electric heater. 
By means of this equipment it was possible to maintain fairly con¬ 
stant soil temperatures in the containers as well as constant air tempera¬ 
tures within the chambers consisting of the submerged cans. Repre¬ 
sentative soil-temperature records are compared with the air tempera¬ 
ture in the greenhouse in figure i. Soil-temperature readings, 2 inches 
below the surface, were recorded several times each day throughout the 
duration of all the experiments, and in many instances soil-thermograph 
records were obtained. The mean soil temperature for the duration of the 
experiments is given in each series of data presented. The Altmann 
incubator was operated at intervals ranging from about 4 0 to 30° C. 
The tanks were operated at regular intervals of.8°, 12 0 , 16 0 , 20°, 24°, 
28°, 32 0 , and 36° C., although the mean given in the table in connection 
therewith may deviate slightly from these temperatures. 
Fig. i.—G raphic record of greenhouse air temperature and of so il temperature in three different tanks 
used in the experiments. 
The approximate quantity of soil necessary to fill all the cans used in 
the series was thoroughly screened and mixed, after which the moisture 
content and moisture-holding capacity of the soil were determined. 
Sufficient water was then added to bring the moisture up to the desired 
point of saturation, after which the soil was again thoroughly mixed. 
After standing covered for a day, the soil was mixed repeatedly and 
potted, a weighed amount being placed in each can. The cans were 
equipped with a small inverted flowerpot placed in the soil at about the 
level of the seed bed. This pot was connected with the surface by a 
short glass tube fitted with a plug to prevent air circulation. The fully 
equipped container was then weighed and the weight recorded, after 
which the can was placed in the temperature tank. 
^ Soil moisture was kept as nearly constant as possible during the progress 
of the experiments by daily weighing and adding water to restore the 
original weight. Evaporation from the soil surface was prevented 
by a dust mulch or coarse sand on the surface. In this way the moisture 
variation was reduced to a minimum, although it was found impossible 
to maintain exactly comparable moisture conditions at various soil 
temperatures. Since the critical range of soil moisture had been pre¬ 
viously determined, it seemed logical to believe that the two factors, 
