WATER IN SOILS 9 



shown in the results. The physiological water was finally 

 determined by deducting the per cent of non-available water 

 from the per cent of physical water in the soil at the time 

 the process of drying out began. The plants were grown 

 for experimentation in three and four Inch porous pots. In 

 order to control the evaporation, the three inch pots were 

 set in soil and kept there during the whole time of exper- 

 imentation. In order to prevent too rapid loss of water, the 

 four inch pots were placed inside painted pots, during the 

 period of drought to which they were subjected in determin- 

 ing the non-available water. All soil samples were taken 

 from the soil in the lower part of the pot, in the vicinity of 

 the roots, the upper part of the soil being discarded, as it 

 contained very few roots. Owing to the great number of 

 pots used in experiments it was impossible to keep a daily 

 record of the physical water in the soil of each. In the 

 comparative plots, soil samples were taken from one pot in 

 each condition, and tbe water in the others estimated from 

 this record. The non-available water was always determined 

 by taking a soil sample from each pot used in the experi- 

 ment. 



3. Water in Plants. In determing the amount of water in 

 plants, the plant was carefully removed from the soil, the 

 small particles of soil adhering to the roots were shaken off 

 where possible, and the plant quickly weighed. It was then 

 subjected to a heat of 212" F. in a drying oven from 8 to 24 

 hours, depending upon the size of the plant. The per cent 

 of moisture was then determined from the loss in weight. 

 All weights both of plants and soil samples were made with 

 a laboratory balance sensitive to one milligram. 



4. Physical Conditions. In order to grow plants under dif- 

 ferent physical conditions, four plant cases with glass panes 

 overhead and in the sides were constructed and placed in 

 the south windows of the laboratory. One of these had a 

 connection with the outside air and by this means was kept 

 at a lower temperature most of the time, the exception 

 being the last week of the experiment. This case is desig- 

 nated as Case I. The light, though less intense than that in 



