76 WLLTIXG COEFFICIENT FOE DIFFERENT PLANTS. 



tractive force upon the soil moisture than another, but to the more 

 perfect root distribution of one variety as compared with another. 

 Drought resistance in certain plants can not, then, be attributed to 

 their ability to exert a greater force upon the soil moisture and so 

 gain an additional water supply. 



A series of comparisons has also been made of the relative time of 

 wilting of different plants grown together in the same pot. In prac- 

 tically every case wilting occurred simultaneously, which is in har- 

 mony with the above conclusions. 



A balancing method has been developed for determining the 

 wilting coefficient for plants in which wilting is difficult to observe 

 owing to structural peculiarities. The plant is potted in a glazed 

 pot and the soil surface sealed with wax, as before. The potted plant 

 is then mounted horizontally in a frame balanced on knife-edges. 

 The water transpired by the plant is at first replaced from the supply 

 jn the soil. During this time the soil end of the system tends to move 

 upward, owing to the loss of water, and is kept balanced by adjusting 

 a counterpoise. A short period then occurs during which the water 

 lost through transpiration is supplied both from the soil and from 

 the storage tissues of the plant. During this time the system appears 

 sluggish. Finally the amount supplied from the soil is insufficient 

 to offset the increasing amount lost from the storage tissues of the 

 plant and the direction of motion of the system is reversed, the 

 plant moving upward. The soil-moisture content at this time cor- 

 responds to the wilting coefficient of that soil for the particular plant 

 used, since the loss of water from the plant tissues would be accom- 

 panied by wilting except for structural peculiarities. 



It has been shown by this method that the olive and the cactus 

 undergo a reduction in the moisture content of their aerial parts at 

 the time other plants wilt, although no outward evidence of such 

 loss is apparent. 



That portion of the soil-moisture content which is available for 

 plant growth is represented by the difference between the actual 

 water content and the wilting coefficient. The latter determination 

 is consequently essential in any critical study of the relation of 

 plant growth to soil moisture. The desirability of a rapid determi- 

 nation of the wilting coefficient in connection with field work led to 

 an investigation to determine whether the wilting coefficient could 

 be computed from physical measurements of the moisture reten- 

 tivity of the soil. A comparison of the wilting coefficient has been 

 made with the moisture equivalent, hygroscopic coefficient, moisture- 

 holding capacity, and the mechanical analyses of a series of soils 

 ranging from sand to clay. From this comparison a series of linear 

 relationships has been established, as expressed in the following 



230 



