THE WATER OF THE H-IBITAT 13 



17. Depth of samples. The general rule is that the depth 

 of soil samples is determined by the distance to which the roots 

 penetrate. The practice is to remove the air-dried surface in 

 which no roots are found, and to take a sample to the proper 

 depth. When the actively absorbing root surfaces are localized, 

 as in deep-rooted plants, it is necessary to confine samples to 

 the area in which absorption occurs. This is especially true 

 when the water supply of a panicular species of plant is to be 

 determined, but. in the case of vegetation in general, roots pene- 

 trate to such different depths that a sample which includes the 

 greater part of the distance concerned is satisfactory. The usual 

 depth of a sample is 3 decimeters. In shallow or wet soils, cores 

 to a depth of 1 or 2 decimeters suffice, while in very dry soils, 

 and in the case of shrubs and trees, deeper samples are required. 



Experiment i. Measurement of water content. Take a 2-din. 

 sample from a dry soil and another from a moist soil: take one also 

 from a sand and a loam. Weigh, dry, and compute percentages upon 

 the basis of the dry soil. 



18. The determination of available water. The amount of 

 water that a plant can absorb from the soil can be readily deter- 

 mined only by finding the amount left when the plant wilts com- 

 pletely. Tins determination can easily be made in the labora- 

 torv or plant-house, though in the field it is attended with some 

 difficulty. A thrifty plant growing in a pot of medium size is 

 the l>est for the purpose. It is necessary that the pot be glazed 

 or covered with sheet rubber in order to prevent too rapid dri-ing 

 of the soil. At the beginning of the experiment, three soil samples 

 are taken in such a manner that the}- will indicate any variation 

 in water content in different parts of the soil. The depth of the 

 core is regulated by the size of the pot and the jx>sition of the 

 roots. The holard is found in the usual way. and is expressed 

 upon the basis of 100 grams of dry soil. e.g.. 2-5 100. The aver- 

 age of the three samples is taken as representative: this average 

 is most readily arrived at by weighing and dr>-ing the samples 

 in one can. The soil is then permitted to dr^- out slowly. Sud- 

 den drouth often impairs the power of al^sorption. and the plant 

 wilts even though considerable available water is present. The 

 proper time to take the second reading is indicated by the thor- 

 ouch wilt ins of the leaves. It is undesirable to wait for com- 



