SOIL 79 
parently increases in strength as the film grows thinner, and explains why it 
finally becomes impossible for the root-hairs to draw moisture from the soil. 
This property, like capillarity, is most pronounced in fine-grained soils, such 
as clays, and is least evident in the coarser sands and gravels. It seems to 
furnish the direct explanation of non-available water, and, in consequence, 
to indicate that the chresard is an immediate result of soil texture. 
115. Mechanical analysis. From the above it is evident that, with the 
same rainfall, coarse soils will be relatively dry, and fine soils correspondingly 
moist. However, this difference in holard is somewhat counterbalanced by 
the fact that the chresard is much greater in the former than in the latter. 
The basis of these relations can be obtained only from a study of the tex- 
ture of the soil. The usual method of doing this is by mechanical analysis. 
This is far from satisfactory, since the use of the sieves often brings about 
the disaggregation of groups of particles which act as units in the soil. 
Furthermore, the analysis affords no exact evidence of the compactness of 
the soil in nature, and tests of capillarity and porosity made with soil sam- 
ples out of position are open to serious error. Nevertheless, mechanical 
analyses furnish resulls of some value by making it possible to compare soils 
upon the basis of texture. For ecological purposes, mi- 
nute analyses are undesirable; their value in any work 
is doubtful. A separation of soil into gravel, sand, and 
silt-clay is sufficient, since the relative proportion of 
these will explain the holard and chresard of the soil 
concerned. The latter are also affected in rich soils, 
especially of forests, by the organic matter present. If 
this is in a finely divided condition, the amount is de- P 
termined by calcining. When a definite layer of leaf- Fig. 25. Sieves for 
mold is present, as in forests and thickets, its water-value oi iad 
is found separately, since its power of retaining water is altogether out of 
proportion to its weight. 
116. Kinds of soils. It is very doubtful whether it is worth while to at- 
tempt to distinguish soils upon the basis of mechanical analysis. Un- 
questionably, the most satisfactory method is to distinguish them with 
respect to holard and chresard, and to regard texture as of secondary im- 
portance. A series of soil classes which comprise various soil types has 
been proposed by the U. S. Bureaut’of Soils as follows: (1) stony loam, 
(2) gravel, (3) gravelly loam, (4) dunesand, (5) sand, (6) fine sand, 
(7) sandy loam, (8) fine sandy loam, (9) loam, (10) shale loam, (11) 
silt loam, (12) clay loam, (13) clay, (14) adobe. These are based 
1Instructions to Field Parties and Descriptions of Soil Types, 35, 1903. 
