THE SOIL 
195 
control, so that none is wasted and the crop is supplied at 
the right time. The loss of water and chilling by evaporation 
are prevented by a “ mulch ” or loose cover of farm refuse, 
or by so hoeing or harrowing the soil that it serves as a mulch. 
The dry farmer keeps the surface broken after every fall of 
rain until sufficient for a crop has soaked underground. He 
then rolls the ground so as to connect the surface with the 
water-table, so that the plants may be nourished. This 
method was practised in early times and its adoption has 
extended wheat cultivation in Australia, beyond the former 
supposed limit of * Goyder's Line,” which marked the edge 
of the area with a rainfall of 14 inches. 
The colour of soils affects their earliness. A black soil, 
such as the chernozen of Russia, and the black cotton soil 
of India, absorbs more heat and is warmer than a pale soil. 
The Indian black soils are coloured by a colloidal silicate of 
iron and aluminium containing some organic matter (Harrison 
and Swan (Mem. Agric. Research Inst., Pusa, No. 5, 1913). 
Some black soils, such as those of the highlands of Benguella, 
though apparently promising, are infertile, because they 
consist of coarse quartz grains coated by a thin film of 
colloidal silicate. Other black soils, such as those of the lava 
plains of East Africa, are darkened by the high proportion 
of iron and humus; they are chemically rich, but are diffi- 
cult to work as the clay in the rainy season becomes semi- 
fluid. It is then impassable to wheeled traffic and horses. 
In the dry season the soil shrinks and a network of deep 
cracks tears across the plant roots and allows the water to 
sink deeper and thus increase the depth with these incon- 
venient properties. 
Soir CoMPOSITION AND SurvVEYs—Agricultural chemistry 
in its early days formed exaggerated expectations of the 
help it could give the farmer; recognition of the limitations 
of soil analysis led to the view that the biological factors are 
most important. The most accurate test available of them 
is nitrification ; but, as pointed out by Burgess (Soil Science, 
vi, 1918, pp. 449-62), it does not explain the differences in 
soils or indicate how poor soils can be improved. The 
address by Dr. Crowther, an agricultural biologist, to the 
British Association in 1923, shows the return to greater 
faith in soil composition; he expressed the * conviction