Relation of Soils to Moisture and Air 47 



samples. Cecil clay was placed in a tumbler and covered 

 with a layer of dry earth an inch thick. Some light loam 

 from St. Mary's County, Maryland, was treated in the 

 same way, and one glass was left to evaporate without any 

 cover. The total loss from the Cecil clay was somewhat 

 less than half as much as that from St. Mary's loam, show- 

 ing the marked difference in the resistance offered by the 

 two soils to the passage of watery vapor. The uncovered 

 loam lost three times as much as the covered. 



From various experiments it is evident that the per- 

 centage of humus in a soil has a marked effect on its 

 capacity for retaining moisture. In general, then, we 

 may assume that evaporation is less rapid from a clay soil 

 than from a loam or sandy soil, and that to retain the 

 moisture brought up by capillarity in any soil it is advisable 

 to keep a layer of loose and dry soil on the surface and 

 to prevent the formation of a crust there which would 

 promote the rapid evaporation. In short, by shallow 

 cultivation during dry weather we can prevent the loss of 

 moisture and hold the capillary moisture where the roots 

 of the plants need it. 



There is, however, one point that has been 

 Selective determined by experimenting with water in 



Ptot Roots ^^^ solution of plant food in the soil. As is 

 well-known, the food which is taken up by 

 the roots of plants must be in complete solution. But the 

 extent to which this solution is due to the water in the soil 

 is far less than is usually imagined. At the Minnesota 

 experiment station grain grown in pots filled with a very 

 fertile soil, grew finely and matured normally. In other 

 pots sand entirely free from plant food was also planted 



