200 § 108. ANxVLYSIS OF SOILS AND KOCKS. 



knead a small portion of it into a thick dough with wateiv 

 and, with a spatula, make several prisms 5 cm. long, and 

 1 cm. square on the end ; let them dry in the air, and 

 then observe what weight must be laid on the back of a 

 knife in order to force it through each one. The same 

 prisms may be used to determine how much the soil 

 shrinks on drying, by noting the difference between their 

 lengths when wet as at first, and after they are dry. 



2. To determine the force with which the soil adheres 

 to wood or iron, fill a cubical zinc box about 6 cm. square 

 and deep, whose bottom is pierced with small holes, and 

 covered with a piece of linen, with the soil, shaking it 

 down frequently while filling ; then immerse the bottom 

 of the box in water, and, when no further increase of 

 w^eight is observed, lay a smooth piece of beech wood, 3 

 cm. square, on the w^et surface of the soil, press it down 

 for the space of ten minutes, by a weight of 100 grms., 

 attach the disk to one arm of a balance, and observe what 

 weight must be put in the pan connected with the other 

 arm, in order to detach the disk from the soil. Try a 

 similar experiment with a disk of iron. 



EXPERIMENTS WITH PLANTS IN CONNECTION WITH ANALY- 

 SIS OF SOILS. 



108. The further development of soil analysis on one 

 hand, and its simplification on the other by eliminating 

 useless or unnecessary determinations, can be accom- 

 plished only by combining suitable experiments with 

 modes of culture and manuring, and growing plants, with 

 accurate analyses of the soils used. 



In the field but little can be done in this way, since so 

 much care and labor are required, in order to obtain a 

 tolerably fair representative sample of the soil, even of a 

 small plot. 



Kot seldom, however, a more luxuriant vegetation is 



