II.] TYPICAL SOILS 59 



If there be much "fine gravel" in the soil, it is best to 

 make a separate determination of its amount on a 

 sample weighing 50 grams, treating with acid as before, 

 and then washing the whole on to the 1 mm. sieve. 

 The result obtained should be taken as the true 

 percentage, and the other percentages found in the 

 analysis of 10 grams only should be recalculated to agree 

 with it. 



The sizes of the particles, the depth of the liquid, and the 

 times adopted above, are purely conventional. The time 

 of settlement required to obtain a fraction of any 

 given range of size can be determined by a series of 

 trials, the material remaining suspended in each case 

 is measured under the microscope until the right time 

 is hit off to secure the desired range of size in the 

 sediment. The relationship between the time of settle- 

 ment, the height of the liquid column, and the diameter 

 of the particles, is governed by the formula : 



r _ 2ff* 2 (q--p) 

 9 v 



where a- is the density of the particle, a its radius, p the 

 density, and 17 the coefficient of viscosity of the liquid. 

 The application of the formula, however, requires to be 

 checked by observation with the microscope, because 

 the particles are not spheres. 



The hygroscopic moisture and the loss on ignition 

 also require determination, which is described under 

 the chemical analysis of a soil. 



Interpretation of Results. 



It is as yet impossible to predict the behaviour of 

 a soil under cultivation from a consideration of its 

 mechanical analysis ; in a general way we can see 

 whether a soil is heavy, whether it is likely to dry 

 " steely," or whether it will crumble readily under 

 proper cultivation, and whether it is more suitable for 



