186 
A soil yielding excellent pasture from the valley 
of the Avon, near Salisbury, afforded one-eleventh 
of coarse siliceous sand ; and the finely divided 
matter consisted of 
7 Alumina, 
14 Silica. 
63 Carbonate of lime. 
2 Oxide of iron. 
14 Vegetable, animal, and saline matter. 
In all these instances the fertility seems to depend 
upon the state of division, and mixture of the earthy 
materials and the vegetable and animal matter ; 
and may be easily explained on the principles which 
I have endeavoured to elucidate in the preceding 
part of this Lecture. 
In ascertaining the composition of sterile soils 
with a view to their improvement, any particular 
ingredient which is the cause of their unproductive- 
ness, should be particularly attended to ; if pos- 
sible, they should be compared with fertile soils in 
the same neighbourhood, and in similar situations, 
as the difference of the composition may, in many 
cases, indicate the most proper methods of improve- 
ment. If on washing a sterile soil it is found to 
contain the salt of iron, or any acid matter, it may 
be ameliorated by the application of quick-lime. 
A soil of good apparent texture from Lincolnshire, 
was put into my hands by Sir Joseph Banks as re- 
markable for sterility. On examining it, I found 
that it contained sulphate of iron ; and I offered 
the obvious remedy of top-dressing with lime, which 
converts the sulphate into a manure. If there be 
an excess of calcareous matter in the soil, it may 
