Geoscience Reference
In-Depth Information
F
IG. 8
Sampling soil profiles with an auger; W.M. Williams on Snowdon. (Photograph D.F.B.)
The second point is that, whatever type of classification is favoured, at whatever
level of detail, actual landscapes are usually more complex and less orderly than
might be assumed from the idealized world of classification. The best any soil map
can do is to show, with greater or less precision, the probability of a specified soil
class existing at any particular spot. Within a single soil class, significant variations
in chemical and other properties usually occur between representative profiles as a
whole, or between equivalent horizons within them. Management and seasonal cli-
matic differences extend the inherent variability that occurs naturally over very short
distances, even within an uncultivated and morphologically uniform soil. Over Bri-
tain as a whole, broad geographic gradients create zones with characteristic groups of
soils. Within these zones, however, much local diversity can be superimposed. From
the resulting complex soil picture, even the most efficient classifications have to use
arbitrary divisions in distinguishing particular types. Plant ecologists (phytosociolo-
gists) have grappled with similar problems in classifying plant communities.
Thus, in any classification, a given profile class or horizon type does not strictly
define its population of soil organisms, in species or numbers. These classes have a
broad two-way relationship with the character of the soil fauna as a whole, but this is
also strongly affected by the particular vegetation cover and its consequent litter fall,
and by past and present land management. The versatility with which some organisms
