S O I L A R T H R O P O D S A M P L I N G
29
The most acute technical difficulties arise when such studies are t o be
based on woodland fauna. Most woodland arthropods are present both
in the litter (Forna) and in the F and H layers beneath it ; some species
such as the Onydhiuridae (Collembola) are confined to the soil and
others such as spiders and perhaps the larger Collembola inhabit the
litter, provided it is present throughout the year. It follows that to
confine the study t o true soil species would limit greatly the choice of
species and at least preliminary investigations should cover both litter
and soil. The high proportion of undecomposed organic matter in the
F layer would usually point to the use of a funnel-type apparatus of
advanced design for the soil fauna investigation but the normal Tullgren funnel is unsatisfactory for most types of litter and careful tests of
the relative advantages of the Lloyd, Kempson and Ghelardi method
against the small funnel extractor with air conditioning would be desirable before coming t o a positive decision on the best means of extracting litter fauna. The vertical division of the medium into litter and
soil layers is obviously important if the distribution of the animals is
t o be followed but it is imperative t o use the same sampling points for
obtaining both the litter and the soil units if errors are t o be avoided.
For this reason it might be preferable to use the same technique for
both soil and litter samples even if a vertical separation proved necessary. It remains true that methods for the satisfactory separation of
arthropods from woodland litter have still to be tested and developed.
VI. SUMMARY A N D CONCLUSIONS
This paper is intended as a practical summary of current methods
used for sampling soil arthropod populations. Broadly speaking such
methods are used in three main types of work: exploratory work which
is not concerned with quantitative population estimates, but is aimed
at finding what species occur in an area ; secondly “community” studies
which are concerned with the relative abundance of a wide range of
species and often over a wide range of habitats; thirdly “trophic”
studies which demand a knowledge of absolute abundance usually of
relatively few species and in a single habitat. Each of these fields of
work makes its particular demands on sampling and extraction techniques particularly regarding the sizes and numbers of sample-units,
the accuracy expected and the range of organisms t o be found and
enumerated. This has not always been recognized in the past and labour
and resources have been inefficient.ly employed both in the field and the
laboratory. The planning of a complete sampling system requires careful
choice and matching of experimental design, extraction technique
and methods of examination and counting in the light of the above
considerations.
29
The most acute technical difficulties arise when such studies are t o be
based on woodland fauna. Most woodland arthropods are present both
in the litter (Forna) and in the F and H layers beneath it ; some species
such as the Onydhiuridae (Collembola) are confined to the soil and
others such as spiders and perhaps the larger Collembola inhabit the
litter, provided it is present throughout the year. It follows that to
confine the study t o true soil species would limit greatly the choice of
species and at least preliminary investigations should cover both litter
and soil. The high proportion of undecomposed organic matter in the
F layer would usually point to the use of a funnel-type apparatus of
advanced design for the soil fauna investigation but the normal Tullgren funnel is unsatisfactory for most types of litter and careful tests of
the relative advantages of the Lloyd, Kempson and Ghelardi method
against the small funnel extractor with air conditioning would be desirable before coming t o a positive decision on the best means of extracting litter fauna. The vertical division of the medium into litter and
soil layers is obviously important if the distribution of the animals is
t o be followed but it is imperative t o use the same sampling points for
obtaining both the litter and the soil units if errors are t o be avoided.
For this reason it might be preferable to use the same technique for
both soil and litter samples even if a vertical separation proved necessary. It remains true that methods for the satisfactory separation of
arthropods from woodland litter have still to be tested and developed.
VI. SUMMARY A N D CONCLUSIONS
This paper is intended as a practical summary of current methods
used for sampling soil arthropod populations. Broadly speaking such
methods are used in three main types of work: exploratory work which
is not concerned with quantitative population estimates, but is aimed
at finding what species occur in an area ; secondly “community” studies
which are concerned with the relative abundance of a wide range of
species and often over a wide range of habitats; thirdly “trophic”
studies which demand a knowledge of absolute abundance usually of
relatively few species and in a single habitat. Each of these fields of
work makes its particular demands on sampling and extraction techniques particularly regarding the sizes and numbers of sample-units,
the accuracy expected and the range of organisms t o be found and
enumerated. This has not always been recognized in the past and labour
and resources have been inefficient.ly employed both in the field and the
laboratory. The planning of a complete sampling system requires careful
choice and matching of experimental design, extraction technique
and methods of examination and counting in the light of the above
considerations.