rate and the fluctuation in abundance. To estimate
the extent of such fluctuations long-term investigations
with samples taken in consecutive years are
needed. Long-term studies describing year-to-year
variability in the catch of each carabid species are
rare, and some of the longer studies do not show
the number of individuals caught per year and
location separately. In open sandy habitats in the
Netherlands Den Boer (1981) found a wide range of
net reproduction rate as well as the relative density
of carabid beetles, which varied by a factor of over
one hundred within 20 years. In more stable
habitats such as recent woodlands, Althoff et al.
(1992) detected a much smaller variation in
population size of a woodland-inhabiting carabid
beetle in the Westphalian lowlands (Germany). But
nothing is known about the fluctuations in abundance
of several woodland-inhabiting carabid beetles
of ancient woodlands. To obtain this
information it is necessary either to estimate the
absolute density or to obtain a whole year catch
with pitfall traps for several consecutive years.
Pimm and Redfearn (1988) concluded that in the
case of insects a period of at least 8 years may be
necessary to describe the extent of fluctuations in
this taxon. Over shorter periods the standard
deviations of the logarithms of annual density
(and other measures) increase with time. The
long-term investigation in an ancient woodland
area of northwestern Germany presented here for
the first time has covered a period of 9 years (1994–
2002).
The aim of this study is to estimate the amplitude
of fluctuations of the relative density of carabid
species in an ancient woodland. These fluctuations
are then compared with those of carabid species
inhabiting other habitats.
Materials and