In simulations of a model with topological sectors,
algorithms which proceed in small update steps
tend to get stuck in one sector, especially on fine
lattices. This distorts the numerical results; in
particular it is not straightforward to measure the
topological susceptibility chi_t. We test
a method to measure chi_t even if configurations
from only one sector are available. It is based on the topological
charges in sub-volumes, which we denote as ``slabs''.
This enables the evaluation of chi_t, as we demonstrate
with numerical results for non-linear sigma-models and for
2-flavour QCD.
