Each parameter was increased
10-fold for tests of linkage disequilibrium. We observed an
excess of homozygotes and some amplification failures
at several loci and thought that null alleles might be a
possible cause. Most but not all of the amplification
failures observed were considered possible artefacts rather
than null homozygotes and therefore the frequencies of null
alleles were estimated using equation 3 of Brookfield (1996),
r
=
D
/(2
−
D
) where
D
= (
H
E
−
H
O
)/
H
E
.