10
Recent emphasis on accurate sampling procedures for
insects in peanuts has revealed the need for better defini
tion of the spatial distribution of these insects. Also,
knowledge of spatial patterns provides insight into the
biology of the species in question. Data on the spatial dis
tribution of insects is necessary before development of se
quential sampling schemes may be effected (Waters 1955).
Spatial distribution models are probability distribu
tions that relate the frequency of occurrence of an event,
depending on the mean of the measurements and in some cases
on one or more parameters.
The most useful statistical distributions in entomo
logical research have been the Poisson and negative binomial
(Southwood 1978). The Poisson distribution describes a
random distribution with variance equal to the mean. In
insect sampling the variance most commonly will be larger
than the mean, indicating that the distribution is aggre
gated or clumped. The most useful distribution describing
a clumped insect population has been the negative binomial.
The versatility of this distribution arises from the fact
that it may arise from at least 5 different models (Waters
and Henson 1959).
The probabilities of a Poisson distribution are given
by
x = 0, 1, 2
P
x
where P is the expected proportion in the xth class and
x 0, 1, 2, . is the value of a discrete random variable.