Biology Reference
In-Depth Information
the example of Darwin's finches on the Galapagos Islands in detail. Thus,
morphological differences and particularly differences in the size of beaks
of species inhabiting the various islands have been interpreted as being
due to genetic drift, interspecific competition, recognition, and physical
and vegetational differences between islands. Null-hypotheses (that spe-
cies are independent of each other) appear to show that interspecific
competition has little effect on birds of the Galapagos Islands and those
of the New Hebrides, but the null-hypopthesis was rejected for birds and
bats of the West Indies. However, the validity of the null-hypothesis
used also has been rejected by some (for a detailed discussion and refer-
ences see Connor and Simberloff
1986
). Nevertheless, Connor and
Simberloff conclude that, for evaluating non-experimental evidence, a
null-hypothesis should usually be tested first. For some communities,
predation may cause effects similar to those due to competition.
Density distributions were used by Janovy et al.(
1992
,
1995
), who
presented a lottery model, in which hosts acquire their parasites by Monte
Carlo type sampling from multiple kind arrays. Assumptions for a null
model were: ''(1) successes are independent events and (2) it is possible to
fail completely to acquire any parasites or their analogs.'' They postulated
that ''claims of parasite interaction need to be supported by studies
showing departures from the predictions of this model.'' The authors
state that their ''model is applicable to any system that mimics a multiple-
kind lottery'' in which the above two assumptions can be made. They
applied their model to data sets from three host/parasite systems in order
to test for the occurrence of interspecific competition, and found that
there were very few (2 out of 15) significant departures from the
expected values. According to the authors, the two departures did not
support the assumption that competition was responsible because they
were the result of a greater and not a smaller than expected number of
hosts infected with many species. The authors explained the departures by
abiotic events. The study is of great value because it shows convincingly
that the arguments given by various authors in support of interspecific
competition are not valid. Nevertheless, the point can be made that,
although a null model using density distributions calculated on the basis
of certain assumptions (e.g., successes are independent events) can be
established, such a model is biologically meaningless since successes very
often are not independent and we cannot know a priori which ones are
and which ones are not. As pointed out by Janovy et al.(
1995
), infective
stages of parasites, for example, often use the same microenvironment,
such as the same intermediate host, and the likelihood of becoming
