Island forms of the same species [of birds] show every
gradation from extremely small differences in average
size to differences as great as those which separate some
of the species. The differences tend to be greater the
greater the degree of isolation. Some of them are adaptive
and others seem non-adaptive.
(Lack 1947a, p. 162)In the island ecology section we explored patterns
and processes while treating species as standard
and unchanging units of analysis. This is a useful
simplifying assumption, but, as we know, species
are not fixed, nor are they homogenous entities. In
this section of the book we consider the special
insights into evolutionary change that islands have
provided.
We start from the point at which a new colonist
arrives on an island, and then consider the micro-
evolutionary processes that go to work on such
populations and which result in a fascinating array
of emergent island effects or syndromes. The
changes with which we are concerned in this chap-
ter are expressed in genotypic and phenotypic char-
acters, sometimes exemplified by island endemic
species, but frequently evidenced at subspecific
level, i.e. they do not necessarily involve speciation
and the recognition of distinct island and mainland
species. The processes include a mix of essentially
chance effects, resulting in the divergence of island
lineages from their mainland source pools, and of
directional evolutionary forces that appear to oper-
ate in a consistent fashion across many remote
islands and taxa. These micro-evolutionary
processes and patterns are the building blocks
involved in the phenomena of phylogenesis and
the emergent macro-scale patterns of island evolu-
tion described in the following two chapters.
7.1 Founder effects, genetic drift, and bottlenecks
Islands form in two fundamentally different ways:
either by arising out of the sea, at which point they
are empty of land organisms, or by once contiguous
areas of land becoming separated from one another
by incursions, as for example when sea levels rise at
the termination of ice ages (Chapter 2). Thus classic
oceanic islands start empty, whilst land-bridge
islands have a full complement of species when
they first become proper islands. However, some
islands have such a complex or poorly known envi-
ronmental history that this simple dichotomy can
be difficult to apply, or to agree upon (e.g. see
McGloneet al. 2001; McGlone 2005; McDowall 2005;
Morrone 2005). Here we set these issues to one side,
mostly taking examples from true oceanic islands,
thus allowing us to examine colonization events
associated with long-distance dispersal to islands
that start (and remain for long periods) species-
poor for their area, and which develop from a
biased subset of the mainland species pool.
Thefounder principle(Mayr 1954) is that typi-
cally a species immigrating to a remote island will
establish by means of a tiny founding population.
This contains only a subset of the genetic variation
in the source population and it subsequently
receives no further infusion. This immediately pro-
vides a bias in the island population, on which
other evolutionary processes then operate. Genetic
variability can be increased after establishment by
mutation and re-sorting. One important form of
re-sorting is genetic drift, the chance alteration of
allele frequencies from one generation to the next,
which may be particularly important under sus-
tained conditions of low population size. In167CHAPTER 7