a particular form of the buttercup can be described
by the following nomenclature:
Kingdom Plantae
Division Spermatophyta
Order Ranunculales
Family Ranunculaceae
Genus Ranunculus
Species bulbosus
Subspecies bulbifer
How is this taxonomy decided? Are the units
actually comparable? The following is a fairly typi-
cal textbook definition of the species unit: ‘groups
of actually or potentially interbreeding natural
populations which are reproductively isolated from
other such populations’ (Mayr 1942, p. 120). This is
a form of the biological species concept—promoted
notably by Ernst Mayr—which has been predomi-
nant for the past 60 years (Mallet 1995). Yet the
question of ‘reproductive isolation’ (Table 8.1) is
often an unknown quantity, as populations may be
geographically separated (non-overlapping distri-
butions are termed allopatric). In such circum-
stances they may exhibit sufficient morphological
differentiation to be considered subspecies or even
separate species, yet actually remain capable of
interbreeding if placed into a common enclosure.
Then again, species may be capable of interbreeding
in the laboratory and have sympatric distributions
(i.e. occur in the same geographic area) but remain
reproductively isolated in the wild because of
behavioural differences. Sometimes a rider is
added to restrict the definition to those individuals
that can successfully interbreed to produce viable
offspring. Yet this rule too can be broken, as species
which are recognized as ‘good’ may interbreed
successfully in hybrid zones (Ridley 1996), an
example being the native British oaks Quercus
petraeaandQ. robur. Studies of the genetics of
British populations have shown that both maintain
their integrity over extensive areas of sympatry, but
form localized ‘hybrid swarms’ in disturbed or
newly colonized habitats (White 1981). Apparently
this is a stable situation, as Miocene fossil leaf
impressions suggest that these taxa became differ-
entiated no less than 10 million years BP. Some pre-
fer to treat such partially interbreeding populations
as ‘semispecies’ within ‘superspecies’ (White 1981).
Hybridization between closely related island
species appears to be fairly common (Silvertown
et al.2005), especially early on in the development
of an adaptive radiation (Grant 1994).
For many groups, moreover, the taxonomy
remains poorly known and subject to revision. For
example, molecular analyses of the Bufo margari-
tifera complex of toads in Central and South
America suggest that toads recognized by this
name actually represent composites of morpholog-
ically cryptic species that are genetically unique
(Haaset al. 1995). If the criterion of being reproduc-
tively isolated is a difficult one to employ on living
organisms, it is even more so with many fossil
organisms. Often the best that can be done is to
determine whether the morphological gaps
between specimens are comparable with those for
living species that are reproductively isolated.
Thus, even the species unit, although central to
so many biogeography purposes, has blurred edges
(see also Otte and Endler 1989; Ridley 1996). The
units below the species are a matter of considerable
debate and strong opinions are held that they are
either important ecologically and therefore matter
or that they are arbitrary and the whim of the tax-
onomist (see discussion in Mallet 1995). The signif-
icance of this issue can be shown by reference to the196 SPECIATION AND THE ISLAND CONDITION
Table 8.1A classification of isolating mechanisms in animals (from
Mayr 1963, slightly modified)
- Mechanisms that prevent inter-specific crosses (pre-mating
mechanisms)
a. Potential mates do not meet (seasonal and habitat (ecological)
isolation)
b. Potential mates meet but do not mate (ethological isolation)
c. Copulation attempted but no transfer of sperm takes place
(mechanical isolation) - Mechanisms that reduce full success of inter-specific crosses
(post-mating mechanisms)
a. Sperm transfer takes place but egg is not fertilized (gametic
mortality)
b. Egg is fertilized but zygote dies (zygote mortality)
c. Zygote produces an F 1 hybrid of reduced viability (hybrid
inviability)
d. F 1 hybrid zygote is fully viable but sterile, or produces deficient
F 2 (hybrid sterility)