A traditional method for categorizing speciation events is on the basis of geography, i.e., as occurring in allopatry, sympatry, or parapatry [4,8,13]. Another way to categorize speciation events is by identifying "isolating barriers" that initiate and maintain separation between incipient species [2,4,14]. Isolating barriers are typically categorized as occurring before mating (premating), after mating but before zygote formation (postmating prezygotic), or after zygote formation (postzygotic). These categories are then further parsed into nested subcategories. Although identifying isolating barriers is a critical part of any research on speciation, we do not discuss these subcategories further because they have been reviewed elsewhere [4,14]. Instead, we focus our attention on one of the ultimate and long-standing questions of speciation: What are the initial causes of reproductive isolation (and thus of speciation) among diverging, incipient species?
Schluter  proposed that the initial causes of speciation can be divided into four general modes: hybridization and polyploidy, genetic drift, uniform natural selection, and divergent natural selection. Sexual selection is not considered a separate mode but rather as a potential contributor within each. Under uniform natural selection, in which different populations are exposed to similar ecological environments, divergence occurs as different advantageous mutations arise and spread to fixation in different populations. These mutations may be incompatible when brought together by interpopulation mating, thus causing reproductive isolation among different populations adapted to similar environments [4,16]. Conversely, under divergent natural selection, similar mutations may arise in multiple populations, but different mutations will be favored and therefore retained in different ecological environments. Adaptive divergence may then lead to initial reproductive isolation in a number of ways. For example, mutations favored in one environment might confer reduced fitness in alternative environments, perhaps also favoring individuals that mate assortatively. Schluter [3,15] refers to this latter mode of speciation — byproduct effects of divergent natural selection — as "ecological speciation." This is the arena within which we consider the role of biomechanics.
Isolating barriers that arise through ecological speciation (or other speciation modes for that matter) may be manifest in two general ways. On the one hand, adaptation to different environments may lead to reproductive isolation that depends directly on features of those environments, such as the availability and distribution of food resources. Such isolating barriers are therefore considered "extrinsic," "conditional," "environment dependent," or "ecologically dependent" [4,17,18]. To illustrate, male displays are often optimized through natural selection for effective transmission in the particular environments that animals inhabit [19-21]. Songbirds living in forested environments, for example, tend to evolve mating songs with lower frequencies and lower rates of note repetition, as adaptations that minimize degradation by reverberation . Optimization of transmission properties in local habitats may thus diminish the effectiveness of particular songs when sung in alternative environments. If the signaling environments inhabited by a species are sufficiently divergent, and the signals are differentially effective in these environments, then females may mate preferentially with males from local environments. A locality-dependent process of reproductive isolation would thus be initiated . Isolation barriers in this example would be premating. Ecologically dependent postmating barriers are also feasible. Consider, for example, hybrids with phenotypes that are intermediate to parental phenotypes. Hybrids may suffer lower levels of fitness in either parental environment because of the difficulty in accessing resources on which parental types are specialized [e.g., 23]. However, as conceptualized in Figure 14.1A, hybrids with intermediate phenotypes might enjoy higher fitness, relative to either parental type, in "intermediate" environments, e.g., in which resource parameters fall in between those in parental environments [e.g., 24].
On the other hand, adaptation to divergent environments may lead to reproductive isolation that is manifest independently of environmental features. Such isolating barriers are considered "intrinsic," "unconditional," "environment independent," or "ecologically independent" [4,17,18]. Ecologically independent premating barriers may arise if traits under divergent selection are also used in mate choice and in ways that do not depend on the mating environment. In stickleback fishes, for example, divergent selection between benthic and limnetic morphs has fostered the evolution of differences in body size. Laboratory mating studies suggest that body size plays an important role in mate choice, in that females appear to choose males with body sizes similar to their own [25,26]. Ecological independence of body size as a mating cue is illustrated by the observation that body size cues are effective not only in the field but also under laboratory conditions, in which natural variation in environmental transmission properties is not present [e.g., 27,28]. As a generalized example of ecologically independentpostzygotic barriers, hybrids of diverging lineages can have genetic incompatibilities that are expressed equivalently (or near equivalently) in any environment. Under such circumstances, hybrids would experience low fitness in nature even if intermediate environments are present (Figure 14.1B), and perhaps even under benign laboratory conditions — although such barriers may be stronger under more stressful conditions . Many studies have demonstrated genetic incompatibilities in hybrids , although we are not aware of any conclusively attributing the resulting isolation to divergent natural selection.
Exploring the distinction between ecologically dependent and ecologically independent isolating barriers is useful because it speaks to the integrity of species in the face of environmental perturbation. Ecologically independent barriers may be more powerful and robust because they should persist even if the environment changes, at least during initial stages of divergence. In contrast, ecologically dependent barriers may collapse immediately after environments change and could therefore represent a more fragile and tenuous route to speciation. For example, in a long-term study of Darwin's finches on Daphne Major Island, environmental changes resulted in increased relative fitness for hybrids, which has led to the
Population X Hybrids
Environment Intermediate Environment X environment Y
FIGURE 14.1 Differences between (A) ecologically dependent and (B) ecologically independent reproductive isolation. Assumed is a case whereby population X individuals are adapted to environment X, and population Y individuals are adapted to environment Y. When isolation is ecologically dependent, hybrids (with intermediate phenotypes) should have a higher fitness than parental types in intermediate environments. When isolation is ecologically independent, hybrids should have lower fitness than parental types in all environments, although the specific value for hybrid fitness relative to parental fitness could vary considerably (indicated by the arrows). Of course, both types of isolation may act at the same time, generating any number of intermediate scenarios for hybrid fitness.
breakdown of ecologically dependent isolating barriers and to the morphological convergence of formerly distinct species . And yet, ecologically dependent barriers can evolve very quickly, simply because adaptive divergence can be very rapid in nature [reviews: 30,31]. As examples, insect herbivores adapting to introduced host plants have evolved ecologically dependent barriers after less than a few hundred generations [32,33; see also 34]. In addition, ecologically dependent barriers should be particularly widespread, and may therefore cause initial reductions in gene flow that allow subsequent ecologically independent barriers to evolve .
Was this article helpful?