Islands of speciation and speciation on islands

The concept of a species

We’ve spent some time before discussing the nature of the term ‘species’ and what it means in reality. Of course, answers to questions in biology are always more complicated than we wish they might be, and despite the common nomenclature of the word ‘species’ the underlying definition is convoluted and variable.

Before we delve deeper into some aspects of speciation – particularly about how it can occur despite violations to one of the most fundamental characteristics of the most common definition – let’s recap the nature of species and the species in nature.

Species are, in essence, a product of human cognitive limitation. That is, the concept of a species is an approach to attempt to delimit organisms into manageable groups. How we decide to define those groups, of course, can vary: in fact, there are dozens of different criteria under which we could define a ‘species’. The most widely accepted definition is the Biological Species Concept (BSC for short), which denotes any two organisms as belonging to separate species if they are reproductively isolated. This means that the two cannot reproduce together to form fertile and viable offspring (i.e. are unable to maintain, long term, hybrids). These caveats are important as just because two animals can form a hybrid offspring (e.g. see ligers, zorses, mules) does not mean these can be sustained over time.

The speciation continuum

As mentioned earlier, divergence of organisms throughout the tree of life progresses in a diffuse, spectrum-like manner. Just because we impose a particular cut-off of reproductive isolation doesn’t necessarily indicate that species comfortably fit into binaries. Additionally, reproductive isolation rarely evolves overnight and can be a slow, accumulating process as genetic differences evolve independently within two separate lineages of the same species. There’s more detail about that in this post, but the importance underlying concept is one of the speciation continuum. As different subgroups (often ‘populations’) slowly diverge from one another, the progressively move along the continuum. At some particular tipping point, this divergence is strong enough to prevent successful reproduction and thus reproductive isolation is achieved.

Speciation continuum figure.jpg — A diagram of the speciation continuum. In this figure, we start with a single ancestral population which splits into separate lineages. Over time, they diverge from one another. This gradual differentiation eventually, at some point, causes the two lineages to become separate species. Where? That’s hard to say.

Tipping the scale: reproductive isolation

The importance of reproductive isolation lies in its inference for the evolutionary history of different species. Under this concept, independent species do not share genetic traits or (direct) evolutionary pathways, and thus may evolve in very different directions with natural selection. Thus, the identification of reproductive isolation (although incredibly difficult to do in empirical, natural study systems) is an important factor in understanding the evolution of new species.

Of course, black-and-white ideologies rarely apply in biological science and to say that all species are 100% isolated from one another (and have been for long enough to achieve full evolutionary independence) might significantly underestimate Earth’s biodiversity. As studies continue to delve into the nature and process of speciation in natural systems, more examples of speciation despite gene flow are detected.

Hybridisation between ‘species’, even distantly in the past, reduces genetic divergence between the two populations by transferring alleles. However, there is a balance between the rate of alleles mixing across lineages (gene flow) and the individual forces of natural selection (differential selection) acting on each population. If natural selection very strongly promotes differentiation between the two populations, particularly in relating to ecological characteristics, the homogenisation effect of gene flow might be suppressed by the overarching impact. There has been significant debate about the balance of these two factors in the literature, but the concept of speciation with gene flow (alternatively, ecological speciation) has gained support in recent years.

Gene flow vs divergence fish figure — Gene flow vs. divergence in a theoretical fish system. On the **left**, we have a physical diagram of the environment and populations of fish. On the **right**, the representative trajectory of the different lineages. Initially, we start with a single widespread population (purple), which moves forward in time down the figure on the right. An isolating event causes the population to split into two separate lineages (red and blue, for simplicity). The two gradually diverge until another event causes them to come back together again (secondary contact). This facilitates gene flow between the populations and reduces their accumulated divergence by sharing unique alleles from red to blue and vice versa. A second isolating event separates them once more and divergence can progress again. How far the lineages diverge from one another is a careful balance of the homogenising effect of gene flow and the accumulation of different genetic variants.

Islands of speciation

So how, despite mixing genes between populations, can reproductive isolation occur? If gene flow is constantly reducing divergence between the two lineages in question, how do they ever progress far enough along the continuum to become new species? This has been an ongoing debate for some time, and there are a number of proposed mechanisms under which reproductive isolation could evolve (even rapidly). The accumulation of genetic differences in each population may gradually lead to reproductive isolation if recombination across populations yields new, previously untested combinations of alleles. These are referred to as ‘Dobzhansky-Müller Incompatibilities’ and are often interpreted in terms of adaptive alleles being mixed onto ‘foreign genomic backgrounds.’

Speciation by DMIs.jpg — An example of how accumulated genetic differences can contribute to speciation through DMIs. The grey boxes represent the average genome of each population, from a single unedited ancestor to two divergent populations. Over time, each population independently acquires new mutations (in red and blue, respectively). After some time, reconnection between the two populations causes hybrids to form with bizarre and new combinations of these mutations. If mutations from Population A and Population B are incompatible (e.g. lead to infertility, physiological issues, etc.) then these hybrids may be inviable in the long term and reproductive isolation (RI) has been effectively achieved. Thus, it is not the individual mutations that cause RI but incompatible combinations of them from across uniquely evolving populations.

The most rapid examples include genomic duplication events, where a rare mutational event causes the entire genome (or sections of it) of an organism to duplicate: if this happened in humans, for example, you would go from having 2 copies of 23 chromosomes (46 total) to 4 copies of 23 chromosomes (92 total). Unsurprisingly, this messes with the internal biochemical mechanisms that allow embryos or reproductive cells (e.g. eggs, sperm) to form. This pattern of speciation has been particularly noted within some plant species (and also in salmon!).

Although dramatic, other particular changes within the genome can similarly initiate reproductive isolation. These often relate to genes that encode traits that are fundamental to the process of reproduction – such as the structure of sex cells or physical characteristics important in copulation – which may fundamentally prevent the formation of hybrids (or even the act of interspecific sex). Regions that are especially divergent between species, and which likely determined their reproductive isolation, like this are referred to as ‘islands of speciation.’ Trying to understand how, and why, islands of speciation form is critical in understanding the process of speciation through the diversity of life.

Islands of speciation — Single, or very few, parts of the genome can cause rapid and strong reproductive isolation through ‘islands of speciation.’ In this example, the yellow/red distributions represent a measure of differentiation between two different lineages of a fish. All it takes is strong differentiation at one very select part of the genome (the coloured box in the genome) to prevent the formation of hybrids, causing rapid RI between the two fish and leading to speciation.

Real ‘hybridised’ species?

Now, these concepts might seem highly theoretical only. How often do they really occur in nature? Well, there are some clear examples of complex mechanisms of speciation in various groups. For example, in some cases the formation of hybrids might actually contribute to new species being formed, weirdly enough.

Hybrid speciation

This can happen if a hybrid is, through one way or another, reproductively isolated from both of its parent species. Dubbed ‘hybrid speciation’, this phenomenon has been observed in arguably the most iconic system for studying speciation: Darwin’s finches! In 1981, a yet-to-mature finch was identified on the largest island of Daphne Major: genetic testing identified it as originating from another species on a different island more than 100km away (Geospiza conirostris, from Española). Now isolated from its home population, this new individual was forced to breed with another species (Geospiza fortis) to maintain a lineage. However, observing the pedigree of the resultant hybrids over generations showed that backcrosses with G. fortis were less fit than hybrids breeding together (i.e. incestuous!) in later generations. Despite the strong inbreeding, hybrids demonstrate unique characteristics such as bill shape and song. The hybrids – with their weird, mixed songs – were unable to successfully woo females of G. fortis and were physically separated from G. conirostris). Thus, hybridisation actually caused a new species to form (under the BSC).

Ring species

Other interestingly complex situations of gene flow and speciation include ‘ring species.’ This occurs when individuals across the distribution of a species show a gradient of differentiation: thus, the two opposite ends of the distribution are the most differentiated. However, in a ‘ring species’ scenario this gradient is not a linear pathway but a circular pattern: where one end of the ring meets the other, the two opposite ends interact. If enough genetic differences have accumulated along the ring, reproductive isolation may have been acquired and the two could be considered different species. This can be especially confusing since there is no clear ‘cut-off’ point throughout the ring and species are only clear at the meeting points of the two ends.