Skip to content

Ecological speciation

Last updated on 18 Sep 2017

ResearchBlogging.orgA new paper in the Journal of Evolutionary Biology (link below) assesses the possibility of speciation by ecological differentiation in conjunction with geographical isolation. The interesting thing here is that it takes two views previously considered as antagonists and combines them into a single model.

“Standard” views of speciation assume that a population has to be divided geographically in order to adapt locally without being swamped by interbreeding – these are called “allopatric” models (I previously summarised these views here). Sympatric models, which posit diversification by local selection on ecotypes, are regarded by allopatrists as unfeasible or impossible, depending who you ask (Sergey Gavrilets, for example, showed that sympatric speciation is mathematically possible. Now the question is when those conditions obtain). The authors of this paper make an obvious point, in retrospect: ranges are never stable. Ecotypes track and are partitioned by geographical changes, including geological barriers, climatological changes, and the effects of other species (such as a particular tree, or an animal that affects vegetation as African elephants do).

Over time, these cause repeated allopatry and sympatry for a single population, so the authors ask what that would imply for speciation by ecological selection. It’s a good question to ask. Since we cannot do experimental observation for processes that would, of necessity, involves hundreds of thousands of years for macroorganisms, they have done simulations based upon some mathematical assumptions. This “experimental modelling” is very common in evolutionary biology. The usual concern is to what degree we can accept the assumptions built into the model – that is, ask when the conditions actually obtain. Still, a proof of concept is interesting and will prove out our other assumptions in the debate over speciation.

They consider two scenarios: a weak tradeoff and a strong tradeoff for the exploitation of two resources, distributed asymmetrically over the two ranges. The weak version leads to specialisation and maintenance of adaptive diversity. Strong tradeoffs lead to the evolution of generalist traits – the ability to exploit both resources (but with lesser efficiency).

I can’t speak to the math, being imatherate, but it looks like some of the usual assumptions in the philosophical debates over speciation are unwarrantedly simplistic. The paper is behind a paywall, so you need to have university access.

Later note: reader Bjørn notes that I misread the paper. Here are his comments:

You write that “The weak version leads to specialisation and maintenance of adaptive diversity. Strong tradeoffs lead to the evolution of generalist traits – the ability to exploit both resources (but with lesser efficiency),” but that’s the opposite of what they find. They write:

“… under stabilizing selection:Under a weak trade-off, diversity can be maintained at a secondary contact if reproductive isolation has evolved in allopatry and is maintained (i.e. if allopatric speciation has occurred). In our model with an evolving mating trait,reproductive isolation is equivalent to assortative mating between ecologically differentiated individuals. Under a weak trade-off, assortative mating is selected neither in allopatry nor at secondary contact.
(…)

Under disruptive selection in sympatry (strong trade-off case), we showed that landscape dynamics allow ecological divergence.”

The logic is that the trade-offs forces specialization, strong trade-offs lead to the evolution of specialist traits.I do wish they had been more explicit about this, and that they had explored the prob. of diversity maintenance as afunction of z – the trade-off parameters.

Aguilée, R., Lambert, A., & Claessen, D. (2011). Ecological speciation in dynamic landscapes Journal of Evolutionary Biology DOI: 10.1111/j.1420-9101.2011.02392.x

13 Comments

  1. joe joe

    The preprint is available here.

    P.S.: Have you seen the target review in 2001 by Chung-I Wu “The genetic view of the process of speciation”? I remember that as qute interesting.

    • Yes, I cite that in my book. But Wu seems to think there are crucial genes involved in speciation – it is not clear that he doesn’t think they are the same genes for all (animal) species. That is simply implausible to me.

  2. Raving Raving

    …without being swamped by interbreeding … Sympatric models, which posit diversification by local selection on ecotypes , are regarded by allopatrists as unfeasible or impossible, depending who you ask …

    Before setting out on a career of modelling spatio-temporal population dynamics, I gave some thought to encouraging development of a new species in a laboratory environment.

    “Boost the rate of mutation and minimize selective pressure”

    • I think you are also going to have to do something about recombination there raving. Adding new variants is one thing, but by itself it will just increase the genetic variance of a single population – since all the new alleles being made are broken up in each generation.

      FWIW relaxed selection was part of “founder flush” models of speciation, but they are now generally considered in the way sympatric models once were 🙂

      • Raving Raving

        Given the specific context and also 25 years ago, I should have been aware of FF. Thanks for letting me know about it.

        Looked at your blog and understand where you are coming from. Nice to see the debate rage on by taking a setback overview.

        Fitness and population dynamics explore consistency of presumed causal mechanisms for empirically supplied end-targets. Attacking natural selection is as denying WYSIWYG.

        Accepting WYSIWYG acknowledges the mundane. Describing the peculiar is more natural for a biological perspective.

        • Raving, your point about increasing the rate of mutation is right on. I have re-analyzed Wright’s finite island model using new mathematics, and I found that the usual analysis has been wrong on major points. The usual analysis concludes that the absolute number of migrants determines whether two or more subpopulations evolve together as a unit or diverge. If there is more than one migrant per generation entering each subpopulation, divergence will not evolve. This analysis is based on Fst as the measure of divergence. But it turns out Fst and its relatives have no relation to divergence (Jost 2008 Molecular Ecology). Using a real measure of divergence or differentiation, I proved that divergence is controlled by the relative migration rate divided by the mutation rate. An increase in the mutation rate by a factor of x is equivalent to a decrease in migration rate by the same factor.
          See the Molecular Ecologist blog, where I post simulations (in the Comment section) confirming that this is what controls divergence:
          http://www.molecularecologist.com/2011/03/should-i-use-fst-gst-or-d-2/#more-662
          Lou

  3. Interesting looking paper,

    I’m not sure that people have completely ignored the way the distribution of populations change. Even in terms of determining if sympatric speciation has happened, Coyne and Orr (2004) [how many times have I typed those letters!] say “the biogeographical history of the species must make a past allopatric phase unlikely” – which I always read as meaning “given the geological and climatic history of this place, might they have been allopatric once”.

    Similarly, lots of ‘ecological’ speciation is presumed to have started with geographical isolation followed by secondary contact and inter-‘species’ competition.

    But perhaps I should just read the paper 🙂

    • One of the issues I have with the way the debate has been framed is that people have set allopatric and sympatric speciation against each other. Clearly a realistic account treats these as alternate tendencies (and I think even possibly simultaneously occurrent – what happens to the sympatric portion of a metapopulation can affect what happens when allopatric portions meet once more). There used to be (perhaps it has gone now) a tendency to single-cause explanations, which were largely a matter of what the boundary conditions of the model used happened to be.

      Toulmin once characterised the catastrophism debate as a sequence of convergence of claims over time separated by differences in vocabulary. I think a lot of this is true in speciation and evolutionary theory in general.

  4. Jim Thomerson Jim Thomerson

    A quibble. The late Louis Rivas coined the term “Syntopic” for situations where members of two species encounter each other and have the possibility of interbreeding. Sympatric means that ranges on a map are the same, but does not necessarily imply syntopy. Rainbow trout and pine trees are sympatric, but not, (usually) syntopic.

  5. John S. Wilkins: There used to be (perhaps it has gone now) a tendency to single-cause explanations, which were largely a matter of what the boundary conditions of the model used happened to be.

    Nature versus nurture, anyone?

    No? Very good…

Comments are closed.

Optimized by Optimole