MacArthur's words still resonate 40 years on

I recently received an old library copy of “Geographical Ecology: Patterns in the Distribution of Species” by Robert MacArthur (1972). It’s the last book that MacArthur wrote before his early death to cancer. It is an ambitious book that connects repeated ecological patterns to mechanisms as broad as the earth’s rotations (producing climate as we experience it) and as focused as organismal behaviour.

But honestly, the thing that has struck me most so far as I read is the timelessness and wisdom in MacArthur's introduction. Issues ranging from focusing on questions versus systems, the value of repeated patterns, complexity, and what generality really means, aren't at all new.

“To do science is to search for repeated patterns, not simply to accumulate facts, and to do the science of geographical ecology is to search for patterns of plant and animal life that can be put on a map. The person best equipped to do this is the naturalist who loves to note changes in bird life up a mountainside, or changes in plant life from mainland to island, or changes in butterflies from temperature to tropics. But not all naturalists want to do science; many take refuge in nature’s complexity as a justification to oppose any search for patterns. This book is addressed to those who do wish to do science. Doing science is not such a barrier to feeling or such a dehumanizing influence as is often made out. It does not take the beauty from nature. The only rules of scientific method are honest observations and accurate logic. To be great it must also be guided by a judgment, almost an instinct, for what is worth studying. No one should feel that honest and accuracy guided by imagination have any power to take away nature’s beauty.

Science should be general in its principles. A well-known ecologist remarked that any pattern visible in my birds but not in his Paramecium would not be interesting, because, I presume, he felt it would not be general. The theme running through this book is that the structure of the environment, the morphology of the species, the economics of species behaviour, and the dynamics of population changes are the four essential ingredients of all interesting biogeographic patterns. Any good generalization will be likely to build in all these ingredients, and a bird pattern would only be expected to look like that of a Paramecium if birds and Paramecium had the same morphology, economics, and dynamics, and found themselves in environments of the same structure.”
--Robert MacArthur

It's interesting that an introduction written in 1972 is so relevant that it could have been written today. The pessimistic view is that ecology is just iterating through the same problems and solutions, or progress is slow. Or maybe classic books remain as classics because their authors understood and explored the issues at the core of the science and had the benefit of being there in the formative years. It's fun to see that when MacArthur thanks particularly four friends who influenced his work most, he means G. Evelyn Hutchinson, E.O. Wilson, Richard Levins, and Jared Diamond. I suppose any book influenced by the combination of all these scientists and written by MacArthur will always have something interesting to say. 

Evidence for the evolution of limiting similarity in diving beetle communities

Remi Vergnon, Remko Leijs, Egbert H. van Nes, and Marten Scheffer. (2013) Repeated Parallel Evolution Reveals Limiting Similarity in Subterranean Diving Beetles. The American Naturalist, Vol. 182, pp. 67-75.

In 2006, Marten Scheffer and Egbert van Nes published a very nice paper showing the outcome of simulated evolution of competing species. Their results showed how patterns of evenly-spaced clusters of species along a niche axis could evolve to minimize competition via limiting similarity. 

From Scheffer and van Nes (2006): Evenly spaced clusters of species along a niche axis (x-axis) evolved in response to competition.

Within any cluster along the niche axis, species tended to be more similar than expected. The results suggested that complex self-organizing clustered patterns might result from simple competitive limitations. Interestingly, although the original paper suggested that clustered patterns in size distributions are common, only now are these theoretical expectations about the evolution of limiting similarity being tested with data. In fact, though theory has long suggested that patterns of limiting similarity should evolve to allow coexistence between competing species, empirical evidence is rather lacking. Despite this, limiting similarity and competition are staples of ecological thought: for example, patterns of overdispersion in traits or relatedness are often used as evidence for the importance of competition.

The follow-up paper -Vergnon et al. (2013)- tests for the pattern predicted in Scheffer and van Nes (2006) using communities of subterranean diving beetles (Coleoptera, Dytiscidae) in Australia. These species have evolved for over 5 million years in isolated aquifers. If limiting similarity structured beetle communities, the authors predicted that there should be regularity in the spacing of species along a niche axis. If competitive interactions determine species' positions on the niche axis, then their absolute positions on the niche axis could vary between communities so long as their relative positions are evenly spaced. If, in contrast, niches are driven by environmental conditions, species in different communities/aquifers should have similar absolute positions along the niche axis.

The authors used a nice combination of statistics, modelling and observational data (34 communities of beetles representing 75 total species) to test for these predicted patterns. They used beetle size as the measure of niche position, since size is often an indicator of niche position and food availability and identity. For almost all aquifers, co-occurring beetles were significantly different in size. Further, species in different aquifers classified as occurring in the same size classes (small, medium, large), had different absolute sizes (i.e. the largest beetle in one 2-species aquifer was not similar in size to the largest beetle in another 2-species aquifer).  

From Vergnon et al. (2013): Absolute sizes of diving beetles in aquifers with 3 species present. The absolute size in a size class (large - black; medium - white; small - grey) varies between aquifers.

Although the absolute size of species differed between aquifers, the ratio of sizes (regularity of spacing on the niche axis) was highly consistent. Further, simulations of evolution of body size due to competition were capable of reproducing the observed size structure of the diving beetles.

From Vergnon et al. (2013): regularity of spacing between competing diving beetles (measured as the body size ratio). 

This paper does a nice job of integrating theory and data, and combining pattern and process. The focus is on testing contrasting predictions, and the authors use complementary approaches to test statistically for the presence of patterns and to demonstrate with simulations the relationship between the evolution of limiting similarity and the observed pattern. The evidence is suggestive that limiting similarity and not pre-existing environmental niches explains the size structure of communities of competing diving beetles. There are still questions about how far these inferences can be extended. For example, do we expect that predefined environmental niches are really the same across aquifers? How important is competition in these communities - at the moment, the authors only have minimal evidence of gut content overlap from a single aquifer. Further the low diversity of aquifer communities (~1-5 diving beetle species) means that the prediction of clusters of multiple similar species made in the original Scheffer and van Nes paper can't be tested. But the fact that aquifer diving beetle communities have low diversity and are very simplistic is beneficial for the authors. Patterns in diverse communities where multiple processes (predation, migration, etc) are important may be too complex to show clear evidence in observational data. Simple systems (including microcosms) are a good place to find evidence that a process of interest actually occurs. Whether or not that process is important across many systems is of course a more difficult question to answer. 

Movement patterns in populations of early academics

Sometimes of the perks of academic life are also the most difficult parts – frequent travel opportunities mean you are also frequently away from friends and family (and spend too much time in airports). The nature of the university job market provides global opportunities for work, but also means that in reality opportunities and circumstances can constrain you to places you wouldn't have chosen otherwise. Your friends will cover the world, but you will rarely be in the same room together. The apprenticeship-like nature of early academic positions means that you will move, probably many times, before you find a permanent position (if you do). 

I have a friend who grew up with diplomat parents, which meant her family moved to a new place in the world every few years. The result was that she often felt like she didn’t have a strong connection to any one place or group of people. Academia isn’t quite so extreme, but you can understand why after moving to one place for undergrad, another for a Masters and/or PhD, one or two more for postdocs, your interactions and place in the world can feel rather impermanent. It also means that, for better or worse, your social circle includes other academics, and they are also shifting from place to place. When I tell non-academic friends and family (who mostly have settled in a single place) about upcoming moves, they are often more excited than I am about the opportunity to pick up and go. No doubt this is a grass-is-always-greener situation, but I often think that the most notable and difficult aspect of academic mobility is that you end up saying goodbye a lot.

I wonder whether some of the academic ambivalence expressed is aggravated by this early, necessary transience. Certainly there is lots of evidence that residential mobility (i.e. moving) relates to higher mortality and lowered health indicators, though some studies suggest that this effect may be more true for introverts than extroverts (presumably because extroverts form new friendships more easily). Academics share this phenomenon with groups like military families and third culture kids. The commonality is that, with every move it becomes harder to define home as a particular place – it is more like an intangible connection to multiple places and people. And maybe that's not so terrible - a good friend who was raised by an academic suggested that the key is to redefine your life and friendships as being global rather than local. And eventually professors settle down (I can think of a few people who have been at one university for 30+ years). But in the interim there is always the not-insignificant tension between the costs and benefits of uprooting yourself every few years, and the slow loss of individuals who are not capable of this mobility, from the academic pipeline.

Felsenstein for SMBE president :)

A highly entertaining, somewhat relevant update to my post about Joe Felsenstein's 'dishonour roll'. Felsenstein is running for President of the Society for Molecular Biology and Evolution, and his personal statement is a must-read career retrospective. If you don't at least crack a smile, you might be taking science a bit too seriously...

For example: "[Felsenstein] has been President of the Society for the Study of Evolution, and imagines that he could be President of the SMBE, even though he has not yet learned the names of all 20 amino acids."

Honestly, I think it gives more insight than most bios into the person and their work.

Another round in Diamond vs. Simberloff: revisiting the checkerboard pattern debate

Edward F. Connor, Michael D. Collins, and Daniel Simberloff. 2013. "The Chequered History of Checkerboard Distributions." Ecology. http://dx.doi.org/10.1890/12-1471.1.

One of the most vociferous recent debates in community ecology started in the 1970s between Jared Diamond and Dan Simberloff (and colleagues) regarding whether 'checkerboard patterns' of bird distributions provided evidence for interspecific competition. This was an early and particularly heated example of the pattern versus process debate that continues in various forms today. Diamond (1975) proposed that the distribution of birds in the Bismark Archipelago, and particularly the fact that some pairs of bird species did not co-occur on the same islands (producing a checkerboard pattern), was evidence that competition between species limited their distributions. The issue with using this checkerboard pattern as evidence of competition, which Connor and Simberloff (1979) subsequently pointed out, was that a null model was necessary to determine whether it was actually different from random patterns of apparent non-independence between species pairs. Further, other mechanisms (different habitat requirements, speciation, dispersal limitations) could also produce non-independence between species pairs. The original debate may have died down, but the methodology for null models of communities suggested by Connor and Simberloff has greatly influenced modern ecological methods, and continues to be debated and modified to this day.

The original null model of bird distributions in the Bismark Archipelago involved a binary community matrix (rows represent islands, columns represent species) with 0s and 1s representing species presences or absences. Hence, all the 1s in a row represent the species present on the island. The original null model approach involved randomly shuffling the 0s and 1s, maintaining island richness (row sums) and species range sizes (column sums). The authors of a new paper in Ecology admit that the original null models didn’t accurately capture what Diamond meant by a "checkerboard pattern". This is interesting in part because two of the authors (E.F. Connor and Dan Simberloff) lead the debate against Diamond and introduced the binary matrix approach for generating null expectations. So there is a little bit of a ‘mea culpa’ here. The authors note that earlier null models captured patterns of non-overlap between species' distributions but didn’t differentiate between non-overlap between species with overlapping ranges compared to non-overlap between species which simply occurred on sets of geographically distant islands (referred to here as 'regional allopatry'). The original binary matrix approach didn’t consider spatial proximity of species ranges.

With this fact in mind, the authors re-analyzed checkerboard patterns in the Bismark Archipelago, but in such a way as to control for regional allopatry. True checkerboarding was defined as: “a congeneric or within-guild pair with exclusive distribution, co-occurrence in at least one island group, and geographic ranges that overlap more or significantly more than expected under an hypothesis of pairwise independence”. This definition appears closer to Jared Diamond's original definition and so a null model that captures this is probably a better test of the original hypothesis. The authors looked at the overlap of convex hulls defining species’ ranges and when randomizing the binary matrix, added the further restriction that species could occur only within the island groups where they were actually found (instead of being randomly shuffled through any island, as before).

Even with these clarified and more precise null models, the results remain consistent. True checkerboarding appears to rarely occur compared to chance. Of course, this doesn't mean that competition is not important, but “Rather, in echoing what we said many years ago, one can only conclude that, if they do compete, competition does not strongly affect their patterns of distribution among islands.” More generally, the endurance of this particular debate says a lot about the longstanding tension in ecology over the value and wealth of information captured by ecological patterns, and the limitations and caveats that come with such data. There is also a subtle message about the limitations of null models: they are often treated as a magic wand for dealing with observed patterns, but null models are limited by our own understanding (or ignorance) of the processes at play and our interpretation of their meaning. 

The slippery slope of novelty

Coming up with a novel idea in science is actually very difficult. The many years during which smart people have, thought, researched, and written about ecology and evolution means that there aren’t necessarily many easy openings remaining. If you are lucky (or unlucky) enough to know someone with an encyclopedic knowledge of the literature, it becomes quickly apparent that only rarely has an idea not been suggested anywhere in the history of the discipline. Mostly science results from careful steps, not novel leaps and bounds. The irony is that to publish in a top journal, a researcher must convince the editor and reviewers that they are making a novel contribution.

There are several ways of thinking about the role of the novelty criterion - first, the effect it has had on research and publishing, but also more fundamentally, how difficult it is to even define scientific novelty in practice. Almost every new student spends considerable effort attempting to come up with a completely "novel" idea, but a strict definition of novelty – research that is completely different than anything published in the field in the past - is nearly impossible. Science is incrementally built on a foundation of existing knowledge, so new research mostly differs from past research in terms of scale and extent. Let's say that extent characterizes how different an idea must be from a previous one to be novel. Is neutral theory different enough from island biogeography (another, earlier, explanation for diversity which doesn’t rely on species-differences) to be considered novel? Most people would suggest that it is distinct enough as to be novel, but clearly it is not unrelated to works that came before it. What about biodiversity and ecosystem functioning? Is the fact that its results are converging with expectations from niche theory (ecological diversity yields greater productivity, etc) take away from its original, apparent novelty? 

Then there is the question of scale, which considers the relation of an new idea to those found in other disciplines or at previous points in time. For example, when applying ideas that originate in other disciplines, the similarity of the application or the relatedness of the other discipline alters our conclusions about its novelty. Applying fractals to ecology might be considered more novel than introducing particular statistical methods, for example. Priority (were you first?) is probably the first thing considered when evaluating scientific novelty. But ideas are so rarely unconnected to the work that came before them, so then we evaluate novelty as a matter of degree. The most common value judgment seems to be that re-inventing an obscure concept first describe many years ago is more novel than re-inventing an obscure concept that was recently described.

In practice, then, the working definition of novelty may be that something like ‘an idea or finding doesn't exist the average body of knowledge in the field’. The problem with this is that not everyone has an average body of knowledge – some will be aware of every obscure paper written 50 years ago, and for them nothing is novel. Others have a lesser knowledge or more generous judgement of novelty and for them, many things seems important and new. A great deal of inconsistency in the judgement of papers for a journal with a novelty criterion results simply from the inconsistent judgement of novelty. This is one of the points that Goran Arnqvist makes in his critique of the use of novelty as a criterion for publishing (also, best paper title in recent memory). Novelty is a slippery slope. It forces papers to be “sold” and so overvalues flashy and/or controversial conclusions and undervalues careful replication and modest advances. And worse, it ignores the truth about science, which is that science is built on tiny steps founded in the existing knowledge from hundreds of labs and thousands of papers. And that we've never really come up with a consistent way to evaluate novelty.

(Thanks Steve Walker for the bringing up the original idea)