Search

The Janzen-Connell (JC) hypothesis, one of the most influential hypotheses explaining forest diversity, is inconsistent with evidence that tree species share the same natural enemies. Through the discussion of seedling diseases from a pathogen-centered perspective, we expand the JC hypothesis to tie in host-pathogen-environment interactions at three levels: local adaptation, host specificity of the combined effect of multiple infections, and environmental modulation of disease. We present evidence from plant pathology, disease ecology, and host-parasite evolution relevant to (but not commonly associated with) forest species diversity maintenance. This expanded view of the JC hypothesis suggests ways to direct new experiments to integrate research on pathogen local adaptation, co-infection, and environmental effects on infection by using high-throughput molecular techniques and statistical models.

The Unified Neutral Theory of Biodiversity (UNTB), proposed as an alternative to niche theory, has been viewed as a theory that species coexist without niche differences, without fitness differences, or with equal probability of success. Support is claimed when models lacking species differences predict highly aggregated metrics, such as species abundance distributions (SADs) or species area distributions (SARs). Here, I summarize why UNTB generates confusion, and is not actually relevant to niche theory (i.e. an explanation for why and how many species coexist). Equal probability is not a theory, but lack of one; it does not include or exclude any process relevant to coexistence of competitors. Models lacking explicit species can make useful predictions, but this does not support neutral theory. I provide s suggestions that could help reduce confusion generated by the debate.

Biodiversity science is unusual in that an emerging paradigm is not based on a specific process, but rather depends largely on stochastic elements, perceived as neutral forces. Here I suggest that these forces, which have been justified, in part, by the concepts of symmetry and equalizing mechanisms, have application to the understanding of stochastic models but do not constitute forces that operate in nature. Another process now regularly classified as a neutral force, limited dispersal, represents a fundamental demographic process that is not neutral with respect to species differences, but rather differs among species in important ways. Finally, I suggest that the dramatic shift in ecological research to focus on neutrality could have a cost in terms of scientific understanding and relevance to real biodiversity threats.

Together, graphical models and the Bayesian paradigm provide powerful new tools that promise to change the way that environmental science is done. The capacity to merge theory with mechanistic understanding and empirical evidence, to assimilate diverse sources of information and to accommodate complexity will transform the collection and interpretation of data. As we discuss here, we specifically expect a shift from a focus on simple experiments with inflexible design and selection among models that embrace parts of processes to a synthesis of integrated process models. With this potential come new challenges, including some that are specific and technical and others that are general and will involve reexamination of the role of inference and prediction.

Halley and Iwasa and I agree that simple models can be valuable [1xNeutral theory as a predictor of avifaunal extinctions after habitat loss. Halley, J.M. and Iwasa, Y. Proc. Natl. Acad. Sci. U.S.A. 2011; 108: 2316–2321CrossRef | PubMed | Scopus (28)See all References, 2xThe coherence problem with the unified neutral theory of biodiversity. Clark, J.S. Trends Ecol. Evol. 2012; 27: 198–202Abstract | Full Text | Full Text PDF | PubMed | Scopus (28)See all References]. Their fine study shows that a model can ignore factors contributing to extinction (habitat, dispersal, life history, physiology, exotic enemies, human activity) and still describe species losses from islands or patches of different size. Their letter [3xNeutrality without incoherence: a response to Clark. Halley, J.M. and Iwasa, Y. Trends Ecol. Evol. 2012; 27: 363Abstract | Full Text | Full Text PDF | PubMed | Scopus (5)See all References][3] acknowledges our shared agreement on the utility of simple models. We part company on what this has to do with the Unified Neutral Theory of Biodiversity.Given the issues I raised in [2xThe coherence problem with the unified neutral theory of biodiversity. Clark, J.S. Trends Ecol. Evol. 2012; 27: 198–202Abstract | Full Text | Full Text PDF | PubMed | Scopus (28)See all References, 4xBeyond neutral science. Clark, J.S. Trends Ecol. Evol. 2009; 24: 8–15Abstract | Full Text | Full Text PDF | PubMed | Scopus (70)See all References], which they objected to in [1xNeutral theory as a predictor of avifaunal extinctions after habitat loss. Halley, J.M. and Iwasa, Y. Proc. Natl. Acad. Sci. U.S.A. 2011; 108: 2316–2321CrossRef | PubMed | Scopus (28)See all References][1], in the Cell Press Discussion on neutral theory (http://news.cell.com/discussions/trends-in-ecology-and-evolution/ecological-neutral-theory-useful-model-or-statement-of-ignorance) and again in their letter, I would expect clarification on how their study supports the Unified Neutral Theory of Biodiversity. In both [2xThe coherence problem with the unified neutral theory of biodiversity. Clark, J.S. Trends Ecol. Evol. 2012; 27: 198–202Abstract | Full Text | Full Text PDF | PubMed | Scopus (28)See all References][2] and [4xBeyond neutral science. Clark, J.S. Trends Ecol. Evol. 2009; 24: 8–15Abstract | Full Text | Full Text PDF | PubMed | Scopus (70)See all References][4] I addressed the many definitions, including that species are functionally equivalent, that all have the same fitness or that all have the same probability of recruitment success. I addressed the belief that the Unified Neutral Theory of Biodiversity is a null model for niche differences or that it represents an alternative to niche or fitness differences. I demonstrated why neutral models do not assume that all individuals obey the same rules of engagement, and why they do not predict that coexistence occurs in the absence of niches. This list of motivations, assumptions, and interpretations may not be exhaustive, but it is close.The statement in the letter by Halley and Iwasa most relevant to Neutral Theory does not respond to the question of how their study supports it:It is not necessary to buy into any interpretation of the niche (or the stronger claims of unification) to apply these mechanisms and get useful results that fit the data well.Halley and Iwasa seem to be saying that it does not matter what niches are. One of the few persistent claims about Neutral Theory has been that it is an alternative to niche theory. Since Neutral Theory does not address niches or lack thereof [5xResolving the biodiversity debate. Clark, J.S. et al. Ecol. Lett. 2007; 10: 647–662CrossRef | PubMed | Scopus (98)See all References][5], proponents now simply avoid saying what a niche is. So how does a study such as [3xNeutrality without incoherence: a response to Clark. Halley, J.M. and Iwasa, Y. Trends Ecol. Evol. 2012; 27: 363Abstract | Full Text | Full Text PDF | PubMed | Scopus (5)See all References][3] support it? What definition of Neutral Theory does it support?My paper that Halley and Iwasa object to in their letter [2xThe coherence problem with the unified neutral theory of biodiversity. Clark, J.S. Trends Ecol. Evol. 2012; 27: 198–202Abstract | Full Text | Full Text PDF | PubMed | Scopus (28)See all References][2] focuses on the need to move beyond claims that the Unified Neutral Theory is being criticized for reliance on simple models. From introductory statistics courses onward we learn to construct simple models and penalize large ones. Nobody wants a complex model when a simple one will suffice. There is no debate about this. The criticism of Neutral Theory concerns vague and shifting definitions and misinterpretation of models.The Unified Neutral Theory finally appears to be no more than a defense of simple models. If proponents now acknowledge that it has nothing to say about niche differences [6xThe case for ecological neutral theory. Rosindell, J. et al. Trends Ecol. Evol. 2012; 27: 203–208Abstract | Full Text | Full Text PDF | PubMed | Scopus (56)See all References][6], the Unified Theory defaults to a statement that simple models can be useful, apparently the point of [3xNeutrality without incoherence: a response to Clark. Halley, J.M. and Iwasa, Y. Trends Ecol. Evol. 2012; 27: 363Abstract | Full Text | Full Text PDF | PubMed | Scopus (5)See all References][3]. All ecologists – including those who do not count themselves as theorists – should be concerned when such basic modeling principles are taken as a unified theory for our discipline.

This, the July issue of TREE, is the second of two issues commissioned to celebrate the 20th birthday of the journal. The people shown on the cover of the issue are a sample of those who will shape our science in the future. They are a non-random sample: we posted a call on two widely distributed e-circulation lists for pictures of graduate students and young post-docs working in ecology and evolution today. Several established figures accused us of ageism and others tried to slip themselves in (including a C. Darwin, who submitted the only painted portrait). In the end, we received ∼375 pictures of people who looked like they had the chronological potential to be driving our field in 20 years. Not all could appear, so we left the final cut to the graphic designer. We hope the cover will at least generate some mirth in 20 years time.But what will ecologists and evolutionary biologists be doing 20 years from now? Predicting science is for the foolhardy, but current opportunities are easier to spot. There is clearly much scope for new work in all areas of current activity: it is hard to identify an area active now that might be solved and abandoned in two decades. To the openings discussed in many of the birthday articles, we add the following more general observations.Destruction, degradation and disasters all provide scientific opportunities. Two decades ago, ecologists actively debated the relative merits of observations versus small-scale experiments, and the need for simple models. The various approaches are still with us, but these are already being enriched by new attention to natural and unintended human-caused ‘experiments’ that provide insights at relevant scales. Invasive species enable us to observe species interactions in the context of climate and soil variation and in situ food webs. Changes in atmospheric CO2 concentrations have provided canopy-scale context for insights gleaned from small chambers, revealing why simplistic interpretations might not apply to productivity on a rapidly changing Earth. Climate change will soon be revealing how fast species can migrate. Chernobyl makes it possible to look at the importance of genetic perturbations in real ecosystems (and whether human occupancy of an area is worse for wildlife than is radiation). Vaccination enables an analysis of the selective removal of part of a biota. And, sadly, habitat loss is already enriching our understanding of demography and genetics at low population densities.People calling themselves systems biologists are emerging in all branches of biology, excited about the prospect of iterative dialogue between mathematical models and huge amounts of molecular and physiological data. Ecologists and evolutionary biologists have been systems biologists for at least 40 years, albeit with different data. Established systems biologists have tremendous opportunities for contributing to the study of suborganismal systems, not least because they understand that mastering complexity is not simply a matter of more data or bigger computers. Inevitably, the new systems biologists will also manage to merge new models and vast data streams in novel ways that analysts of ecological systems will surely be able to exploit. And, ultimately, suborganismal mechanistic function is of interest because of what it does at an organismal or even ecosystem level. Yet the analysis of genotype × environment interactions, something that many TREE readers specialise in, is experimentally demanding, and beyond the experience (and even current interest) of most molecular and cell biologists.Biologists working at suborganismal scales continue to reveal a swathe of natural history that receives far less attention from evolutionary biologists and ecologists than is deserved. For instance, the natural history of mammalian immunology continues to be understood in ever more intricate detail. Yet there is no quantitative understanding of the population biology of interacting cell types and no predictive explanation for why the immune system is designed as it is. Much human misery is caused by immunopathology [1xEvolutionary causes and consequences of immunopathology. Graham, A.L. et al. Annu. Rev. Ecol. Evol. Syst. 2005; 36: 337–397Crossref | Scopus (154)See all References[1], yet we have no understanding of why our immune genes often harm us and, indeed, whether immune self-harm occurs in anything other than us and laboratory mice. Analogous questions apply to most areas of suborganismal physiology and cell biology.Indeed, it is striking that the attention that ecological and evolutionary biologists pay to something seems to be inversely related to the amount of mechanistic detail known about it. Is this because we have been slow to wade into the jargon-laden arena of biomedicine? Or is it because our theories do not work well when there are lots of facts? Becoming involved in biomedicine requires a detailed understanding of the experimental techniques and the jargon involved, a genuine respect for biological reality, and an ability to sort signal from noise. For ecologists and evolutionary biologists who can do this, enormous potential exists.A tension in some areas of organismal biology, which is reflected in several of the birthday articles, concerns the use of model organisms. Model organisms that are well understood by mechanistic biologists (the odd bacterial, yeast, fly, worm, rodent or weed species) have, in some cases, been used with great success to address issues in ecology and evolution, and there is undoubtedly considerably more mileage to be gained that way. However, model organisms are not typically studied in their ecological context (indeed, most have essentially no known ecology outside a laboratory container). By contrast, little mechanistic detail is known about organisms that are well studied in the wild. With technological advances, it will be increasingly possible to carry out detailed mechanistic and genetic work on non-model organisms in nature, ultimately weakening the case for studying ecology and evolution in jars. For instance, it should be soon feasible to study bacterial evolution in great detail in contexts where they cause disease. And interesting genetic polymorphisms identified in mouse and human studies could be subjected to detailed ecological genetic analyses in field populations of wild animals with known genealogies. A detailed understanding of selection on such polymorphisms has to move out of the lab or away from the constraints of a human study. However, one could argue that some areas of our work could make less use of wild, non-model organisms, and work more on laboratory models. It would be good to resolve quantitatively the evolution of sex for at least one organism.Need-driven research offers huge opportunities, not least because ecologists and particularly evolutionary biologists have been slow to fully address real world problems. For instance, had SARs persisted in the human population, would it have evolved to be more or less virulent for humans? We have no quantitatively successful explanations of virulence change for any disease. Over one million people die each year of diseases transmitted by insects but most TREE-reading entomologists do not work on these species. And, with a few notable exceptions, we have almost entirely left the evolution of drug resistance (one of the basic challenges of 21st-century medicine, and one of the prime examples of evolution in real time) in the hands of people who have no formal training in either evolution or population biology.Finally, there are enormous opportunities for communicating beyond our community. Perhaps more than ever, there is a need to get our science across in an accessible, jargon-free way that society can use. Climate change and evolution are now household words, no longer viewed as hypothetical possibilities, irrelevant to ordinary lives. Society often has strong views on our science, or wants to know about it, or has no views about things that it ought to. More than ever in the history of our subject, there are important opportunities in teaching, popularising and policy.Our subject is now so large that this monthly review journal flourishes. But a danger with large disciplines is that those in them keep busy talking only with each other. We hope that a substantial proportion of the cohort represented on the cover of this issue will consider looking outward as their careers develop. For those prepared to get out a bit, there are very real opportunities to make a difference to our discipline, to other sciences, and to society as a whole.