Monday, November 24, 2014

PITCHFORK SCIENCE: Guppies, Stickleback, and Darwin’s Finches.

[This is a cross-posting with the EXEB blog at Lund - thanks for the reciprocal opportunity Erik. And thanks for your earlier post here on Eco-Evo-Evo-Eco.]

I study Trinidadian guppies, threespine stickleback, and Darwin’s finches, surely 3 of the top 10 evolutionary biology “model” systems - for vertebrates at least. I thus fall at one extreme (or is it three extremes?) on the “pick a model system and use it to answer my question” versus “develop a brand new system all my own” continuum. Many students and postdocs find themselves facing their own decisions about where to position themselves along this continuum. Should they take the shortcut of working with an established system so they don’t have to work out the simple details and can get right to addressing the big general questions? Or should they forge their own path and become an expert in something brand new? It might seem, based on the above listing, that I consciously took the first approach but the reality is something quite different. In truth, I used a “follow your nose” coincidence-and-serendipity approach to study system choice. I here trace my own personal history in these research areas before closing with some general thoughts on how to choose a study system.

Why I study salmon: a 16 year old me with a steelhead from our cabin on the Kispiox River (BC, Canada).
I worked on salmon for my MSc and PhD, largely because I grew up with salmon fishing as my primary passion. Thus, I began studying salmon simply because I liked them and liked fishing for them. This led me to choose an institution (University of Washington - UW), department (School of Fisheries), and supervisor (Tom Quinn) ideally suited to immerse myself in salmon work. As my graduate work progressed, I very gradually became more and more interested in general questions in ecology, largely through exposure to the research of other people in the department. I even started subscribing to Ecology in addition to – of course – Fisheries. Yet my thinking remained salmon-centric: “what can ecology tell me about salmon”? Nothing wrong with that, of course. Then, when visiting home for Christmas in 1994, my mother gave me a book: The Beak of the Finch by Jonathan Weiner. When your Mom gives you a book for Christmas and you then spend the next week at home… well, you better read it.

The laboratory for my PhD: Lake Nerka, Wood River, Alaska.
The book was amazing. It described in wonderfully readable prose the research of Peter and Rosemary Grant on Darwin’s finches in the Galapagos Islands. What struck me the most, while reading beside the heater vent looking out at the blowing snow and -40 C weather (literally!), was Jonathan’s description of how the Grants had documented generation-by-generation rapid evolution of finch beaks in response to natural selection resulting from environmental change. Wow – you can actually study evolution in real time! It was my own eureka moment and, in short order, I became captivated by the idea. As soon as I got back to UW after Christmas, I went to the library and photocopied EVERY paper on Darwin’s finches (ah, libraries and photocopying – the good [and bad] old days). From then on, almost as though my brain had achieved an alternative stable state, my thinking was inverted to become: “What can salmon tell me about evolution?” 

My MSc and PhD work focused on sockeye salmon - this one in Knutson Bay, Lake Iliamna, Alaska.
Salmon did tell me a lot about evolution. I even edited a book (Evolution Illuminated, with one of my evolutionary idols, Steve Stearns) about merging evolutionary theory and salmon research. However, when one starts focusing on a topic (evolution) rather than an organism (salmon), one starts to become irked by aspects of the organisms that are not optimal for addressing the topic. Most notably, it is very hard to do experiments with salmon unless you have lots of water, lots of space, and lots of time. So, when thinking about a postdoc, I started talking to folks about which systems might allow me to better address basic evolutionary questions. I ended up moving in two directions.

The laboratory for my first postdoc. For more than a month of glorious weather, I camped on a small island in a small lake (Mackie Lake) at the end of a 4-wheel drive road. Those are my mesocosms floating in the lake and projecting from the island.
The first was the University of British Columbia (UBC) – because I didn’t want to go too far from my girlfriend (now wife) who was still at UW. I visited UBC and went from prof to prof telling them of my interest in a basic evolutionary question – the balance between divergent selection and gene flow – and asking if they knew of a system that would be good for testing my ideas. Many great suggestions were made, but Rick Taylor insisted he had the perfect system: Misty lake-stream stickleback – and he was right. So I started working on stickleback not because they were a model system, but because someone suggested they would be well-suited for my question and because it let me stay reasonably near my sweetheart.

A threespine stickleback guarding his nest.
The second direction came about through a conversation with Ian Fleming, who suggested that I should work with David Reznick on guppies. I hadn’t even considered this possibility, but I knew a bit about the system (which is also described in The Beak of the Finch) and it seemed cool. So I went to UCR and met with David and talked about how we might use guppies to study the interaction between selection and gene flow. David said he would be happy to help me with this work but that he didn’t have any money for me – and so I offered to write a full NSF proposal. I was just gearing up to do so when I heard that I had received an NSERC (Canada) postdoctoral fellowship to work with Rick Taylor on the Misty system – so off I went to stickleback, leaving guppies behind.

My favorite wild guppies captured in my first year of sampling, 2002.
UBC was great, an outstanding place for nurturing interest and insight into general questions in evolutionary biology, but one must eventually move on. My next postdoc was the Darwin Fellowship (I applied because of the title) at the University of Massachusetts (UMASS) Amherst, working with Ben Letcher on salmon again (hard to shake the habitat). While at UMASS, my guilt started building about telling David I would write an NSF grant and then not having done so, so I went ahead and wrote one, which got funded on the second shot (after bringing in my salmony lab-mate from Tom’s lab, Mike Kinnison). So my work on guppies eventually developed owing to guilt about not carrying through on something I said I would do.

The laboratory for our guppy work - here the Paria River, Trinidad
While at UMASS, my office happened to be near that of Jeff Podos, who was working on Darwin’s finches. Near the end of my Darwin Fellowship, Jeff received an NSF Career grant and had money to burn – I mean invest. Jeff knew of my interests and asked if I wanted to come along to the Galapagos on the project (he recalls me asking – or perhaps begging – to come with him), and of course I immediately said yes. So my work on finches was simply a case of being in the right place at the right time. The experience was every bit as exciting as promised in my day dreams that cold winter back in 1994. Several years later, Jonathan Weiner called to talk about my salmon work and I was able to tell him how influential his book had been and how it actually brought me (without any plan) to work on finches.

The laboratory for our finch work, presided over by a marine iguana.
In short, a large amount of coincidence and serendipity determined my choice of study systems. Once in each of the three systems, I became enamored with them and never left. I have now 25 papers on stickleback, 22 papers on guppies, and 11 papers on finches, and I have no intention of ever pulling back from any of these systems. I have also published 33 papers on salmon, and I continually look for new opportunities for additional work on them.

Perhaps my favorite finch photo.
Peter Grant once told me that, in conversation with Daniel Pauly at UBC, Dan told him that he (Peter) was a “point person” whereas he (Daniel) was a “line person”: a point person being someone who takes a single subject/system (finches) and looks at every aspect of their ecology and evolution, and a line person being some who takes a single subject (fisheries) and looks at it across many systems. I guess that makes me a pitch-fork person – trying to go into depth in three systems. Of course, this means that I can’t get too deep in any one system, much to my frustration. However, comparing and contrasting results from the three systems has proved fascinating. For instance, I study ecological speciation in all three systems with essentially the same methods (catching, banding/marking, measuring, recapturing, genotyping) focused on revealing the same processes (disruptive/divergent selection, adaptive divergence, assortative mating, gene flow). The similarities and differences in results obtained from the three systems has proved very instructive and motivational. In fact, my favorite research talk involves walking through a comparative story of ecological speciation in the three systems.
The title slide of my pitchfork talk.
Beyond how many systems one works in, I need to return to the question of working with model (developed) versus new (undeveloped) systems. As noted earlier, a benefit of working in a model system is that one doesn’t have to do as much background work (although every system is nowhere near as well-understood as the impression given by the literature), whereas a cost is that you are never known as the expert in that system (because the experts are the senior folks working on the same thing). The cost-benefit payoff is not easy to calculate and so the temptation for many students and postdocs is to spend a lot of time debating the pros and cons of the different approaches. I think all this angst is a mistake (or at least suboptimal) and that one should instead follow one’s nose (and Mom’s book recommendations). I think everyone should work on the systems and with the people that they find the most interesting and inspiring – not the systems that have the best-described genomes (as an example). These inspiring systems might be model systems or they might be new systems or both (I also have students who work on non-model systems), but they are – most importantly – the systems that feel right at the time, not the systems that have been rationalized based on a logical calculation of optimal career advancement. It worked out fine for me (and many others) – although I am sure my colleagues would argue I could still use considerably more career advancement.



An interesting perspective by Joe Travis on question-based versus system-based science: Is it what we know or who we know? Choice of organism and robustness of inference in ecology and evolutionary biology

Friday, November 14, 2014

Why stop there? Probing species range limits with transplant experiments

[ This post is by Anna Hargreaves; I am just putting it up. –B. ]

Understanding why species occur where they do is a fundamental goal of ecology.  Predicting where they might occur in the future is also an increasingly important goal in conservation, as invasive species spread and native species respond to climate change.  One approach to explore both issues is to study the edges of species distributions and the processes that currently limit them.

Do species stop occurring where things suck too much?
A satisfyingly simple explanation for range limits is that each species has an optimal set of conditions under which it thrives.  As it moves away from that optimum, individual fitness declines and populations dwindle.  At some point populations are unable to sustain themselves (we call this break-even point the niche limit) and the species disappears from the landscape.

Canada’s humans show a classic niche-limited northern distribution, huddling along the southern border for warmth. Statistics Canada. 

The best way to test why a species doesn’t occur somewhere is to move it there and see what happens (provided one is not dealing with humans). Transplant experiments comparing fitness within and beyond the range can test for predicted fitness declines.  Assuming experiments are adequately replicated in time and space, transplant success at the range edge but failure beyond suggests the range limit coincides with the species’ niche limit.

Adequate replication is no small feat, however, and makes good transplants labours of love (often minus the love by the end).  Since strong experiments will only ever be conducted on a small subset of species, those of us studying range limits must eventually ask ourselves, “is there any hope of predicting across species, or is it all just ‘stamp-collecting’?”.

To address this slightly uncomfortable question, we tested for patterns among transplants of species or subspecies across their latitudinal, longitudinal, or elevational range (111 tests from 42 studies).

A smattering of the 93 taxa transplanted beyond their range.  Most studies used plants, which obligingly stay where you transplant them.  

We tested how often range limits involve niche constraints by testing how often performance declined from the range edge (ideally) or interior (if there were no edge transplants) to beyond.  To compare among studies that measured everything from lifetime fitness to clam respiration, we calculated the relative change in a given performance parameter:

`(text{performance within the range } – text{ performance beyond}) / text{mean performance}`

What did we find?

Fitness declined beyond species’ ranges in
75% of 111 tests

and the average decline was significant.  The percentage was even higher when studies measured lifetime fitness (83% of 23 tests).  This strongly supports the importance of declining performance (niche constraints) in limiting species distributions.

How often do range limits coincide with niche limits?  We restricted this analysis to studies that included transplants at the range edge.  Without them one cannot tell if range limits coincide with niche limits, or exceed them via sink populations (middle vs. right panel Fig. 1).  Although most range limits involved niche constraints, only 46% coincided with niche limits.

Fig. 1. Hypothetical results comparing transplants at the range interior, limit, and beyond. Fitness declines beyond the range in all cases, but only the middle scenario suggests range and niche limits coincide.  Numbers give % of 26 meta-analysis tests that fit each RL vs. NL  pattern. Click to view at full size.

Discrepancies are generally explained by dispersal
If species fail to occupy suitable habitat beyond the range, they are dispersal limited.  If edge populations occupy unsuitable habitat (a phenomenon for which range limits have been nicknamed “the land of the living dead”; Channell 2000, Nature) they must be maintained by dispersal from the range interior to persist (Fig. 2).

Figure 2.  The full array of range limit vs. niche limit possibilities. Click to view at full size.

So, while most range limits involve niche constraints, dispersal decouples many from the species’ niche limit.  Interestingly,
while latitudinal ranges were often dispersal limited, elevational ranges were more likely to exceed niche limits via sink populations.
This makes sense given the much longer dispersal distances needed to traverse a geographic gradient than an equivalent elevational one.

Which niche constraints matter?
Amid growing concern over modern climate change, a lot of effort is spent predicting how species distributions will respond, with most models assuming range limits are imposed by climate.

Beech trees in Patagonia, Argentina don’t like cold winters either.

While climate is undoubtedly important, there are many examples of ranges limited by other factors, including interactions among species.  As species interactions are messy to include in range shift projections, it would be useful to know how important they are, and when.

Interactions with other species are important
We compared transplants in natural environments to those that softened potential biotic interactions (e.g. reduced competition or herbivory).  Beyond-range fitness declines were more severe when transplants were subject to all possible biotic interactions (Fig. 3).

Figure 3. Allowing all possible biotic interactions results in bigger fitness declines beyond species range limits (P = 0.0097). Click to view at full size.

We also tested an old hypothesis that biotic interactions are especially important at a species’ low-elevation and low-latitude range limits (click to view at full size):

We compared the drivers of high vs. low elevation limits and high vs. low latitude limits.  As predicted, most high-elevation range limits were imposed by purely abiotic factors (e.g. climate), whereas
> 50% of low-elevation limits
were imposed at least partially
by species interactions
(Fig. 4).  The same pattern exists for equatorial vs. polar limits, but there were too few studies to test it statistically.

Figure 4. Interactions among species are more important in imposing low-elevation (and low-latitude) range limits. Analyses included only transplants into natural environments that provided enough data to assess the causes of the range limit studied. Click to view at full size.

Not just stamp collecting
Our meta-analysis of transplant experiments revealed broad geographic patterns in the relative importance of niche constraints and dispersal, and in biotic vs. abiotic constraints (it also revealed that really good experiments are rare and sorely needed, if you’re tempted).

Implications for predicting climate-driven range shifts
First, we might expect faster relative shifts of high-elevation vs. polar range limits.  Dispersal is better at keeping range and niche limits in equilibrium across elevation gradients, and sink populations common at elevational range limits may provide a head start.  At the other end, ranges limited primarily by other species will respond less predictably to climate change, so we should not be surprised to see a mess of contrasting responses at lower limits.

Anna Hargreaves, Queen’s University

Monday, November 10, 2014

Plasticity in mate preferences and the not-so-needed Extended Evolutionary Synthesis (EES)

[ This post is by Erik Svensson at Lund University; I am just putting it up.  –B. ]

Andrew Hendry at McGill was kind enough to invite me to write a guest post at his blog, where I would explain why odonates (“dragonflies and damselflies”) are great study organisms in ecology and evolution, and I happily grabbed this opportunity. I will also re-publish this post at our own blog, Experimental Evolution, Ecology & Behaviour. Here I will try to put our research and our study organisms in a somewhat broader context, briefly discuss the role of plasticity in evolution and whether we would need a so-called “Extended Evolutionary Synthesis” or EES, as has recently been argued by some.

I am writing this from Durham (North Carolina), where I am currently at a so-called “catalysis-meeting” at NESCent (the “National Evolutionary Synthesis Centre”). The title of our meeting is “New resources for ancient organisms – enabling dragonfly genomics”. Briefly, we have gathered a fairly large group of researchers working on various aspects of odonate biology (including ecology, evolution, behaviour, systematics, population genetics, etc.) to create a genomics consortium, with the long-term goal of making genomic resources available for these fascinating insects so that we can recruit new talented postdocs and PhD students to our research community. This would be needed – I think – as evolutionary biology is suffering from somewhat of a low diversity in study organisms. A few classical model systems tend to attract a disproportionate number of researchers, such as Drosophila, sticklebacks, Anolis lizards, guppies, etc. But odonates are cool too! Please consider joining us, if you read this and are a young scientist who is looking for some relatively unexploited research organisms.

As an example of research in this group and in my laboratory, I would like to highlight our recently published paper in Proc. R. Soc. Lond. B.  entitled “Sex differences in canalization and developmental plasticity shape population divergence in mate preferences”. This is a study that contains experimental field data that were first collected back in 2003 – over a decade ago! – which has later been complemented with population genetic analyses and laboratory experiments.

Our study organism is the banded demoiselle (Calopteryx splendens; male in A above, female in B), which co-exists with its congener the beautiful demoiselle (Calopteryx virgo; male in C, female in D, above) in a patchy network of sympatric and allopatric populations in southern Sweden. What we show in this paper is that there is pronounced population divergence in both male and female mate preferences towards heterospecific mates, in spite of these weakly genetically differentiated populations being closely connected through extensive gene flow. Whereas females learn to recognize mates, males do apparently discriminate against females already when being sexually naive, revealing differential and sex-specific plasticity in mate preferences. Males are therefore more canalized and females more plastic in their mate preferences.

Interestingly, these sex-differences in developmental plasticity and canalization are also scaled up and shown at the between-population level: females show strong population divergence in mate preferences compared to males, presumably related to their higher plasticity. This suggests that plasticity can and does play some role in population divergence, even in the face of gene flow, which is of some principal interest to evolutionary biologists, and fits with ideas proposed by Mary Jane West-Eberhard in her book “Developmental plasticity and evolution”, but also with a recent population genetic model by Maria Servedio and Reuven Dukas on the population genetical consequences of learned mate preferences.

Given our results in this study, one could perhaps expect me to show some enthusiasm for the recent opinion-paper by Laland et al. in Nature entitled “Does evolutionary theory need a rethink?” But, as a matter of fact, I do not like the opinion piece by Laland et al., and I think it is one of those opinion articles that would fit better as a blog post. As it stands now, the opinon article by Laland et al. gives a misleading impression of a very divided scientific community and results in a confusing discussion for discussion’s sake.

Laland et al. argue that developmental plasticity, niche conservatism and some other factors are important in evolution, and so far I agree with them. They then go on to make various strong (but in my opinion very biased and sometimes unsubstantiated) claims that evolutionary theory needs to be changed substantially and radically. They argue for an “Extended Evolutionary Synthesis” that should replace the current Modern Synthesis. It is a bit unclear to me, first why we need an EES, second to what extent the current paradigm stops anyone from doing the research he or she wants, and third, what this EES would actually contain that makes it so urgently needed. The authors are quite vague on this point. In my opinion, far too many opinion articles have been published about the need for an EES, and far too little rigorous empirical or theoretical work has been performed, in the form of critical experiments, formal theory or mathematical modelling.

The EES is actually not an invention of Laland et al.; the term was first coined by former evolutionary biologist Massimo Pigliucci, who is today a professor in philosophy, after he has left evolutionary biology. During his relatively brief career as an evolutionary biologist, Pigliucci produced a steady stream of opinion articles and edited volumes in which he constantly questioned and criticized what he saw as “mainstream evolutionary biology” or “The Modern Synthesis”. His efforts culminated in a meeting he organized entitled “Altenberg 16”.

This meeting at Altenberg gathered a selected group of (self-proclaimed) scientific “revolutionaries” and resulted in a book entitled “Evolution – The Extended Synthesis”. What struck me, as an experimental evolutionary ecologist, was the rhetorical tone of the whole effort, the grandiose worldview of  put forward by the group and the seemingly na├»ve belief that scientific synthesis can be organized and commanded from above, and thus be declared, rather than growing naturally from below. The meeting at Altenberg was also quite biased in terms of who were invited – further strengthening the impression of an old boys network with a very biased view of evolutionary biology, mainly grounded in philosophical, rather than empirical arguments.

However, even if we accept that science in general, and in evolutionary biology in particular, evolves and changes over time, and even if we believe philosopher Thomas Kuhn’s theory about “paradigm shifts” and “scientific revolutions”, it does not follow that a revolution will happen just because there are willing revolutionaries. This is not how political revolutions happen either, such as the French, the American, or the Russian Revolutions. Having dedicated revolutionaries is not enough; such revolutionaries are only a subjective factor. What is also needed is the objective factor: the material (or scientific) conditions necessary for a revolution (political or scientific).

Neither Laland et al. nor their predecessor Massimo Pigliucci have have convinced me that they are the leaders we should follow, or that the time for the scientific revolution or a substantial paradigm shift is waiting around the corner. Although I do not consider myself an orthodox population geneticist at all, in this case I tend to agree with population geneticist Jerry Coyne, who has previously criticized Pigliucci for being committed to BIS – Big Idea Syndrome. One symptom that somebody is suffering from BIS is initiating debates for debate’s own sake. I  feel that the same criticism can be directed to Laland et al. Their rather rethorical opinion piece contains very few concrete suggestions of how to do research differently than we do today. This gives me the impression that this is mainly a debate about how to interpret the history of science, rather than being useful or providing practical advice to evolutionary biologists in their daily work.

Both Laland et al. and Pigliucci have painted a picture of evolutionary biology and the Modern Synthesis as a monolithic and dogmatic scientific paradigm that prevents researchers from asking heretical questions, such as addressing problems about plasticity. The Modern Synthesis clearly did not stop me and my co-workers from initiating our study on mate preference plasticity in damselflies. Neither is it clear to me that an EES (if it had it existed) would have helped us in any way to design our study differently than we actually did in the end. Given these considerations, I am quite convinced that the debate about the EES is truly academic (in the negative sense), as it will not lead us anywhere or provide us with any new analytical tools, tools being either empirical or theoretical. I therefore do not think that the proposed EES will have any long-lasting effect on the field of evolutionary biology – at least not as much as its proponents wish.

I am also quite frustrated by the poor scholarship of Pigliucci and Laland et al. regarding the history of the Modern Synthesis. Their rather negatively biased view of the Modern Synthesis strikes me as being a good example of a straw man argument wherein they set up the scene by making a caricature of something they do not like in the first place, and then go on to criticize that caricature. But their caricature is far from the more complex reality, richness and history of the Modern Synthesis.

A few years ago Ryan Calsbeek and I edited a book entitled “The Adaptive Landscape in Evolutionary Biology”, in which we and many others discussed the contrasting views between the population geneticists Sewall Wright and Ronald Fisher, and their legacy which still influences evolutionary biology and population genetics today. It is simply wrong to claim that was a monolithic paradigm that did not allow for radically different views on genetics, plasticity, and micro- and macroevolution. Had Pigliucci and Laland et al. read the various contributions in our book, many of which had radically different views, some of their misleading arguments could have been avoided. Critical views similar to those I have expressed in this post can be found on the blog “Sandwalk”, such as here and here.

However, I would say that there might already be an ongoing synthesis  in evolutionary biology – but it is not led by Laland et al. To see what I mean here, Steve Arnold published an interesting paper earlier this year in the American Naturalist entitled “Phenotypic Evolution: The Ongoing Synthesis”. In this article, Steve argued that evolutionary biology is now in the midst of a true synthesis, wherein micro- and macroevolution are finally coming together through the integration of quantitative genetics with comparative biology, largely driven by the explosion of phylogenetic comparative models of  phenotypic trait evolution.

Unlike the case for the EES, there are many more "silent" revolutionaries in the field of comparative biology who are now busy in developing analytical methods for phylogenetic comparative methods in the form of R packages and other useful tools. These new methods enable us to directly study and infer evolutionary processes and test various models and evolutionary scenarios. This is the sign of a healthy and dynamic research field: people do things, rather than just talking about the need for revolutions. Researchers in this and other fields are busy making quantitative tests, rather than spending time on verbal reasoning on the need for new syntheses. To paraphrase  a legendary revolutionary (anarchist Emma Goldman): “If you can’t do any rigorous experimental procedures or statistical tests, it is not my kind of scientific revolution”.

In summary: science evolves over time, and so does evolutionary biology. Our field is very different from what it was in the early days of the Modern Synthesis – in spite of some of the claims by Pigliucci and Laland et al. Without a doubt, plasticity, niche construction, and many other phenomena mentioned by Laland et al. are worthy of study and certainly very interesting. The mistake Laland and other proponents of EES make is that they think that they are the only ones who have realized this, and that other folks outside the EES camp are not thinking deeply about these problems. I end this blog post by citing another true revolutionary (quote taken from Jerry Coyne’s blog “Why Evolution is True”):
I close with a statement by my old mentor, Dick Lewontin, who of course as an old Marxist would be in favor of revolutions: “The so-called evolutionary synthesis – these are all very vague terms. . . That’s what I tried to say about Steve Gould, is that scientists are always looking to find some theory or idea that they can push as something that nobody else ever thought of because that’s the way they get their prestige. . . they have an idea which will overturn our whole view of evolution because otherwise they’re just workers in the factory, so to speak. And the factory was designed by Charles Darwin.”

Final note: I am fully aware that both Laland et al and Massimo Pigliucci are likely to strongly disagree with my criticisms above. The views are entirely my own and do not necessarily represent other authors of this blog. 

Tuesday, November 4, 2014

How to be a reviewer/editor

Many articles have been written about how to be a good/responsible/fair/rigorous/timely reviewer or editor. Having now reviewed more than 400 papers and having been an editor for 100 more, I find myself developing rather strong opinions on the subject. If those opinions meshed nicely with the ones previously published, a blog wouldn’t be needed – but they don’t. Instead, I find myself holding rather different views on how to be a reviewer and editor. As time has gone on, these opinions have strengthened, not weakened, and so perhaps it is time to get them out there.

How to be a reviewer – 1 simple rule.

Don’t reject papers!!!!!!! How’s that for a minority opinion? Even before we start our reviewing careers, we are told to be very stringent and critical and to only accept the very best stuff. But – as I will explain – this does not work.

As a reviewer, your goal is to improve the scientific literature, which you can achieve by helping good papers get published, by stopping bad papers from getting published, and by improving papers before publication. The straight-up reality is that the second option is out: you simply can’t keep stuff out of the literature. Hundreds of journals exist and so rejecting a paper at one journal just means it will end up getting published in some other journal (Fig. 1), especially in this new age of pay-as-you-go open access publishing. Worse yet, if you reject a paper, the authors have no obligation to follow your suggestions for improvement. Thus, rejecting a paper actually makes the scientific literature WORSE. Instead, you want to keep whatever paper you are reviewing in play at the same journal. That way, the author will be encouraged/required to follow your suggestions for improvement. You and the authors can work together to craft the best possible paper – what a wonderful world (Fig. 2).

Fig. 1. If at first you don't succeed, try try again. Network of submission flows among journals. From Calcagno et al. 2012 (Science).
Several exceptions to this rule might seem necessary. First, some papers are just irredeemably bad in the sense that no amount of re-analysis/re-writing will make them tolerable. In these cases, you have no choice but to reject them – but remember that they will likely just pop up in some other journal in close to the form you previously rejected them. Thus, you really need to be somewhat relaxed about what you consider irredeemable. By this I mean that the paper really has to be fraudulent or completely (not just partially) incomprehensible. In practice, I think such heinousness applies to a vanishingly small subset of papers.

The second exception occurs when the paper just isn’t at all suited for a journal. By this I don’t mean that it “isn’t good enough” (but see below); I instead mean that it is a paper about behavior in a journal about morphology, or some such. Again, however, this is extremely rare as it is the job of the editors to catch this mismatch.

A third exception might occur when you think the paper is far below the quality expected for the journal. The most obvious example is Nature/Science, where we often look at papers and think “Sheesh, I had 10 papers rejected from these journals that were way better than this one.” It is very hard to resist this sentiment and so, yes, sometimes you won’t be able to avoid the temptation to suggest rejection simply because you think the paper “isn’t good enough for the journal.” In addition, papers in such journals tend to get a lot of attention – and so by rejecting them, you will certainly reduce the attention paid to them, thus in essence "keeping them out of the literature" in another sense.

I should make sure to clarify that “don’t reject” means don’t reject without the possibility of resubmission. In many cases, the paper really does require a ton of work and so it really should be rejected “in its current form” or “without prejudice” and resubmitted. Note, however, that specifying this as a reviewer is likely – given the editors different objectives (see below) – to get the paper rejected. Thus, a better option is usually “major revision” – that is, if the authors really can do what you say for improvement, then the paper should be publishable.

Fig. 2. The different types of reviewers, according to what I am suggesting is a hybrid "heavy weapons guy" - "medic" - "engineer" - but without the bad parts.

How to be an editor – no simple rule.

In contrast to your role as a reviewer, your role as an editor becomes a balance between your desire to improve the literature (Don't reject papers!) and the journal’s desire to have you improve their journal in particular. Thus, you now end up rejecting papers because your journal wants to publish the very best work and thus increase its impact factor and prestige and subscriptions (and money) and so on. Nowadays, a lot of pressure is placed on editors to reject papers (ideally without review) so that you don’t end up accepting more papers than the journal has funds to publish. (And so you don’t end up wasting everyone’s time with a review process that is likely to fail anyway.) This necessity can be quite frustrating when one tends to fall more on the “improve the literature” side of the balance. Indeed, I see many papers that could be good being rejected simply because they aren’t as good as other papers that are being submitted.

So how does one decide which papers should be rejected and which shouldn’t? In my opinion, acceptance or rejection should not in any way depend on the actual results of the study. If the study is well-motivated, interesting, well-designed, well-executed, and well-analyzed, then it should be published regardless of whether or not it confirms a specific prediction or hypothesis or theory. Perform this thought experiment: take a study that has a negative (e.g., non-significant) result and imagine a positive (e.g., significant) result instead. Would you want to publish the paper? If so, accept it even though it has a negative result. Often, people complain about the design of a study with negative results (“they should have done this, they shouldn’t have done that”) but the reality is that they would not have complained if the result had been positive.

Other reasons to NOT reject a paper (as a reviewer or editor) are: poor stats, poor writing, poor citation of the literature, poor graphics, and so on. All of these things can be fixed with revisions. Just tell the authors what they need to fix. If they can do it, great, if they can’t, fine, you can always reject it next time.

Reasons TO reject a paper from your awesome journal: samples sizes that are too small (but you can encourage the authors to collect more data), a study design that is incapable of testing the hypothesis (but perhaps the hypothesis/study can be rephrased/reframed in a way that it can be tested), and lack of replication at the level where inference is being attempted. For me, this last problem is often the most damning. For instance, if one wants to make inferences about populations in two environments (north vs. south, cold vs. warm), then at least two independent populations of each type must be studied. Of course, studying only two populations (one of each type) is still a good suggestive study; it just might not be good enough for my awesome journal.

How to be an open access editor – 1 simple rule

Accept everything!!!!! In sharp contrast to the above, your role here is to make money for the journal. Yes, I know this is a cynical perspective in a growing culture where open access is considered a paragon of virtue: “make your science freely accessible to everyone, and the world will be a better place.” Having now worked with several open access journals, however, it is very clear that the entire goal is to make money. Consider this: pay-as-you-go open access journals don’t make a cent unless they publish your paper. Stated another way, they lose money every time they reject a paper. Indeed, that is why so many publishers have started open access journals to which they “refer” papers they have rejected from their flagship subscription-based journals. Before these new ventures, every rejected author simply went and paid someone else (if all else fails, PLoS ONE!) to publish their paper instead (Fig. 3) – so the publishers said, “hey, cool, we can also get money from the papers we reject – how awesome is that.” By the way, if you like open access, check out my new super-easy, super-fast, and super-cheap open-access journal “MyScience”:

Fig. 3. PLoS ONE publications (100,000 as of June 2014) = lost revenue to for-profit publishers. The solution: start your own open access journal. (Graphic from PLoS Blogs.)
In my opinion, a better solution is simply to pay for open access at the subscription-based journals, which is nearly always an option, or simply publish in the subscription-based journal and then put the PDF on your website. Yes, I know the publishers imply you shouldn’t do this but I have been doing it for 20 years and no one has complained yet.

Oh s**t.

Having just written the above, it now strikes me that authors reading this post will have a more prosaic inspiration – “hey, I should recommend Hendry as a reviewer/editor – he won’t reject my paper.” Go for it. These days I receive so many requests to review that I turn down most of them anyway: instead only accepting reviews for papers that are squarely in my areas of expertise, which some folks suggest are quite circumscribed. And now I am struck by another realization: editors reading this post might have a different prosaic inspiration – “hey, I better not recommend Hendry as an editor/reviewer as he won’t reject enough papers.” Go for it. I get too many requests to review anyway. 

Monday, October 27, 2014

An invasive species drives rapid evolution in a native

Anolis carolinensis male, dewlapping. Photo by Ambika Kamath.

In 1956, W.L. Brown and E.O. Wilson proposed the following eco-evolutionary process: two closely-related species come into contact, interact strongly (usually over food and other resources), and thereby experience natural selection to diverge from one another--ecology influences evolution. Then, if such divergence resulted in sufficient resource partitioning, the species’ population dynamics would stabilize and the two (or more) species would coexist--evolution influences ecology.

They called this particular eco-evolutionary process character displacement. During my dissertation, I spent several years investigating character displacement in Anolis carolinensis, culminating in publication last week. Thus, E-E-E-E regular Andrew Hendry asked me to describe the study for Eco-Evo-Evo-Eco.

Anolis carolinensis, marked 47 with Sharpie during a capture-mark-recapture study. Photo by Todd Campbell.
Anolis carolinensis is the only anole native to the southeast United States. Adults are about 3-5cm in length, weigh 2-5 grams, eat arthropods, and are active during the day. Males are territorial and use a dewlap along with push ups and headbobs to communicate during territorial and mating interactions.  

The sister species of A. carolinensis, called A. porcatus and A. allisoni, live in multi-species anole assemblages in Cuba today. Likely because of interspecific interactions, they are restricted in their habitat-use to high parts of tree trunks and the tree canopy. Thus, when the ancestor of A. carolinensis, also from Cuba, arrived to the US 2-3 million years ago, it probably experienced ecological release: with no other anole competitors, it shifted its habitat use towards low parts of the tree trunk and the ground. At least, that's where we find it today.

In the late 1940s/early 1950s, Anolis sagrei arrived to south Florida, likely as a stowaway in agricultural and other pre-embargo trade shipments between the US and Cuba. Like A. carolinensis, this lizard is 3-5cm long, 3-6 grams heavy, active during the day, eats arthropods, and maintains territories. Moreover, like A. carolinensis in Florida, this species likes to use the ground and low parts of the tree trunks.

Thus, where they overlap, the two species are likely to interact strongly with one another, setting the stage for character displacement.

Shortly after its colonization, A. sagrei established and spread northwards quickly. Today, it is well into southern Georgia and has even founded populations in Louisiana, Texas, and Hawaii (such jump-dispersal from Florida/Georgia is likely facilitated by horticultural traffic - the best place to find A. sagrei anoles in Houston a few years ago was Home Depot's garden department).

Anolis sagrei male, dewlapping. Photo by Adam Algar.
With its colonization and spread throughout Florida, A. sagrei has become arguably the most abundant vertebrate by biomass in the state and it must therefore affect its close relative, A. carolinensis. My colleagues and I tested two predictions for how: one ecological prediction, and one evolutionary prediction.

Prediction 1: Ecology
By the time of our study, Collette (1961) and others had noted that A. carolinenis tended to perch higher in the canopy whenever it was in sympatry with A. sagrei in Florida. Collette predicted that A. sagrei was responsible for this habitat-use shift in A. carolinensis, but the definitive evidence remained elusive, as there were many alternative explanations. A field experiment was need, and in 1995, my colleague and co-author, Todd Campbell found the place to do it: Mosquito Lagoon.

Dredge-spoil islands in Mosquito Lagoon, viewed from Oak Hill, looking south towards Cape Canaveral and the Kennedy Space Center. Note the houseboat in the foreground for scale; larger islands are about a hectare in size. There are about 80 islands in this part of the lagoon. Photo by Todd Campbell.
In the 1950s, the US Army Corps of Engineers dredged a channel through the lagoons that line the Atlantic coast of the United States. This channel, the Intracoastal Waterway, was meant to provide a sheltered lane for shipping traffic. The dredging machines (nicknamed 'clinkers' for the sound they made) would suck material from the bottom of the lagoon and then deposit that material alongside the channel. As a byproduct (spandrel?) of dumping the dredge spoil material, an island was formed, and this process was repeated regularly every 50 meters or so for hundreds of kilometers of coastline. These islands provide a replicated system excellent for testing the effects of A. sagrei introduction on A. carolinensis.

It took until the late 1980s for A. sagrei to arrive to the mainland bordering Mosquito Lagoon, which is about halfway up the Atlantic coast of Florida. At that time, the dredge-spoil islands there had established plant communities (mostly cabbage palm, eastern red cedar, buttonwood, and mangroves) and supported multiple arthropod species. The islands also had A. carolinensis on them.

Todd wished to know what the demographic impact of A. sagrei introduction would be on those A. carolinensis island populations. In 1995, he picked three islands (one small, one medium, and one large) as experimental islands and three similar islands as controls (using a random blocked design). That May, he conducted capture-mark-recapture studies for A. carolinensis on all six islands. Then he introduced A. sagrei (collected from the nearby mainland) to the three experimental islands.* He followed the populations with capture-mark-recapture surveys for the rest of that summer as well as summers of 1996, 1997, and 1998 - a colossal effort.

At the same time, and crucial for our story here, Todd also monitored perch heights in A. carolinensis. Thus, he could ask, is there a perch height shift in A. carolinensis following A. sagrei invasion? Here's what he found.

Within a few months, A. carolinensis perched higher in the presence of A. sagrei. This provided compelling experimental evidence confirming Collette's prediction: that interactions with A. sagrei would drive a habitat use shift by A. carolinensis.

Prediction 2: Evolution
Collette also made a second prediction; he had observed that A. carolinensis, in regions with A. sagrei, had larger toepads with more specialized scales, called lamellae. Thus, Collette predicted that a habitat shift in A. carolinensis, driven by A. sagrei, would result in the evolution of larger toepads. (Anoles that have larger toepads with more lamellae are better at clinging to surfaces. Across the ~400 species of anoles, those species that live higher in the canopy tend to have larger toepads. Together, this suggests that it is adaptive to have larger toepads when living higher in the canopy.)

The right hindfoot of an A. carolinensis male. Note the expanded scales on the distal portion of the toes. These are the lamellae. Photo by Yoel Stuart.
In 2010, I went to Mosquito Lagoon to test this prediction. My plan, as proposed to my thesis committee, was to go back and compare A. carolinensis toepads on the experimental versus control islands from Todd's demographic study. This would be great experimental evidence for character displacement.

Unfortunately, this wasn't feasible, primarily because A. sagrei had naturally colonized Todd's control islands. In fact, my surveys in 2010 and surveys by Nathan Turnbough (at UT Knoxville) a few years prior showed that A. sagrei had reached all but five islands in the lagoon - more than 70 natural colonizations.

Being unable to use the experimental islands was a disappointment to be sure, but a nice long conversation between myself, Todd, and co-author Jonathan Losos convinced me that we weren't sunk yet. (We had this conversation in the field, which is I think the best place to plan such work, as you really do get a sense for the challenges you'll be facing).

We decided that though I couldn't look for evolution in a true experiment, I did have a replicated setting where I could compare A. carolinensis toepads on un-invaded islands to toepads on islands with the invader. Moreover, I could bound the invasions in time. In preparation for the 1995 study, Todd surveyed most of the islands in the lagoon for both species, finding that most islands had just A. carolinensis. My 2010 survey showed that these islands had been invaded sometime in the intervening 15 years, setting the amount of generations A. carolinensis could have been evolving with A. sagrei on those islands to approximately 20. It wasn't a true experiment, but this natural experiment was pretty darn close.

We collected our data as follows. My field help and I headed out in a small boat every morning, landing on an island as the sun was coming up. We walked through the islands slowly until we saw a lizard that was undisturbed by our presence. We noted the perch for perch height measurements and then tried to capture the lizard so that we could measure its toepads. To catch the lizards, we used extendable fishing poles with fishing-line lassos tied to the end--get that lasso around the head, give a little tug, and you had your anole. (The lizards are very light, so this doesn't injure them). In the afternoons, we returned to our lodging and collected toepad images using a flatbed document scanner available for purchase in any office supplies store, ramped up to 4800 dpi (see image above). During that process, the lizards were anesthetized; after scanning, we let them wake up and recover overnight, and then put them back where we caught them the next day. We often wondered if their friends believed their abduction stories.

Stuart (foreground) and Campbell (background) pursuing lizards on Hook Island. Note the fishing pole pointing upward from Stuart's hand. The lizard must have been high up, as the pole is nearly fully extended.
Once I had the perch height and toepad data, first, I wished to double check that A. carolinensis on invaded islands did indeed perch higher than on un-invaded islands. They did, confirming the experimental result.

Then I tested whether A. carolinensis from invaded islands had larger toepads with more lamellae. Consistent with prediction, they did, suggesting that A. carolinensis was adapting to its arboreal lifestyle!

As noted above, I knew that A. sagrei couldn't have gotten to the islands earlier than 15 years, or about 20 generations, before 2010. I calculated the rate of divergence in haldanes over those 20 generations, and found that populations on invaded islands were diverging from populations on un-invaded islands at about 0.08 standard deviations per generation for each trait. To put that in human perspective: the average height of the American male is about 5'9". If American male height were increasing at 0.08 standard deviations for 20 generations, the average American male would be 6'4", or the size of an NBA shooting guard (assuming basketball was still around). This divergence was substantial and fast.

However, because the evidence for toepad divergence was observational, there were several alternative hypotheses, other than evolution, that might explain it instead. In the interests of space, I'll just say that: (1) we used a common garden experiment to show that there is an evolved genetic component to the observed divergence; (2) we conducted random habitat surveys to show no appreciable differences in environment between invaded and un-invaded islands; and (3) we used RAD-seq data to show that the islands were evolving independently of one another, so that the observed divergence wasn't the result of ecological sorting. This was a huge amount of work to collapse to a single paragraph; one paragraph doesn't suffice to give enough credit to co-authors Graham Reynolds, Liam Revell, Paul Hohenlohe, and Jonathan Losos for all the work they did here.

Alright, we're nearing the end. Let me sum up. With experimental and comparative evidence, we showed that the arrival of A. sagrei results in a perch height shift in A. carolinensis, suggesting a strong interaction between the two species. We then ruled against many alternative hypotheses, allowing us to say confidently that this habitat use shift resulted in rapid, morphological evolution by A. carolinensis, likely as adaptation to maneuver better on small, thin, and slippery arboreal perches during feeding, mating, and anti-predator behaviors. The major step now, and a focus of my future work, will be to determine exactly what kind of interaction is happening between the two species. It's likely that they compete for food and space resources. They may also interact agonistically. Moreover, adult male A. sagrei will eat hatchlings of A. carolinensis, so perhaps there is also some intraguild predation at work here, not to mention the possibility of indirect interactions through shared predators and parasites.

Nevertheless, regardless of the nature of the interaction between the two species, the invasion of A. sagrei is driving the rapid evolution of character displacement by A. carolinensis.
The exhaust trail of the space shuttle Endeavour drifting in the skies over Mosquito Lagoon after lift-off on July 15, 2009 (STS-127). The shuttle returned on July 31, accompanied by its typical double sonic boom. Photo by Yoel Stuart.
Y.E. Stuart, T.S. Campbell, P.A. Hohenlohe, R.G. Reynolds, L.J. Revell, and J.B. Losos. 2014. Rapid evolution of a native species following invasion by a congener. Science 346: 463-466 
DOI: 10.1126/science.1257008

* Over the last week or so, I've had a number of folks inquire about the ethics of introducing invasive species. This is a very valid concern; the spread of invasive species should be limited as much as possible. In this case, however, by 1995, A. sagrei was already highly abundant on the mainland and was starting to get to the spoil islands (which, recall, are man-made and didn't exist 40 years prior). Our opinion was that the lizards were going to get to the rest of the islands eventually anyways, so we might as well learn something from the invasions in a controlled, experimental framework. In retrospect, that almost all the islands in the lagoons have A. sagrei on them today substantiates our reasoning. However, we were also lucky that there weren't any unintended consequences (unlikely as those seemed at the time)--I doubt permission from the local permitting agencies would be granted today.

Saturday, October 18, 2014

How to write/present science: BABY-WEREWOLF-SILVER BULLET

As an editor, reviewer, supervisor, committee member, and colleague, I have read countless papers and proposals and have seen similarly countless presentations. Some work well and some don’t. Beyond the picky details of slides that are too wordy, speaking that is too fast, sentences that are poorly constructed, and so on – the most critical problem is making clear why the work is interesting and important. Why should we read further rather than moving to the next paper on the pile? Why should we give you money as opposed to your competitor? Why should we listen to your talk instead of tweeting about the party last night? This simple and yet pervasive inability to engage the reader and have them buy into your work is likely the single greatest flaw in the writing of every student (and many postdocs and faculty members). In this post, I will explain a simple metaphor that can help you to solve this problem in each and every one of your papers/proposals/presentations.

The metaphor emerged from a comment by McGill’s Dean of Science, Martin Grant, about what makes a good proposal. He suggested that you need a werewolf and a silver bullet. With a werewolf, a funding agency and their reviewers can see the problem that needs solving. With a silver bullet, the agency and reviewers can see that you have a realistic chance of solving the problem. When translating this logic to my own students, I have modified it somewhat to better fit the ideal outline of a paper/proposal/talk. The basic idea is that the structure of your paper/proposal/talk should follow this sequence:
  • CUTE BABY: First explain to the reader/listener the general umbrella under which the work falls – some umbrella that will make the reader/listener sit up out of their sleep-deprived torpor and say to themselves “Oh, OK, this talk is about something that is interesting and important. I better pay attention to what new insight they might bring.” Cute babies can be things that are important, such as ecosystem health and human well-being. A common cute baby here is biodiversity and its contribution to ecosystem services. Cute babies can also be things that are interesting, such as theories, with examples from ecology and evolution being the equilibrium theory of island biogeography or the ecological theory of adaptive radiation. In developing this cute baby, it is critically important to not overtly state that the baby is cute. “One of the most important topics in ecology is the maintainance of biodiversity” – this is a “motherhood” statement ( that just annoys the reader by making them feel manipulated. Instead the reader should make up their own mind when reading about the baby that it is indeed cute. Stated another way, if you have to say that your study area/work is important, then people will think you are trying too hard. Instead, the reader should think “oh, that is important” without you having to say it.
Cute babies.*
  • SCARY WEREWOLF: Next explain how that cute baby is somehow threatened, so the reader/listener shifts forward on their seat and begins to empathetically furrow their brow in shared concern, thinking “Yes, that’s true, that really is an unsolved problem that could hurt the baby.” Scary werewolves for biodiversity and ecosystem services might be climate change and habitat loss and, well, pretty much anything. Scary werewolves for theories are things like potentially inappropriate assumptions, or the lack of empirical tests, or the failure to include an important idea, or low explanatory power. Here (as opposed to the above) it is more often OK to state that the werewolf is scary but it is still more effective to avoid motherhood statements and let the fear emerge within the reader’s mind. (How often do stories say “the werewolf was scary”? Instead they say that “a hulking beast with dark, tangled mats of hair emerged from the darkness dripping blood from its fangs with its eyes glinting in the moonlight.”)
Scary werewolf.*
  • SILVER BULLET: Finally, explain how the work that you did/will conduct has the potential to slay – or at least severely wound – the werewolf. If you do this effectively, the reader/listener will begin to unfurrow their brow and nod: “Ah, yes, that would be a great way to solve the problem.” Silver bullets can be applied solutions to problems, such as a new design for corridors that reduces the negative effects of habitat loss. Or they can be new experiments that address outstanding gaps in knowledge, or particular study systems that are ideally suited to show how some theory needs to be modified. Once again, you ideally don’t say “I have the silver bullet”; instead, the reader has this emergent thought while reading about the study system. And, of course, NEVER say your system is “ideal,” which means “couldn’t possibly be better,” as everyone who works on a different system will immediately think “no it isn’t.” Instead your system is “excellent.”
Silver bullet.*
  • DEAD WEREWOLF: For work that has already been conducted, one would ideally show that the silver bullet (new method/theory/experiment/observation) has killed the werewolf and thereby saved the cute baby. In reality, however, it is just as likely – and effective – to show how you have wounded or exposed the werewolf or how you have shown that the werewolf is scarier than originally thought or how you have found a new werewolf. These alternative end points nicely establish the need for further work. For work that has not yet been conducted, such as in a proposal, the dead/injured/new werewolf is not actually shown, but the reader has to see what it might look like. That is, they can visualize the werewolf lying dead on the ground while the cries of the baby fade into giggles while the baby bounces up and down on the werewolf’s belly. (Or alternatively, the werewolf is just a hairy uncle bouncing the baby on its knee.)
Dead werewolf.*
In implementing this schema, I suggest working from the goal that your Introduction will follow the above structure. The first paragraph (or section) describes the cute baby (the general area of research, subtly making clear why it is important), the next section describes the scary werewolf (the problem/gap/limitation of current knowledge/work), the next section suggests a silver bullet (the study system/experiment/new theory), and the final section postulates what the werewolf might end up looking like (the predictions/questions/hypotheses). In the context of a presentation, the entire talk should follow this structure, with the methods falling into the silver bullet part and the results/discussion taking the form of the dead/injured werewolf. Note also that studies sometimes examine multiple questions, in which case the baby-werewolf-bullet approach can take on a fractal appearance: the whole study, the individual components, and sometimes even within individual sections.

At this point, I am sure you have some thoughts or criticisms of the above plan. Although I presumably can’t predict all of them, here are some likely ones.
  • But my work just isn’t that important – it won’t solve world hunger, it won’t halt the loss of biodiversity, and it won’t overthrow the ecological theory of adaptive radiation. Surely I shouldn’t try to pretend it will. Indeed you shouldn’t, but don’t throw the cute baby out with the scary werewolf. Instead, you simply need to scale your baby/werewolf/silver bullet accordingly. If you have a relatively small problem, give us a clearly defined but only modestly scary werewolf. If your silver bullet is unlikely to slay the werewolf, then give us some silver pepper spray. The key point is that the above logic and outline applies regardless of the size of the problem or the actual outcome of the study – yet, it is true that you can’t oversell your baby-werewolf-bullet.
  • But I don’t want to oversell my work – reviewers will see through my attempt to make it seem more important than it is. This concern is related to the one immediately above and, again, it is correct that you shouldn’t promise something you can’t deliver or outline a werewolf you can’t kill. However, you can outline nested werewolves – like Russian doll werewolves where you can slay some small ones thus getting closer to the big one. Conveniently, the solution is the same as above – scale the baby/werewolf/bullet to the scope of your study and what you can deliver.
  • But I didn’t actually kill the werewolf. No problem. Explain how your silver bullet was tarnished (polishing will fix it up) or was made of aluminum (I need a new experiment) or how it missed the werewolf altogether (the werewolf was in our imagination or was really just a hairy uncle seen in low light). This outcome is just as satisfying in most instances.
Maybe it wasn't that scary after all.*
Or maybe it is the baby that is scary.*
Well, there it is – a suggested plan for writing every paper/proposal/presentation for the rest of your career. I hope it helps. I have certainly found it immensely helpful for improving the logical flow and engaging narrative of my students’ work, as well as my own. Of course it doesn’t always work and of course it doesn’t guarantee acceptance (rejection can occur for many other reasons), but I think it solves the problem of how to structure the presentation of ideas and how to make clear the importance of your study. And, even within this framework, many other improvements can be made to the grammer/writing/presentation. Here are my own suggestions: and I list some additional ones below.
The Science of Scientific Writing

On whimsy, jokes, and beauty: can scientific writing be enjoyed?

Five ways to improve your science writing


* I did not take these photos but found them on google - the original source (and copyright holder) is not clear.