There are a few “story behind the paper” style posts that I have in mind, and this one is the first of them. For this one, I’m going to focus on a paper that I wrote in collaboration with Spencer Hall. The hook for this story, as the title indicates, is that it’s a paper that we originally submitted to Ecology, where it was rejected. We then submitted it to AmNat, where it was accepted. That AmNat paper is what later won the Mercer Award. This sounds like a classic case of “Those stupid reviewers got it wrong,” right? But it’s not. To me, this is a story of a success of peer review. The paper that ended up getting published is much, much better than the one we originally submitted to Ecology. That paper would not have won an award. So, to me, this story actually shows the value of peer review. It’s part of why I would be pretty hesitant to go with the “no revisions” option Jeremy posted about recently.
But first, let’s back up. A quick overview of the paper: It is based on a chapter from my dissertation. In fact, it’s based on the chapter of my dissertation that I struggled with writing the most, and that I really came to hate. (Tip for the grad students: arrange things so that the chapter you hate the most is NOT the first chapter you discuss at your defense. It was really stupid of me not to move this chapter to the end, just so I could get on a roll with the other chapters first.) The key question of the dissertation chapter and the Ecology submission was asking how predation influences parasite prevalence and host population dynamics. The paper combined a bunch of different things: characterization of the selectivity of fish predation on Daphnia infected with two different parasites; studies on individual-level effects of those two parasites; detailed population-dynamical studies on 5 lake populations, carried out during epidemics of those parasites; and an epidemiological model trying to link all the empirical data together. It was a whole lot of field work in one paper, and I had completely exhausted myself collecting it. The dynamical sampling involved sampling lakes just about daily (there actually were more lakes that I sampled, but that ended up needing to be dropped for various reasons — including, in one case, the addition of a massive amount of copper sulfate to the lake during the dynamical study). And, on top of that, I needed to go fishing at dawn to get the fish selectivity, and do Schindler series at day and night to get data on habitat use so I could figure out water temperatures to use for egg development time for the dynamics data. And all the samples need to be counted live (we can’t see the parasites in preserved samples), which didn’t leave a whole lot of time for sleep. All of which is to say: I had a whole lot invested in this paper. I should also say that part of why I could do all that field work is my father was my field assistant for this study — he was a really fantastic field assistant, and I highly recommend exploiting family members whenever possible.
The big problem for this paper in review was the modeling component: in short, the model did a bad job of capturing the dynamics of the system. It predicted that the parasite should become endemic (that is, that it should persist indefinitely in the host population). In nature, we always see epidemics, where infections go from rare to common to rare in a fairly short period of time. So, we were hinging our whole explanation of things on a model that did a really bad job of describing the system. In our defense, we acknowledged this, and only analyzed the invasion of the parasite, where the model did a good job. But, still, the model wasn’t a good one for our system. So, after two rounds of review at Ecology, the paper was rejected, in large part because of the model.
It became clear, then, that we needed to do something to make the model better. By this point, I had already developed the evolutionary epidemiological model that was the focus of my 2007 paper with Lena Sivars-Becker. That model captures the dynamics of our system quite nicely. So, we decided that we needed to combine the model in the original paper that added selective predation to an epidemiological model with that evolutionary epidemiological model from the Duffy & Sivars-Becker paper. In our initial submission to AmNat, we only modeled one of the two parasites, because we only had data on genetic variation in susceptibility to that one parasite. The reviews came back mostly positive, but really wanting us to use the model to compare the two parasites. So, we revised the paper yet again to add the second parasite species in to the modeling component of the paper. This allowed us to understand the joint effects of selective predation and rapid evolution on the impacts that these two parasites have on the host population, and helped explain the different impacts we had observed these two parasites having in our natural populations. And that is a really, really important part of the paper. But I wouldn’t have gotten there without a whole bunch of prodding from reviewers and Associate Editors (especially Yannis Michalakis at AmNat).
So, as I said at the beginning, while it’s fun to say that the paper that won a major award from ESA was rejected by their society journal, it probably deserved it.