We’ve just posted a new preprint! Like our recent funny-titles study, it’s a pandemic pivot project. Like our funny-titles study, it’s a little weird – but also exciting. I’ll tell you a bit about the preprint, and then use it to make a point about collaborations.
Have you ever wondered if names only label things, or if they also influence the way we think about those things? Actually, I hope you’re well past wondering, because there’s extensive evidence that it’s the latter: for example, names are one source of unconscious bias in how we think about each other. But what about the Latin (scientific) names of species? Could the scientific name of a species influence the kind of research that gets done on it?
Told you it was a weird one. Surely not, right? The job of a scientific name is to label a species so we can refer to it unambiguously.* Scientists are too objective and too smart to let that name then determine whether or not they study that species, right?
Here’s what we did. We focused on plant-feeding arthropods (insects and mites) and on studies of host-associated differentiation (HAD). HAD** is an evolutionary process of dietary specialization, by which a once-generalist species diverges into two (or more) specialist races or species. For example, an insect that once fed on two goldenrod species might give rise, via HAD, to a pair of closely related but genetically distinct races, one feeding on each of the two goldenrods. Evolutionary ecologists are very interested in this process, because plant-feeding arthropods are astonishingly diverse and often tightly specialized, and HAD may be a big part of the group’s diversification. But there are hundreds of thousands of plant-feeding species that we might study, to ask whether or not they’ve undergone HAD.*** As a field, we’ve studied just a tiny subset of those. Which ones? Could whether or not we’ve studied a species depend on its name?
We compiled a list of 30 insect and mite species that have been tested for HAD, and then we compiled a much longer list of all the species in their genera (about 2700 of them, in total). For each, we recorded the etymology of the species’ Latin name. Some names derive from a geographic place (Oxya japonica); some from morphology (Thrips alatus); some are eponymous (Diplolepis mayri); and some refer to the plant the species attack (Eurosta solidaginis, which attacks Solidago.) We asked whether species named for their hosts were more likely to have been tested for HAD – by flipping the question over, and asking whether species tested for HAD were more likely to be named for their hosts. And they were! 47% of HAD-tested species have names based on their host plants; but only 23% of the other species do.
This is a startling result, because we’re not used to thinking this way. How scientists decide what to study is sort of mysterious, but I think we generally assume that it’s a rational process. We study a species that offers a good test of our hypothesis, or one that lives in an accessible location, or one that’s economically important. Our study suggests there’s more to it: we may be unconsciously influenced by what species are called. Actually, I don’t think that’s implausible. I started studying HAD with the species Gnorimoschema gallaesolidaginis, which makes galls on Solidago. I had noticed the galls, looked up what they were, and wondered if it was a generalist on multiple Solidago species or a complex of specialists. Did the species’ name pique my interest in its relationship with its host? Good question.
I promised a point about collaborations and author contributions. At the same time as I posted the preprint, I also submitted the manuscript to a journal.**** Part of the submission process involved listing the contributions different authors had made. I was lucky with this project: I got to work with some really fabulous scientists (and people): Julia Mlynarek, Chloe Cull, Jess Vickruck, and Amy Parachnowitsch. (As a side note, here’s some career advice: find smart, kind collaborators and everything is better.)
I found it easy enough to list who had curated the data, done the analysis, or contributed to writing; but being asked about about study design and conception really made me think. The idea for the project goes back a long way. Julia and I had been talking about her leaf-miner HAD data and how lots of leaf miners are named for their host plants. But whose idea was it to ask whether name etymology influences whether HAD gets studied? I don’t know; my guess is mostly Julia, but it was back and forth so much that neither of us is really sure. And who came up with the approach to answer the question? I don’t know that either; my guess is mostly me, but it was back and forth so much that neither of us is really sure. So it’s a good thing that author contribution statements can be pretty vague, because even we can’t offer a straightforward list of our own contributions.
Does it seems strange that we can’t remember whose idea our study was? It won’t, if you’ve been lucky enough to have the very best kind of collaboration. In the very best collaborations, that’s how it works: you bat ideas back and forth, you scribble over each other’s sketches on a whiteboard, you change your minds repeatedly, you argue back and forth until you realize you’re swapped positions and are still arguing. Eventually, it doesn’t even make sense to ask which one of you came up with the idea – but a new idea exists, and it’s one that neither of you could have come up with alone.
As scientists we’re in the business of finding stuff out. We desperately want to know things. But one thing I’m very happy not to know – not to be able to know – is “whose idea was this”? Not knowing that is one thing that makes science fun.
© Stephen Heard July 5, 2022
Images: The preprint in question; and a gall of Gnorimoschema gallaesoldiaginis on Solidago gigantea, © S Heard, CC BY 4.0
*^The job of a scientific name is not to describe a species. Yes, some scientific names are descriptive, like Setophaga cerulean, the cerulean (blue) warbler. But that’s not a necessary component of scientific names, and here’s why I think those who insist it should be are barking up the wrong tree.
***^Such a study normally involves collecting individuals of what appears to be a single generalist species, from two or more different plant hosts, and then using genetic markers such as allozymes, restriction fragments, microsatellites, or single-nucleotide polymorphisms to screen for genetic differences associated with the host plant from which the individual was collected. You asked. OK, you didn’t ask. But I told you anyway.
****^Despite arguments that posting preprints allows for feedback before submission to a journal, it rarely works that way – most preprints receive little if any feedback; and realizing that, most authors don’t wait to submit. There’s a fuller discussion of preprinting in Chapter 25 of The Scientist’s Guide to Writing.