What are the greatest neglected classics in ecology and evolution? Truly great papers or books that deserve to be much more widely known than they are?*
Part of the challenge here is deciding how little-known a paper has to be before it counts as “neglected”. I’d be very surprised if any truly great paper was totally unknown–say, never cited, or cited only a handful of times. Indeed, lots of my very favorite papers are hardly obscure, even if they’re not nearly as famous as I think they should be.
The other part of the challenge here is deciding how good and important a paper has to be in order to deserve to be called a “classic”. There’s no single right answer here. Major league baseball fans fall on a continuum from those who’d prefer a small Hall of Fame, to those who’d prefer a large Hall of Fame. Similarly, I’m sure ecologists and evolutionary biologists fall on a continuum from those who think that only papers on this level count as “classics”, to those for whom “classic” means “any paper I like a lot”.
Note that a “neglected classic” needs to be old enough that it could’ve attained “classic” status. Ok, I don’t know exactly how old that is. But probably at least a decade, right? Maybe even more?
Here’s an opening bid: Levin 1970. A huge step forward for coexistence theory, I think. Mark McPeek thinks so too. I agree 100% with Mark that this paper totally supersedes older verbal ideas about coexistence. If you want to understand coexistence as a dynamical phenomenon, and so appreciate all the ways in which Hutchinson’s famous “n-dimensional hypervolume” metaphor is limited and unhelpful–even the opposite of helpful–you need to read Levin 1970. It’s also very accessible, and it’s got an unusual and charming epilogue. And Levin 1970 has only been cited 262 times according to Web of Science, about 1/10th as often as the most-cited ecology papers from that time. Perhaps Mark’s paper looking back on Levin 1970 will raise its profile to “classic”.
Armstrong and McGehee’s work on nonequilibrium coexistence via the mechanism now known as “relative nonlinearity” might qualify. This is hugely important work and was a big conceptual leap for community ecology at the time (though see next paragraph for related work from around the same time). But Armstrong and McGehee 1980 (the paper most ecologists cite, even though they first published the idea in 1976) has been cited 572 times, which arguably means it’s not really “neglected”.
And as long as we’re talking about neglected classics in coexistence theory, I’ll throw Levins 1979 out there. Terrific paper, only been cited 278 times. But some of the key ideas are in Levin 1970 (Levins basically extends the argument to non-equilibrium situations), so if you forced me to pick one I’d pick Levin as the neglected classic.
Until not too long ago, Price 1970 was definitely a neglected classic. But thanks to the work of Steven Frank and others, the Price equation is now much more widely known. Price 1970 has now been cited 651 times, and unlike most papers it’s been cited much more often in recent years than it was when it was first published.
I can think of numerous others just off the top of my head, though I don’t know that any are quite on the same level as Levin 1970. Armstrong 1979 (ahead of its time, though it has antecedants including Levin 1970; Robert Armstrong has numerous underrated papers). Kaunzinger and Morin 1998 (ok, I’m laughably biased on this one since Christina Kaunzinger was a labmate of mine and Peter Morin was my supervisor, but I still think it’s maybe the greatest food chain experiment ever and should be in every ecology textbook). Chesson and Huntly 1997 (one of my favorite papers ever, a big influence on me and very well-known in the circles I move in, but not as widely known as it should be). Lots of others I’m dying to name–I could go on and on. But I’ll shut up now and open the floor for comments.
*Maybe we should call them “should-be classics”, since “neglected classics” is something of an oxymoron.
Dynamic Ecology 
Van Valen, L. (1976) Energy and Evolution. Evolutionary Theory vol. 1 (7) 179-221.
The paper has only been cited 121 times. Very interesting in that it provides an energetic definition of fitness. Every time I read it I take something new away from it. Not surprising it is not cited more as it buried in his own journal and typical of Van Valen papers can become quite obtuse in sections.
Interesting. If memory serves, back in the early 90s Jim Brown also proposed an energetic definition of fitness, in a book chapter (at least, that’s where I encountered it). I found it intriguing but ultimately not convincing. It just seemed like too big a leap from “energetic considerations are among the important determinants of fitness” to “energetic considerations can be used to *define* fitness”.
In light of Joachim’s comment, it’s interesting to speculate on what might happen if an energetic definition of fitness were ever to be widely adopted. Possibly, Van Valen’s paper might come to be seen as a classic, much like George Price’s work on what’s now called the Price equation. Or perhaps more likely, whatever work rediscovers or revives Van Valen’s idea might end up being regarded as the classic, with Van Valen (and Brown, and maybe others I’m unaware of) regarded as historical footnotes.
Karl Düsing proposed “what is undoubtedly the first formal sex ratio model and perhaps the first mathematical model in evolutionary biology” according to Seger and Stubblefield (2002. Pp. 2-25 in ICW Hardy (ed) ‘Sex ratios.’ Cambridge Univ Press). The model is almost identical to that of Shaw and Mohler (1953. ‘The selective significance of the sex ratio.’ American Naturalist 87:337-342) except for notation.
It’s been published in a journal and later been expanded to book length (see below). Rediscovery is due to AWF Edwards (1998. ‘Natural selection and the sex ratio: Fisher’s sources.’ American Naturalist 151:564-569), who reconstructed the unacknowledged sources of RA Fisher (1930. ‘The genetical theory of natural selection.’).
1. Düsing, Karl. 1883. ‘Die Factoren, welche die Sexualität entscheiden.’ Jenaische Zeitschrift für Naturwissenschaft 16. Bd. Neue Folge X. Bd, 1883.
[online here: https://archive.org/stream/jenaischezeit1691882medi#page/428/mode/2up%5D
-> The mathematical sex allocation model starts at page 435
2. Düsing, Karl. 1884. ‘Die Regulierung des Geschlechtsverhältnisses bei der Vermehrung der Menschen, Thiere und Pflanzen.’ Jena: Gustav Fischer.
Except for citations in historical contexts, these are cited roughly zero times.
Yes, rediscovery of ideas seems like an important source of neglected classics–the rediscovery becomes regarded as a classic, while the original source is neglected.
Of course, the difficulty is that often (not always), there are some differences between the original idea and the rediscovered version, which at some point become sufficiently large that the “rediscovery” becomes an original idea rather than a rediscovery. Think of all the claims that Darwin’s Origin wasn’t original because others had previously made passing remarks or brief proposals that could be interpreted as anticipating Darwin’s ideas. Which neglects the fact that Darwin developed those ideas, and the evidence for them, much more completely than anyone else had, which is why I think it’s right to regard his work as the “classic”.
In ecology, the origin of the intermediate disturbance hypothesis is controversial. Connell (1978) coined the term and is the most-cited source, but others had the same basic idea earlier. The basic idea of apparent competition has been independently discovered numerous times, before (and probably after) Bob Holt coined the term and published the first thorough theoretical study of the phenomenon in 1977. I’d say the case of apparent competition is somewhat like Darwin’s–others had had the basic idea before, but I think Holt was the first to thoroughly study it and work out its implications, so Holt 1977 deserves the label “classic”. And neutral theory in ecology has several independent origins (leaving aside its long history in evolution)–an Ecological Monographs paper by Hal Caswell, and then later and roughly simultaneously, the work of Steve Hubbell and Graham Bell. But it’s Hubbell’s monograph that ended up being the influential “classic”.
While I agree that anticipations are often not neglected classics, I think Düsing’s model does fit here.
If you took the time to read Edwards’s reconstruction of how Düsing’s model was widely known before 1930, how it reached Fisher and how Fisher therefore did not even see a need to mention his sources, I guess you’d come around to take it as fitting into your category of neglected classic.
It’s a failure of citation practice that eclipsed Düsing’s work and an instance of the ‘The Matthew effect in science’ (Merton R.K. 1968. Science 159:56-63) meaning the fact that famous scientists often get credit for ideas that aren’t theirs.
Darwin’s first edition of the Descent of Man also features in this history, by the way.
For a German translation of Düsing’s model see a later paper by Edwards linked to at my blog.
Re: the Matthew Effect, yes, I’m sure that’s what’s going on in many cases. Interesting history re: Fisher, of which I wasn’t aware.
That’s actually part of what makes the history of neutral theory in ecology so interesting–it’s a case that can’t be explained by the Matthew Effect, because all concerned were famous. Hal Caswell was (and is) really famous. And he published his ideas in what was, and is, one of the most high-impact journals in ecology. Graham Bell is really famous too, though more famous in evolution than in ecology. And he first published his ideas as a major perspectives-type paper in Science, with follow-up papers in high-impact ecology journals like Am Nat. Steve Hubbell is also famous, but when he first published his ideas about neutral theory, he published them in a specialized journal (Coral Reefs, 1997), to little notice. It was his 2001 Princeton Monograph that finally caused neutral theory to take off in ecology. I’m not quite sure why. Possibly, because writing a book allowed Hubbell to develop the idea at length and work out all its implications. Except that Caswell and Bell both published lengthy papers working out many different implications of neutral theory.
Dear Joachim,
Your thoughts on Düsing’s work are very interesting. Thanks for sharing that.
But I would say that the first mathematical model in evolutionary biology is the one proposed by Johannes Friedrich (“Fritz”) Müller in his 1879 paper “Ituna and Thyridia; a remarkable case of mimicry in butterflies”. His “number-dependent model of mimicry” describes the fitness gains of a species adopting specific traits, in a mimicry system.
Just quoted Seger and Stubblefield on that matter, but you’re correct, Müller’s model is earlier.
Thanks for this interesting blog.
Hope it’s ok I have added your references to the community curated collection on mendeley.
http://www.mendeley.com/groups/574761/biology-classics/overview/