Friday, September 29, 2017

Evolution and Historical Contingency ~ A Review of Improbable Destinies


Stephen Jay Gould threw down the gauntlet when he argued in Wonderful Life:  The Burgess Shale and the Nature of History (1989) that, were we able to replay the tape of life – that is, erase the metaphorical tape back to some point in the distant past – then “any [subsequent] replay of the tape would lead evolution down a pathway radically different from the road actually taken.”  (Wonderful Life, p. 51)  Evolution, in Gould’s view, is clearly not deterministic, rather, historical contingency holds sway.  By historical contingency, Gould meant that any particular state (of history, of the evolutionary paths of species) was “dependent, or contingent, upon everything that came before – the unerasable and determining signature of history.” (Wonderful Life, p. 283)

Thirty years later debate over the validity of Gould’s proposition continues.  Are the various constraints on evolution through natural selection such that, regardless of where we re-start the tape, the results shown in the replay eventually would be remarkably similar to what we see around us today, or does contingency play such a powerful role that ending up in the same place is not only improbable but perhaps out of the realm of possibility, dependent as it is upon myriad events occurring in just the right order?

This is the context for evolutionary ecologist and herpetologist Jonathan B. Losos’ new book Improbable Destinies:  Fate, Chance, and the Future of Evolution, a riveting and remarkably accessible exploration of convergent evolution and its import for Gould’s thought experiment.


Convergent evolution is the phenomenon in which different species evolve similar features, that is, they converge on very similar morphological features, say, wings in insects and wings in birds among other vertebrates.  The list of examples of convergent evolution in the natural world is rich and long and growing.  As Improbable Destinies makes clear, anole lizards, an early example and the centerpiece of Losos’ research, have been joined by a host of other instances of convergence involving many different kinds of organisms, ranging from snails to guppies, from bats to stickleback fish.

The wealth of such examples has led some scientists, paleontologist Conway Morris prominent among them, to conclude that (as Losos writes), “evolution is deterministic, predictable, following the same course time after time.  The reason, they argue, is that there are only so many ways to make a living in the world,” and so the tape actually would end up in the same place every time it is replayed.  (p. 5)

At the same time, Losos shows that the list of “evolutionary idiosyncrasies” isn’t short and often features creatures, such as the kiwi or the solendon, that have evolved on isolated land masses such as New Zealand, Australia, and Madagascar.  There’s a diverse richness here.  Losos notes, “Conway Morris and his colleagues have made long lists of examples of convergence, but it would be just as easy to make comparable catalogs of species without counterparts.”  (p. 87)  He argues that these one-offs may appear because “natural selection is either not as predictable or as powerful as some make it out to be.  That is, even when species experience identical environments, they might not evolve in the same way.”  (p. 88)

Losos describes many research projects relevant to our understanding of convergent evolution.  These include observation of convergence in the wild, particularly in settings showcasing natural experiments (principally islands), deliberate experiments in the wild as scientists transfer organisms to different locations and analyze their subsequent evolution (yes, as he makes abundantly clear, evolution need not proceed at the snail’s pace that Darwin would have had it), and finally experiments in the laboratory, often in petri dishes or vials involving microorganisms, such as E. coli.

The highlight of the book for me (and here it’s a page-turning pleasure) is Losos’ account of his research on the Anolis lizard (commonly referred to as the anole lizard) and his field work in the Greater Antilles (Cuba, Hispaniola, Jamaica, and Puerto Rico).  Perhaps their most spectacular feature (reserved in most species to males) is the dewlap, “a flap of skin under their throat,” usually hidden from sight.
But when the lizard has an announcement to make – ‘Get lost, buddy, this is my territory” or “Hey, ladies, come check me out.  I’d make a good baby daddy’ – out comes the dewlap, arching downward from the jaw, forming a semicircle so large that the lizard often has to straighten its legs, pushing its body off the ground, to provide clearance.  (p. 58)
A nice example of the prose that graces the book.  Losos is able to make what is certainly tedious and physically difficult research on these lizards into an exciting narrative, a narrative that shows why the anole lizards are robust examples of convergent evolution.

He found that, although each of the four islands in the Greater Antilles has a different array of anole species, those species generally occupy similar ecological niches (bush, or different parts of trees) across the islands.  Those in each niche have evolved to be so similar from island to island that each of these “habitat specialists” can be mistaken for each other.  Yet all of the species on any of these islands are more closely related to each other than they are to any of the similar habitat specialists across the islands.

There is a great deal to absorb from Losos’ walk through the relevant research from the wild to the laboratory, much of it, particularly those natural and deliberate experiments in the wild, thoroughly engaging.  Yet two aspects of the research covered here are, to me, disquieting.

First, I appreciate what can be learned from projects that involve researchers moving organisms in the wild from one place to another (e.g., guppies from pools with predators to those without, or anoles to islands with no lizards), but, given how complex the array of contingencies is in the natural world, what are the real consequences of this manipulation?  I am reminded of how off putting I found ecologist E.O. Wilson’s research that involved fumigating entire mangrove islands in the Florida Keys to kill off all the arthropods.  Yes, he gained insight into how populations recover and how quickly, but at what cost?  Certainly I’m being overly sensitive about this and I’m sure my concerns would be discounted by those involved.  Still the deliberate interference bothers me.

Second, though much of Losos’ explication of the relevant laboratory research is extremely well done and informative, I had a nagging unease about this research.  It didn’t help that I often lost track of what the various projects were trying to prove and what they did prove.  Perhaps the more substantive part of my discomfort with the laboratory portion of the book has to do with my sense that test tube evolution with microbes is somewhat analogous to running computer programs that seek to replicate evolution:  a limited set of conditions are programmed in, with some random function added to mimic the mutations that natural selection has to work with.  The results can be instructive, yet they are, ultimately, artificial.  The full force of contingency that occurs in the natural world is largely missing in either the lab or the computer program.

None of this is to gainsay the importance of tracking myriad generations of microorganisms in controlled environments in the laboratory, not only to see if identical populations always end up in the same place, but also because, as at least one project was structured, researchers could, literally, replay the tape, taking a population to some previous generation and letting it reproduce from there. The message from such experiments, as I understand it, is that, yes, largely the tape replays exactly as it has run before.  But, and it’s a big but, not always.

Indeed, though most of the research from the field and from the laboratory would appear to support the evolutionary determinist camp – that the replay ends up in the same place – closer examination of the results, often over a longer range of time, shows that isn’t always true.  It’s most likely to be true when the populations begun with are closely related; such relatives are likely to evolve similarly when placed in new environments.  Not so when they are more distantly related.

But I didn’t think that Gould’s thought experiment envisioned the tape replaying exactly as it had before and the only thing in play being how evolution would be affected by random mutations in organisms.  Losos, citing philosopher of science John Beatty, acknowledges as much, positing that Gould actually embraced two concepts of contingency.  The first being that replaying of the tape with nothing changed to see if evolution would or could take a different course.  The second being the replaying of the tape from a specific point with the full panoply of historical contingencies allowed to be at work affecting living organisms.  Though these contingencies may be small events with important or unimportant consequences, they could, as well, be more dramatic things such as a storm wiping out the sole representatives of a species or an asteroid hitting the planet and decimating a host of species.  Each of these specific contingencies cannot be anticipated when we replay the tape – they may or may not occur – and they are likely to make all the difference in the world and we are likely not to end up in the same place with any replay.

In fact, and this needs no spoiler alert given the title of the book, I think Losos comes down largely in the Gould camp though his stance is somewhat nuanced.  As he writes,
So, can we predict evolution?  In the short-term, yes, to some extent.  But the longer the passage of time and the more different the ancestors or conditions, the less likely we are to prognosticate successfully.  (p. 336)
He adds, regarding our species,
If any of a countless number of events had occurred differently in the past, Homo sapiens  wouldn’t have evolved.  We were far from inevitable and are lucky to be here, fortunate that events happened just as they did.  Asteroids, of course, but what other events critically tipped evolution’s path in our favor?  Who knows how slight a difference in the past – a tree falling on great-great-great-to-the-millionth-degree-grandpa Ernie, a forest fire, a mutation – might have snuffed out our future existence?  (p. 335)
As he notes, we weren’t fated to evolve though possibly something like us might have.  Ultimately, no matter the outcomes one might anticipate from evolution, it is highly improbable that any specific one will occur.

In closing, I have to explain why I felt compelled to buy and read Improbable Destinies.  I had no choice after reading his interview in Current Biology (August 21, 2017).  He had me from the moment he answers the question:  How did you get started in biology?

Sure, he mentions dinosaurs as a gateway drug, but he goes wonderfully so much further, offering credit to:
a particular episode of Leave it to Beaver, the one in which the Beaver purchases a mail-order baby alligator.
According to the IMDb synopsis of this particular episode (frankly, I don’t remember this one which explains, I guess, why I didn’t become a biologist), Wally and the Beaver acquire a Florida alligator sub rosa and the laughs (such as they are) stem from their efforts to raise an alligator in the house without their parents discovering.  In contrast, Losos, smitten by the idea, was above board in his campaign to be allowed to acquire this kind of reptile.
. . . I knew that local pet stores sold baby caimans, the neotropical relative of alligators. The question was: how to convince my mother to allow a crocodilian in the house. Fortunately, my mom — not liking to say ‘no’ — passed the buck to the local zoo director, a family friend, expecting him to put the kibosh on the idea. To her dismay and my delight, however, he said that having an alligator was how he got his start in herpetology, and the next thing you know, I had a caiman in a kiddie’s wading pool in the basement.
In closing, I must say that my encountering this interview (which in turn prompted me to read and review the book) was contingent on a long, long chain of events that included spotting a turtle in my backyard several years ago and deciding to write a blog post on the turtle shell.  Current Biology had a relevant article and I actually paid for access to it (shows how desperate I was) which got me snared in the publisher's email web which meant the index to this August’s issue reached me.  Then again, even if I hadn’t seen the interview in Current Biology, given what I read I still would have seen reviews of the book and I might very well have ended up in the same place.
 
Nature Blog Network