The Beak of the Finch and Experimental Evolution
“You can’t ‘preserve’ a species.” – Grant and Grant
Passing through an experimental evolution lab this summer, I got used to the idea that we can experimentally test evolutionary theory and the operation of selection (only?) in small, simple systems, like the yeast the lab used, bacteria, or viruses. “Experimental evolution” to me meant evolution studied at the level of genes and especially single mutations, fitness determined by who remains standing in the petri dish. So the single most striking experiment described in Jonathan Weiner’s Beak of the Finch was the observed drastic decrease in elephant tusk length among populations subject to poachers looking for ivory. I don’t know how elephant tusks work–if growth is based on Bmp4 expression, it’s not too surprising to me that a “macro” trait like tusk length could quickly evolve in response to selection pressure at the genetic level–but with all the debate and discussion about the unit of selection and macroevolution, the tusk example was enlightening.
This book is the story of Peter and Rosemary Grant’s 20-year vigil on the Galapagos Islands, particularly Daphne Major. For being that, it’s amazing how much the book focuses on the research: there is a human drama that unfolds in these pages, but it’s almost entirely told through the long progression of experiments and scientific life in the field, with some recourse to Darwin’s history. I want to learn that narrative trick!
The acceptance of quantification as rigorization makes me a little uncomfortable in all these “experimental evolution” and especially ecological studies. Mostly I think it’s good–taking detailed measurements has surely led to striking demonstrations of the power and quick-acting scale of natural selection, and how big a difference a tiny variation in beak trait matters. That’s an important lesson for all of us who would constantly wonder how selection on variation, which many think of as a piecemeal, slow process of accumulating very slight benefits, could result in qualitative differences in who survives, the generation of a major phenotypical innovation, or the origin of species. And I can see how this minute empirical measurement is partly an antidote to Darwin’s florid Victorian prose and the geometric landscape theories of the great mathematical population geneticists of the early century. But it also seems that desire for illuminating numbers can blind researchers studying evolution like this to the need for sophisticated interpretation and to hold down the fort against the allure of trumpeting a murkily visible trend. For example, there’s a recurring theme of one researcher declaring that there’s been no natural selection in Darwin’s finches, and the Grants coming back with minute year-by-year measurements and saying that, while there’s little net change in characteristics, there’s actually strong selection oscillating rapidly. Certainly, that’s a vital distinction, especially as a corrective to the idea that nobody could ever observe natural selection in action in the timespan of a human lifetime. But I think there’s a danger of swinging too far in this shiny new direction: the fact that there’s little net variation is also very important.
Some highlights of the drama of observing Daphne: The description of the difficulty of landing on Daphne, everyone’s least favorite part of the trip–there’s only really one ledge to land on, and that after repeated partial offloadings as the swells move the boat above and below the ledge. The story-cum-legend about the scientist who was walking clad only in shoes on the island and got attacked by a barnacle that clamped onto his balls. The description of the Grants back in Princeton analyzing the data with such detailed, knowing affection, as they swap labelled stories: “‘He’s been a good producer of fledglings, 2666,’ Peter says, this time without even looking at the screen.” (118) The sense of a wunderkammer somewhere in two Princeton offices, full of vegetation and food and families and songs, all in tables of data. The times when the data really does seem to make a clear point: for example, how Darwin’s finches really are much more variable than most, a natural laboratory for natural selection–sparrows, closely related to finches, rarely deviate in beak length more than 10% from the mean on the remote island of Mandarte, B.C. (ostensibly similarly isolated conditions for accelerated evolution to take place), whereas in the Galapagos 4% of the cactus finches differ from the average beak by more than 10%. (p. 47) The sheer carefulness of the data collection and care taken in what to measure is stunning: there’s an index for difficulty-of-eating among the seeds (and every seed on the island is accounted for), and the Grants know just how much force it takes to crack the toughest seeds. The careful correlation of measurements and behavior: telling what kind of finch discovered the mericarp by whether the cover is peeled back or bitten through, and how many seeds are left. Careful experimental design: Peter Boag tested for heritability of beak dimensions, ruling out that big-beaked parents get more food for their babies (not with Darwin’s finches, but on Mandarte–the egg-swapping would be quite catastrophic in such a small, fragile ecosystem!)
I got to revisiting some of my half-remembered thoughts about the difficulties in Darwin’s theory while reading this book. Prime among them–and I look forward to going back to this when I read Grant & Grant’s book about their research, How and Why Species Multiply–is observing “the ever-turning sword,” the actual origin of new species. It was never quite observed on Daphne–but there were big steps. The image of the evolutionary tree as having loose, tangled webs, and not clean breaks, at the branching points, is a powerful one–and this summer, I learned how important the visualizations we rely on are in evolutionary thinking (witness the insidious, helpful idea of a fitness landscape). Dolph, one of the researchers, even made an actual empirical fitness landscape out of all the measurements for Darwin’s finches!!!!
Two particular points of interest: the light shed on Darwin’s gradations between kind, variety and species, and how they’re borne out on Daphne–many of these finches “are so intermediate in appearance that they cannot safely be identified…In no other birds are the differences between species so ill-defined.” There’s a saying at the Charles Darwin Research Station: “Only God and Peter Grant can recognize Darwin’s finches.” One of the impetuses of Grants’ research was a monograph published by David Lack, Darwin’s Finches, which put to rest the idea that prevailed for a while that they aren’t new species at all, but “a hybrid swarm” of varieties on the Galapagos, “offering no scope for natural selection.” Lack saw that the birds weren’t breeding together, but the ground finches were eating the same seeds–he saw that in a wet season, when seeds are plentiful. Then looking over the data at home, he saw that the closest together species in their beaks never live together on one island, and inferred competition. The Grants were the first to really see competition and the principle of divergence in play on the island, and not just infer it happened in the past.
On the problem of adaptation, a neat experiment with crossbills showed proof-of-concept that little fitness advantages can accumulate to make big adaptations: the beaks of the crossbills were filed down, and each generation grew a little more of a crossbill back; each generation was more fit than the last in terms of its ability to crack the pinecones it’s adapted to. The other one, more shocking, is the key role for hybrids that the Grants posit as a result of their observations: they found, contrary to all received wisdom, that sometimes the hybrids were successful far beyond what any of the “pure-breds” could do, and it wasn’t uncommon for two different species to attempt mating. Number 006, the tiniest fulginosa on the island, always pairs with a fortis, and is the most successful of her species on the island breeding-wise. (123) The numbers for bird hybridization are striking: there are 10,000 bird species known, and 1,000 are known to have mated with other species. In some cases the rates are even higher: 67 of 161 species of ducks and geese have interbred that we know–and probably more, given the patchy state of our knowledge (we didn’t even notice that the best-studied birds ever, on the Galapagos, were interbreeding until after a 20 year vigil). This extends the role of hybridization and its power, known to produce ne wplant species “literally overnight,” to the animal kingdom. At least half of the world’s flowering plant species came from interbreeding. Does a number on the biological species concept! Hybridization is also common among Bufo toads, many insects, and many fish. Evidence also points to the role of human motions and ecology in increasing the rate of hybridization: we disturb habitats and introduce invasive species that mix with the local gene pool, and hybrids can back-cross (“introgressive hypbridization,” Edgar Anderson called it) to mix the gene pool even more. Hybrids can fill special niches their parents can’t, and humans are creating such mixed up niches all over the planet: we see this in wildflowers in the Delta, where different species inhabit different fields, which the farmers have treated with different chemicals. Anderson also argues that “ecologically dominant” species, like humans currently are, could have driven the evolution of new niches in the past too: e.g. at the colonization of new islands or continents, the first land vertebrates invading terrestrial vegetation, the first large herbivorous reptiles or the first large land animals. (Stebbins and Anderson, “Hybridization as an Evolutionary Stimulus”)
On the modern speed of evolution, Weiner goes beyond the pat “we have lots of drug-resistant strains” line. He writes of two people who took antibiotics for a few days, and then sampled the bacteria in their body: almost all were drug resistant. As they note, it’s different seeing fast evolution in a lab and in our own bodies. There are examples of all the different ways a pest can avoid the effects of a pesticide: dodging it, not letting it get inside, developing an antidote, or inactivating it once it’s inside.
And there are interesting stories to liven things up (it’s really a page turner!) Not of Darwin’s personal life or the Grants’ per se, but personal anecdotes in the course of doing research. Darwin could’ve seen natural selection in action if he’d been of a mind to, as he kicked stones in his garden to count laps and noticed the death of 4/5ths of the bird species on his grounds one winter, something like the effect of an El Nino or a drought on the Galapagos. Darwin discovered 537 species of plant in 3 tablespoons of mud in his tabletop experiment. Two engineers created “two new letters to the alphabet of life,” adding synthetic X and K to A, C, T, and G. The book doesn’t get personal, but there’s a great sense of personal investment and the importance of these scientific stories in it. I’ll be trying to get my hands on Weiner’s other books.
“Direct Demonstrations of Natural Selection” in Natural Selection in the Wild, for more examples of experimental evolution studies
*Grant and Grant, How Species Multiply