The lab work can be pushed another step, one that Hans and a few other researchers have just begun to approach, and that is to create an atavism. We can try to change the course of development in ways that would turn back the evolutionary clock. Once we have established, in the case of feathers, or digits, a sequence of development, and have a hypothesis about how changes in this sequence occurred during evolution, we can test our hypothesis. We can intervene to make the sequence of development, at the molecular level, what we think it was before the evolutionary change. We can tweak the developmental instructions given the embryo to see if the ancestral state can be re-created.
A danger is that you could simply create an effect that looks something like the ancestral state, but you might have found another route to produce a superficially similar result and not have rewound evolution at all. There are ways to protect against such a result, but this a murky area and one little explored so far. Nonetheless, in principle, if you can turn back one evolutionary pathway, you ought to be able to turn back several. If you can do it for one trait, why not for several? Why not turn a chicken into a dinosaur?
Arguments can go on forever about evolution and whether it should be taught in the schools and about the abstract nature of science and evidence. But just as a picture is worth a thousand words, I thought a living dinosaur would be worth a thousand court cases in the visceral effect it would have on schoolchildren.
6
WAG THE BIRD
THE SHRINKING BACKBONE
Most species do their own evolving, making it up as they go along, which is the way Nature intended. And this is all very natural and organic and in tune with mysterious cycles of the cosmos, which believes that there’s nothing like millions of years of really frustrating trial and error to give a species moral fiber and, in some cases, backbone.
H
ans Larsson is a fast walker and a fast talker. You need to be fit if you want to keep up with him on the hills of the McGill University neighborhood in Montreal, let alone on the remote islands of the Canadian Arctic where he searches for fossils in summer fieldwork. He talks the way he walks, freely swinging in a fast-paced lope from the philosophy of science to genetic probes to the rich Cretaceous ecosystem he is exploring at another field site in Alberta.
Like many paleontologists he has been fixated on dinosaurs since childhood. He is, however, unusual in the breadth of his intellectual interests. Just as he seems impatient with a slow walking pace, he is impatient with the limitations of traditional paleontology. He is one of the scientists in the forefront of merging paleontology and molecular biology in an effort to connect major evolutionary changes—the development of new species and new characteristics, new shapes and structures, new kinds of animals—to changes in specific genes and their regulation.
He came to his current mix of research because he found himself unsatisfied with the business of collecting and categorizing fossils and drawing inferences about evolution from the fossil record.
“The reason that I was initially disenchanted with dinosaur paleontology is that these things were not testable. Just sort of stories and scenarios. Anybody and their mom and dog could come along and join the party. So it needs to be rigorous. And there needs to be some testing and developing these things across interdisciplinary approaches. And so including experimental embryology and ecological approaches to it, that’s keeping me satisfied with it.”
Anybody who wants to shake up traditional scientific approaches, bridge disciplines, ask new questions in new ways, is a researcher after my own heart. Using embryology to test ideas developed through paleontology seems to me to be a big part of the future of evolutionary biology. And Hans has been doing research right at the heart of the transition from dinosaur to bird. What is even more intriguing to me is that he is interested in pushing experimental embryology forward by reactivating dormant genes or changing the regulation of active genes to bring back ancestral traits that have been lost in evolution.
Another part of his approach to science, consistent with the desire to make paleontology more rigorous, was a concern with the philosophy of science, with the nature of proof and evidence and experiment. This kind of concern, rare among experimentalists and field paleontologists, is another aspect to his unwillingness to accept the status quo. He wants not only to make paleontology testable by laboratory experiment, he wants to define the nature of testability and what constitutes an experiment.
Over the forty years or so that I’ve been deeply involved in dinosaur research, what satisfies me the most has changed. First it was finding new fossils. Next it was changing paleontology by pushing it to bring new sorts of research techniques into practice. For the past few years it has been teaching. The greatest pleasure now is watching graduate students who are smarter than I am turn over old ideas and break new ground.
All teachers hope to pass on something to their students. For me, it’s not specific knowledge or technical expertise. My graduate students quickly outstrip me in lab skills and knowledge of molecular biology. My goal is more like that of a high school science teacher who recently introduced his class to the theories of Georges Cuvier, a genius whose career straddled the eighteenth and nineteenth centuries. He was the first paleontologist, and proved the reality of extinction. He did not suggest that new species evolved, however. His theory was that the earth was incredibly old, and stayed largely the same, with periodic catastrophic changes, or revolutions, that caused extinctions. He did not see the reality of gradual geological change over time that molded the earth.
The teacher presented Cuvier’s ideas to the class as solid, well-proved science. Some students disagreed, but he argued them down and, with a combination of his greater knowledge and his position as the teacher, eventually convinced all but one student. Once he had done so, he made a sudden about-face, now revealing to his class that Cuvier’s ideas were, in fact, erroneous. He congratulated the student who had held on to his point of view, and, turning to the class again, warned them never to believe something just because a teacher said it was true. That’s what good teachers, at any level of science, or any other field, for that matter, hope to pass on to their students.
Hans is not a student of mine. In fact, I could be a student of his in development and molecular biology. But he has exactly the kind of agile and restless mind that any teacher would look for in a student, and that any scientific discipline needs to keep it alive. He also has the invaluable ability to merge disciplines, to jump from fossil collecting and analysis, to laboratory experiment, to the philosophical examination of how science should proceed in its investigation of the past. A physics or chemistry experiment can be done in the laboratory and reproduced until the standard of scientific proof is clearly met. The history of life is more elusive. We have the fossils. We have developmental biology. We have molecular biology. All are now being merged in the study of the history of life in evolutionary developmental biology. Along with some of his colleagues, like Günter Wagner of Yale, he has been concerned with establishing an accepted set of logical parameters for forming and testing hypotheses in evo-devo.
When a field is new and attractive, it is easy to make leaps beyond what we actually know. It is so clear in general that any evolutionary change in an animal must be a change in development that it is very tempting to start connecting the dots quickly, linking specific developmental and genetic changes to changes seen in the fossil record. But over the past two centuries the use of embryology to discover clues to evolution has faded in and out of favor. One reason is that for a scientific field to prosper there must be agreement on how to assess the evidence, and what logical steps lead to falsifying or supporting a hypothesis.
Collecting and cataloging fossil bones, the heart of vertebrate paleontology, has been primarily a historical enterprise, one of collecting information and looking for patterns in that information. You could test some conclusions, but experiment was not really part of the discipline. Instead you might conclude, on the basis of discoveries of dinosaur eggs and young in preserved well-drained highlands in Montana that dinosaurs preferred this kind of territory for nesting. You could predict that similar formations would produce young and eggs around the world and that wetter locations would not.
That is a self-serving example, since it was my prediction and more eggs and young were indeed found in those kinds of locations. Still, this is more like history than chemistry. And I’ve had my fair share of hypotheses that have been proved wrong.
Laboratory science, in particular the study of microevolution, has been conducted in a different fashion. You could suspect, say, that a particular growth factor is important in the formation of the tetrapod hand. So the hypothesis might be that if that gene were absent or nonfunctional, the hand would not develop. With mice, one can knock out a gene. You can engineer the mice so that the gene is absent or silenced and see what happens in development of the embryo. If the hand develops perfectly, you have falsified your hypothesis. If it does not, you have good evidence that the gene in question does what you thought it did.
With flies and worms such hypothesis and experiment is relatively straightforward. The genetic systems are simpler, the generations shorter. All sorts of evolutionary hypotheses can be tested. But these all have to do with small changes. What about a significant change in form in which something new is introduced that hasn’t been seen before in evolution, something like the appearance of limbs, or hair, or feathers, or lactation?
Well, it can be investigated in the laboratory with intellectual rigor by adhering to the notion of symmetry that Hans used in tracking the evolution and development of the vertebrate hand and that Richard Prum identified in the evolution and development of feathers.
REWINDING EVOLUTION
But there is another way that it can be tested that has hardly been attempted, and that is to run the tape of evolution over again, to use our ability to intervene in the course of development in the chick embryo (or other embryos) to reverse evolution. This is a profound advance in the kind of experiment available to test evolutionary theory, and it depends entirely on the progress that has been made in evolutionary developmental biology. It is only because we can match developmental events to evolutionary events, only because we now have both the fossil record, which shows us the path that evolution has taken, and the developmental record in extraordinary detail, that we can link the two.
It is hard to overestimate the importance for understanding evolution that a detailed record of development offers us. We have the ability to map precise developmental pathways, not just in terms of the observed patterns of how cells organize and differentiate, but in terms of which genes are activated, and when, and which growth and signaling factors are present in different areas of the embryo, and when, and at what levels. We have the tools, with probes that tell us what proteins are directing growth and development, to discover and write down the entire program of growth for an organism. The complexity of this for a human being would be overwhelming. But in principle we could acquire that information, and the computing power to organize and understand the information that grows by leaps and bounds.
To be realistic, of course, we are not close to such an achievement. Look at the progress with
C. elegans,
the first multicellular organism to have its genome mapped. Researchers have also mapped its development cell by cell. That is to say that for the body cells (apart from the gonads) every cell in the worm’s body can be traced from the fertilized egg to the fully formed worm. But we do not have the accompanying set of instructions, when each gene turns on and what concentration of each growth factor and signaling factor occurs at each location and at each stage in development. In fact, although the genome has been mapped, that does not mean that scientists know each gene. The sequence of every section of DNA is known, but it is another thing to fish out of that database which sequences are genes, and what their function is. That process itself is an enormous challenge. But it is conceivable that a truly complete instruction book for the development of
C. elegans,
from one cell to fully formed adult, can one day be compiled.
But a full instruction book is not necessary to try to rerun evolution. With birds, for instance, the absence of a tail, the difference between wings and grasping forearms, the absence of teeth, are all subtle evolutionary changes on a basic dinosaur plan. Perhaps if we imagine the dinosaur plan as a house plan. We might be going from a Cape Cod to a saltbox. Or you might think of the evolution of cars, since the way technology develops has some parallels to organic evolution.
Perhaps we could look at four-wheeled contrivances, like the simple cart, as similar to the first tetrapods on land, and the modern profusion of motorized vehicles as the modern world’s profusion of reptiles, amphibians, and mammals. The birds, as dinosaurs, are included with the reptiles. Perhaps they are sports cars. This analogy clearly does not hold up if you look at it carefully. But the point is that trying to reverse-engineer a Model T from a Corvette is not as complicated as going all the way back to the invention of the wheel. Perhaps a closer analogy would be to changing the manufacturing process to leave out fuel injection and vary body shape. The chassis would stay pretty much the same.
Development is tougher than car manufacturing. Cars don’t really evolve. They don’t self-assemble starting with one part. And we have fully accessible and detailed manufacturing plans for cars with and without fuel injection. But if we have a detailed manufacturing plan for one part of chick development, the wing or tail, why couldn’t we go run back the development process again, this time triggering the signals to produce a grasping forelimb or a long tail?