Evolutionary biologists ask two kinds of questions: what happened, and why did it happen. First, what is the history of life as revealed by fossils and phylogenetic systematics. And second, why do organisms look and act as they do - what are the ultimate forces driving evolution? Until recently, few evolutionary biologists have been concerned with the "how"questions - how are organisms built and how do they work?

Following on from Darwin's enormous contributions, evolutionary biology has made great strides in the 20th century. We have a quantitative, predictive understanding of how natural selection acts. We can explain why the peacock has overblown plumage, why honeybees live in large, mostly sterile colonies, and why chimpanzees have dominance hierarchies. We even have a pretty clear understanding of the process of speciation, the problem that Darwin ostensibly tried to tackle.

Two themes run through this work. First, natural selection is the primary creative force in evolution and second, organisms are only skin deep. That is, these "organismal" studies have focused on the external morphology of the whole organism plus behavior. The implicit assumption has been that the mechanisms that build and operate the organism will evolve in response to natural selection by any means necessary.

This paradigm is based on the work of the great discoverer - naturalists who concentrated on finding, describing, and cataloging the external features, bones and behavior of the diversity of life. These data were the inspiration for Darwin and subsequent evolutionary theorists - and this paradigm has been very successful. However, 20th-century biology is primarily the story of the mechanistic biologists, the "how" biologists. With, for example, the rediscovery of Gregor Mendel's work, the discovery of the structure of DNA, the unravelling of metabolism and photosynthesis, the rise of molecular biology, and the fusion of molecular genetics with embryology, we have learnt a fantastic amount about how organisms work. I would maintain that the "how" biologists are the great discoverer - naturalists of the 20th century. They have been discovering the taxonomy and natural history that occurs within organisms, and it is to their data that evolutionary biologists will turn in the 21st century.

The evolutionary biologists will again ask two questions: how did these mechanisms evolve (the "what happened" question) and what are the general rules governing the evolution of these entities (the "why" question). This is all fairly obvious, and there are clear trends that research is going in this direction. But what specifically will this shift in evolutionary biology teach us? Let's focus on my main interests, the evolution of development. And I will give my predictions a head start in the inaccuracy stakes by talking about organisms I know very little about, dinosaurs. The most profound biological discovery of the past twenty years is the fact that most of the genes that pattern animal embryos are conserved over the entire animal kingdom. Over the past century, biologists have believed that species diversity was generated by the evolution of developmental mechanisms, but no one guessed that all the animals would use the same genes to pattern their bodies.

But what do these genes do and what does it mean to say they are conserved across the animal kingdom? For this discussion, there are two relevant classes of genes: genes that autonomously instruct cells to carry out a set of instructions, and genes that mediate cell to cell communication. Development is communication and commitment.

The first group of genes, the autonomous instructors, produce proteins whose job is to turn other genes on and off, that is they mediate transcription which is why they are called transcription factors. I call them autonomous instructors because they instruct cells to do things only when the protein is present in the cell. It is the action of unique combinations of transcription factors working in a single cell that specify the fate of a cell.

The second group of genes produce proteins that mediate cell-to-cell communication. Communication is crucial for development: for example, neighbouring cells can confirm that they have adopted the correct fate, they can induce one another to adopt new fates, and they can help one another align into the correct position. Thus the regular spacing of bristles on flies is generated by cell-to-cell communication and it is difficult to see how it could be otherwise. That is, it must be very difficult to precisely pattern the details of an organism without such communication.

Communication occurs over two spatial scales, locally between neighboring cells, and over the entire organism. Local communication is made possible through molecules that are either attached to the cell surface or diffuse a short distance. Long-range communication is controlled by hormones that coordinate events throughout the organism.

Those are the basic mechanisms. The biggest surprise has come from the discovery that many of the molecules that perform these tasks in flies are the "same" molecules that perform them in humans. What this means is that the common ancestor of flies and humans possessed these genes, presumably for doing the same sort of jobs. In the time since this common ancestor lived, about 500 million years ago, these molecules appear to have retained the same general function.

The future lies in understanding the details. Exactly how do these molecules pattern an organism and how do these systems evolve? The goals are to achieve a historical understanding and a functional understanding. How did these conserved molecules generate observed diversity? And are there any general rules governing how these systems can evolve? Back to the dinosaurs.

Michael Crichton's approach to reconstructing a dinosaur, although lucrative for him, is really rather inelegant. Reconstructing a dinosaur by sequencing fossil DNA is not only prohibitively difficult, if not impossible, but this brute-force approach treats the genome and development as a black box. It also assumes that dinosaur development is so different from the development of modern organisms that the only possibility for reconstruction lies in capturing the entire genome of an extinct species and plugging it into a fertile receptacle.

We can do better. We now know that all animals probably share the vast majority of developmental genes and that the differences between animals lie in the developmental regulation of this conserved set of genes. The elegant approach combines progress in phylogenetic systematics and comparative developmental biology. The first step is to reconstruct a likely complement of genes for our dinosaur and their regulatory regions. This could be done by estimating ancient DNA sequences by comparing living species. In general, if two living species share the same DNA sequence then it's likely that their common ancestor had the same sequence. So to reconstruct an extinct dinosaur's genome, we would rely on comparisons of bird genomes (since birds are the only existing dinosaurs) and their close relatives, the reptiles.

In 25 years, it will be as easy to sequence the genome of a bird as it now is to sequence a single gene. Phylogenetic methodology will have advanced to such a stage that we will be able to generate the most likely reconstruction of a dinosaur's DNA sequence with an estimate of the error in various regions. By this point, comparative developmental genetics will have advanced such that we will understand how variation in the structure of molecules influences their properties and, more importantly, how variation in regulatory regions creates morphological and behavioral diversity.

This information will be crucial both to understanding how the reconstructed dinosaur sequence would have functioned and to understanding how variation in this sequence created variation among different dinosaurs. We should be able to achieve all this for at least several of the key features of dinosaurs. It will probably take another 50 years for the remaining genes. The crowning accomplishment will be to combine all these features in the first dinosaur built by humans, perhaps in 200 years from now.

Am I serious? The problem is more subtle than just understanding how dinosaurs were different from modern birds. Essentially, we want to understand how different dinosaurs were constructed. What made Tyrannosaurus different from Velociraptor? Consider humans and chimpanzees: they share all the same body parts, but they are clearly different species. The primary differences lie in the relative sizes, or shapes, of different organs (such as the brain) and in quantitative differences such as the amount and distribution of hair. Quantitative change dominates evolution, but the mechanisms responsible are largely unknown. This will have to change (and I expect it will as the tools for studying developmental mechanisms become more sensitive) if we hope to really understand the mechanisms of morphological evolution.

I expect that progress in the related fields of comparative developmental biology, genomics and phylogenetics will allow new approaches to "reconstructing" the history of life on Earth, and will bring the past "alive"in our imaginations as never before.

I would also like to predict the rise of a new sub-discipline: applied evolutionary biology. I was inspired by other talks in this symposium to reflect that evolutionary problems and solutions permeate diverse fields. The basic principle of evolution by natural selection - variation with selective amplification - is being applied to computer science, combinatorial chemistry and drug discovery. These disciplines are essentially "borrowing" evolutionary ideas and applying them to problems of optimal design, illustrating that, in fact, selection is more efficient than "rational"design for finding solutions to difficult problems.

Evolutionary phenomena can also be found in pesticide resistance by crop pests, multiple antibiotic resistance in hospitals (the so-called superbugs) and malarial resistance to various drugs. I am surprised that evolutionary theory has not been used more to combat these problems. So it is here that a discipline of applied evolutionary biology could really do most good. These problems are typically attacked by developing new weapons (antibiotics, drugs, pesticides) rather than by using new tactics based on the considerable body of evolutionary theory available.

Perhaps the single most important fact is that any treatment that does not accomplish complete worldwide eradication of a communicable disease or pest simply selects for strains that are resistant to the latest weapon. My general impression is that this message, although heard and understood by many, has not altered the behavior of those developing new drugs and chemicals. I believe we must change the approach to such problems. The goal cannot be complete eradication, but instead minimization of occurrence and of damage.

For example, not only have past malarial treatments failed to eradicate malaria, but they have selected for ever-more dangerous strains of malaria. I believe that more realistic goals would be best accomplished by the application of our weapons within the framework of population and evolutionary biology. I see great opportunities for applying evolutionary theory to important and pressing societal problems in the coming century.

David L. Stern is a biologist studying the evolution of developmental mechanisms at the Museum of Zoology and Churchill College, University of Cambridge.

Andrzej Krauze is an illustrator, poster maker, cartoonist, and painter who illustrates regularly for HMS Beagle, The Guardian, The Sunday Telegraph, Bookseller, and New Statesman.

---

Send us your comments and ideas for future articles.

Endlinks

Phylogenetic Analysis Computer Programs - a collection of tools for scientists studying evolutionary relationships on the molecular level. Part of the extensive Tree of Life resource.

Flies Invade Human Genetics - describes the research that led to knowledge of the evolutionary conservation of molecular pathways from flies to humans. From the June 22, 1998 issue of The Scientist.

Evolutionary Theory - a detailed and well-written explanation of contemporary evolutionary theory. Part of the Evolutionary Psychology site at the University of California at Santa Barbara.

Related HMS Beagle articles:

• Do Genomes Enhance Their Own Evolution? - a discussion of the conference "Molecular Strategies in Biological Evolution."
• Waking Up in the Universe - a review of the Royal Institution Faraday Lectures given by Richard Dawkins in 1991, now available on video.