Ernst von Haeckel, an early champion of evolutionary theory, was the best-known scientist to propose that similarities among embryos contain important information. Specifically, Haeckel proposed that "ontogeny recapitulates phylogeny." This pithy bit of jargon, when translated into English, asserts that as an embryo develops, it passes through stages that are equivalent to the adult forms of its ancestors. For example, according to Haeckel, a human embryo would pass through a stage in which it has features of an adult fish, then features of an adult amphibian, and so forth.
That would be a mighty tall order, of course, for at least two reasons. First, many adult animals (including extinct species) are highly specialized, and carry scores of complicated structures that would have to be assembled and disassembled as an embryo progresses from one stage to another. Second, any species alive today has an awful lot of ancestors that stretch back over billions of years. To pass through the adult stages of all of them would make for a long, tortuous (and wasteful) embryonic life!
The intriguing point is that among all the millions of animal species alive today, there are only a couple of dozen really different body-part tool kits.
Biologists have known for decades that ontogeny doesn't strictly recapitulate phylogeny—at least not in the precise way that Haeckel thought it did. Haeckel was wrong in his insistence that embryos resemble ancestral adults. But many embryos do pass through stages during which they look a lot like embryos of their ancestors—and therefore, embryos of related species. And many scientists agree that events during development are vitally important in evolution. Leigh von Valen, a prominent modern evolutionary thinker, went so far as to suggest that ".. evolution is the control of development by ecology." The subject is fascinating, complicated, and not at all easy to summarize. (Stephen Jay Gould devoted several years and a lengthy book to the topic.) But here are a few important points to think about.
The embryos (fishes, birds, pigs and humans) in NOVA's "Odyssey of Life - The Ultimate Journey" resemble each other because they all belong to animals that biologists call vertebrates. Any textbook can tell you that, but what does grouping us together that way actually mean? The simple answer is that we all share common ancestors who evolved a successful body plan based on a backbone, two pairs of limbs, and body systems set up in a certain basic manner. But what does that mean? And what does it have to do with the importance of embryos in evolution?
Think about it this way. Each major animal group has evolved a unique combination of particular body-parts that perform essential functions. Take, for example, the fact that all animals must breathe. Many land animals use lungs like ours, but insects and spiders use quite different devices. Some aquatic animals use various styles of gills, while others just let oxygen and carbon dioxide pass across soft, wet skin. You can think of lungs, gills, and other body parts that help animals breathe as the "breathing tool" component of their body-part kits. There are similar "tools" for feeding, movement, defense, reproduction, and so on.
The intriguing point is that among all the millions of animal species alive today, there are only a couple of dozen really different body-part tool kits. Each is the hallmark of a major animal group—a collection of related species that biologists usually call a Phylum. Mollusks—snails, clams, octopi and their kin—are one such group. Insects and their relatives are another. But with only two dozen or so basic body plans, where do the many thousands of species within each group come from? You can think of each group's basic body plan as the biological equivalent of a major musical theme. The slightly different body plans of species within each group are like variations on that theme. Just where do these themes and variations come from? Here's where things get interesting.
An embryo grows and develops under the control of its genes. Some genes work fairly simply, directing cells to churn out products and assemble those products into structures. But these relatively simple genes couldn't produce a complex organism by themselves. Their actions are coordinated by master control genes that act like orchestra conductors—determining which genes are turned on and which are turned off, in what cells, at what stages in development, and for what lengths of time. Of course, the task of orchestrating the entire process of constructing a fish, pig, or human makes directing the most complex musical score look like playing with nursery rhymes!
This is one reason why entirely new basic body plans don't evolve very often. Each body-part is assembled by a group of genes acting under the direction of particular control genes. Those control genes are themselves controlled by higher level control genes. And those genes are controlled by still higher-level control genes. (In some ways, these controls-within-controls are set up almost like a military chain of command.) The long and the short of it is that making major changes in body-part tool kits requires wholesale shakeups in these complex genetic programs. And in order to survive the test of natural selection, these shake ups need to happen in ways that don't introduce any fatal flaws.
So it isn't surprising that completely new body plans haven't evolved very often. In fact, these sorts of shakeups have occurred only a few times in the entire 3.5 billion year history of life. The best-known was the "Cambrian Explosion"—a period roughly 600 million years ago during which the basic body plans of most major groups of living organisms (and those of many extinct groups) arose. Of course, those original body plans were controlled by genetic programs that have been passed down over time to species alive today.
It is even more fascinating, however, to realize that minor modifications in the timing and ordering of events during embryonic life can produce enormous differences in adults.
How does all this explain why vertebrates pass through an early stage that resembles a fish embryo? As Nietzsche once wrote, "Ye have made your way from the worm to man, and much within you is still worm." Our "fish-like" early stages are directed by parts of the same genetic program that built early fishes—some of the first members of our branch of the animal kingdom. By the time fishes evolved, genetic control of development was already a very complicated business, because fishes are complex animals. And once a genetic program passes a certain level of complexity, it becomes difficult for major changes early on not to have "domino effects" that knock things out of kilter down the line.
So, over millions of years, evolution operated mainly by adding on to and fiddling around with later stages in development, rather than by making radical changes in genetic programs that substituted one type of body part for another. The result is that some aspects of the earliest stages in the human developmental program remain rather similar to those found in living fishes.
It is even more fascinating, however, to realize that minor modifications in the timing and ordering of events during embryonic life can produce enormous differences in adults. In fact, nearly all the sorts of evolutionary changes most of us usually think about—ancient fishes giving rise to amphibians, amphibians to reptiles, dinosaurs to birds, and so on—have occurred through relatively small changes in timing and orchestration of genetic controls during development. That's how ancient limbs evolved into wings, or feet into flippers. It is also how ape-like ancestors evolved into humans. Why do humans share an astonishing 98% of our genes with chimpanzees? Because, when it comes to differences between such closely related species, timing is everything!